WO2013089073A1 - 情報解析装置、電子聴診器、情報解析方法、測定システム、制御プログラム、および、記録媒体 - Google Patents
情報解析装置、電子聴診器、情報解析方法、測定システム、制御プログラム、および、記録媒体 Download PDFInfo
- Publication number
- WO2013089073A1 WO2013089073A1 PCT/JP2012/081960 JP2012081960W WO2013089073A1 WO 2013089073 A1 WO2013089073 A1 WO 2013089073A1 JP 2012081960 W JP2012081960 W JP 2012081960W WO 2013089073 A1 WO2013089073 A1 WO 2013089073A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- sound
- waveform
- determination
- information
- unit
- Prior art date
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7253—Details of waveform analysis characterised by using transforms
- A61B5/7257—Details of waveform analysis characterised by using transforms using Fourier transforms
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7282—Event detection, e.g. detecting unique waveforms indicative of a medical condition
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B7/00—Instruments for auscultation
- A61B7/003—Detecting lung or respiration noise
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B7/00—Instruments for auscultation
- A61B7/02—Stethoscopes
- A61B7/04—Electric stethoscopes
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B90/00—Instruments, implements or accessories specially adapted for surgery or diagnosis and not covered by any of the groups A61B1/00 - A61B50/00, e.g. for luxation treatment or for protecting wound edges
- A61B90/36—Image-producing devices or illumination devices not otherwise provided for
- A61B90/361—Image-producing devices, e.g. surgical cameras
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0002—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network
- A61B5/0015—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network characterised by features of the telemetry system
- A61B5/0022—Monitoring a patient using a global network, e.g. telephone networks, internet
Definitions
- the present invention relates to an information analysis apparatus, an information analysis method, a control program, and a recording medium for analyzing body sound information collected by a stethoscope.
- vital sound collected by a stethoscope is not information that a doctor listens to and disappears on the spot, but important patient information that can be recorded and recorded as vital sound information in an electronic medical record or the like. It became one.
- Such biological sound information is not only used for reproduction and listening by a doctor, but also becomes an object of analysis processing by the apparatus.
- Patent Document 1 discloses a respiratory sound data processing apparatus that analyzes respiratory sound data.
- the respiratory sound data processing device detects the presence or absence of secondary noise based on the sample data and the actually acquired respiratory sound data.
- Patent Document 2 discloses a lung sound diagnostic apparatus that collects lung sounds and diagnoses the presence or absence of abnormal lung sounds.
- the lung sound diagnosis apparatus determines the presence or absence of abnormal lung sounds by comparing with reference data with a known disease name.
- the body sound which is information that disappears on the spot
- the doctor auscultates the body sound on the spot determines the patient's condition from the body sound based on his knowledge and experience, and provides appropriate treatment.
- biological sound information is recorded as one piece of patient information and can be used at any time.
- a user other than a doctor who examines (auscultates) a patient on the spot can use it in any medical scene. It is assumed that biological sound information is used.
- a user is a person other than a doctor, for example, a medical worker other than a specialist doctor related to the patient, or a person related to a patient who does not have medical skills. including.
- Patent Documents 1 and 2 analyze body sound information and assist users. However, these conventional techniques compare abnormal sound or complex sound by comparing sound data to be diagnosed with sample sound data stored in advance (normal / abnormal learning data, sample data, similar reference data, etc.). Or the like.
- the present invention has been made in view of the above-described problems, and an object of the present invention is to objectively analyze body sound information collected by a stethoscope and to efficiently use the analysis result.
- the information analysis apparatus, the information analysis method, the control program, and the recording medium to be presented are realized.
- the information analysis apparatus is a waveform feature that indicates a reference for distinguishing a feature of a sound waveform from a sound waveform included in biological sound information collected by a stethoscope. Applying the determination criteria, a waveform feature determining means for specifying the characteristics of the sound waveform, and a type of sound to which the biological sound information belongs based on the characteristics of the sound waveform specified by the waveform feature determining means. And a sound type determining means for determining.
- the information analysis method provides a waveform feature that indicates a reference for distinguishing a feature of a sound waveform from a sound waveform included in biological sound information collected by a stethoscope.
- a waveform feature determination step for specifying a characteristic of the sound waveform by applying a determination criterion, and a type of sound to which the biological sound information belongs based on the characteristic of the sound waveform specified in the waveform feature determination step
- a sound type determination step for determining the sound quality.
- the information analysis apparatus is a waveform feature that indicates a reference for distinguishing a feature of a sound waveform from a sound waveform included in biological sound information collected by a stethoscope. Applying the determination criteria, a waveform feature determining means for specifying the characteristics of the sound waveform, and a type of sound to which the biological sound information belongs based on the characteristics of the sound waveform specified by the waveform feature determining means. And a sound type determining means for determining.
- the information analysis method provides a waveform feature that indicates a reference for distinguishing a feature of a sound waveform from a sound waveform included in biological sound information collected by a stethoscope.
- a waveform feature determination step for specifying a characteristic of the sound waveform by applying a determination criterion, and a type of sound to which the biological sound information belongs based on the characteristic of the sound waveform specified in the waveform feature determination step
- a sound type determination step for determining the sound quality.
- FIG. 5B is a diagram illustrating an autocorrelation function derived by the autocorrelation analysis unit
- FIG. 5B is a diagram illustrating another example of the autocorrelation function derived by the autocorrelation analysis unit using another respiratory sound waveform as an input. It is.
- FIG. 4 It is a figure which shows one specific example of the autocorrelation function which the autocorrelation analysis part of an information analyzer outputs, and shows the autocorrelation function which the autocorrelation analysis part derived
- FIG. It is a figure which shows an example of the waveform characteristic determination reference which the periodicity determination part of an information analyzer references, and an example of the waveform characteristic determination result which a periodicity determination part outputs. It is a figure which shows a specific example of the spectrum which the Fourier-transform part of an information-analysis apparatus outputs, and is a figure which shows the spectrum derived
- FIG. 1 It is a figure which shows the other specific example of the body sound information acquired by the body sound acquisition part of an information analyzer, and is a figure which shows the respiratory sound of an asthma patient especially. It is a figure which shows a specific example of the spectrum which the Fourier-transform part of an information-analysis apparatus outputs, and is a figure which shows especially the spectrum which carried out the Fourier-transform of the respiratory sound of the asthma patient shown in FIG. It is a figure which shows an example of the waveform characteristic determination reference which the spectrum determination part of an information analyzer references, and an example of the waveform characteristic determination result which a spectrum determination part outputs.
- (A) is a figure which shows an example of an envelope with a high continuity
- (b) is a figure which shows an example of an envelope with a low continuity.
- Embodiment 1 Embodiments relating to the information analysis apparatus of the present invention will be described below with reference to FIGS.
- the auscultation system means that a subject's body sound is acquired by an electronic stethoscope, and the acquired electronic data, that is, body sound information is analyzed by the information analysis apparatus of the present invention and used for the subject's medical care. It is a system that makes it possible.
- a subject who is examined by an electronic stethoscope is referred to as a patient.
- the subject (patient) is assumed to be a human, but an auscultation system in which any living body other than a human being is a subject (patient) also falls within the scope of the present invention.
- the information analysis apparatus analyzes a patient's respiratory system sound (biological sound) information and determines a patient's condition related to lung disease.
- the information analysis apparatus of the present invention is not limited to the above-described example, but analyzes other body sound (heart sound, abdominal sound, intestinal sound, blood flow sound, fetal heart sound, etc.) information and relates to each part.
- the patient's condition may be determined.
- the information analysis apparatus of the present invention is not limited to the above-described example, and can be introduced into any other system that can acquire and use biological sound information from a living body for purposes other than medical treatment.
- FIG. 2 is a diagram showing an outline of the auscultation system in the embodiment of the present invention.
- the auscultation system 200 includes at least an electronic stethoscope 3 for the operator U to collect the body sound of the patient P (that is, auscultation), and an information analysis apparatus 100 used by the operator U during auscultation. And is built.
- the operator U is present at the medical treatment site 1 for treating the patient P, and performs medical treatment of the patient P at the medical treatment site 1 using various devices such as the electronic stethoscope 3.
- the various devices may include, for example, an oxygen saturation meter, an electrocardiograph, a blood pressure meter, a thermometer, an arteriosclerosis meter, a blood vessel health meter, and the like.
- the information analysis apparatus 100 and the electronic stethoscope 3 are connected to each other so that they can communicate with each other wirelessly or by wire.
- the operator U can operate the information analysis apparatus 100 to read out and refer to information necessary for medical treatment of the patient P, for example, information related to the patient P (such as an electronic medical record). Further, the operator U can store the body sound information collected from the electronic stethoscope 3 in the information analysis apparatus 100.
- the information analysis apparatus 100 is realized by an information processing terminal apparatus possessed by the operator U and excellent in portability, or a desktop personal computer (PC) installed in the clinical site 1.
- the information analysis apparatus 100 of the present invention is realized by a multifunctional mobile communication terminal such as a smartphone.
- the operator U When the operator U has specialized knowledge, skills, and authority as a doctor, the operator U examines the patient P using the electronic stethoscope 3 and the information analysis device 100, and A final diagnosis of the condition may be made and treated.
- the auscultation system 200 including the electronic stethoscope 3 and the information analysis apparatus 100 also falls within the scope of the present invention.
- the auscultation system 200 may be constructed to include the electronic stethoscope 3 and the information analysis device 100 in the clinical site 1 and include the management server 4 in the remote support center 2.
- the information analysis apparatus 100 and the management server 4 are connected to be communicable with each other via a communication network 5 such as the Internet.
- the operator U does not have advanced knowledge, skills, and authority as a doctor, or is a non-specialized medical treatment, but under the guidance of a professional doctor, the electronic stethoscope 3 and the information analysis device It is conceivable that the user has the skill to perform simple diagnosis and treatment immediately at the clinical site 1 by operating 100.
- an electronic stethoscope 3 and an information analysis apparatus 100 operated by an operator U who is an advanced nurse (NP; Nurse Practitioner) or other medical personnel are provided in the medical treatment site 1 of the auscultation system 200.
- the management server 4 which manages the electronic medical record of each patient managed in the said auscultation system 200 is provided.
- a doctor D having specialized knowledge and skills is stationed, and provides guidance to the operator U using an information processing terminal device or a communication device such as a telephone (not shown). Can support medical care.
- the body sound information directly collected from the patient P by the operator U using the electronic stethoscope 3 is stored in the management server 4 via the information analysis apparatus 100.
- the doctor D can access the management server 4 to obtain the body sound information of the patient P at a remote place, and can conduct examination and treatment.
- the operator U can perform a simple treatment under the guidance of the doctor D, or can introduce other cooperative hospitals that can be handled when the medical treatment site 1 cannot handle them.
- the information analysis device 100 realized by a smartphone has a function of analyzing body sound information collected from the electronic stethoscope 3 and outputting the analysis result to the own device or the management server 4.
- the information analysis apparatus 100 of the present invention having a function of analyzing biological sound information may be realized as the management server 4 in a remote place.
- FIG. 1 is a functional block diagram showing a main configuration of the information analysis apparatus 100 according to the present embodiment.
- the information analysis apparatus 100 includes at least a control unit 10, an input unit 11, a display unit 12, a storage unit 13, and a communication unit 14 as a hardware configuration. Furthermore, the information analysis apparatus 100 is not shown in order to realize a function that is inherent to a smartphone, an audio input unit, an external interface, an audio output unit, a call processing unit, a broadcast image receiving unit (such as a tuner / demodulation unit), You may provide various components with which the smart phone is equipped normally, such as GPS, a sensor (an acceleration sensor, an inclination sensor, etc.), an imaging part.
- the input unit 11 and the display unit 12 are integrally formed to constitute a touch panel.
- the display unit 12 may be realized by a liquid crystal display monitor
- the input unit 11 may be realized by a keyboard and a mouse.
- the input unit 11 is used for a user to input an instruction signal for operating the information analysis apparatus 100 via the touch panel.
- the input unit 11 determines a touch surface that receives contact of an indicator (such as a finger or a pen), contact / non-contact (approach / non-approach) between the indicator and the touch surface, and the contact (approach) position. It is comprised with the touch sensor for detecting.
- the touch sensor may be realized by any sensor as long as it can detect contact / non-contact between the indicator and the touch surface. For example, it is realized by a pressure sensor, a capacitance sensor, an optical sensor, or the like.
- the display unit 12 displays a result of processing the body sound information by the information analysis apparatus 100, or displays an operation screen for a user to operate the information analysis apparatus 100 as a GUI (Graphical User Interface) screen. is there.
- the display unit 12 is realized by a display device such as an LCD (Liquid Crystal Display).
- the information analysis device 100 may include an operation unit (not shown) that allows the user to directly input an instruction signal to the information analysis device 100.
- the operation unit is realized by an appropriate input mechanism such as a button, switch, key, or jog dial.
- the operation unit is a switch for turning on / off the power of the information analysis apparatus 100.
- the communication unit 14 communicates with external devices (such as the electronic stethoscope 3 and the management server 4).
- the communication unit 14 first includes a short-range communication unit for performing short-range communication with the electronic stethoscope 3.
- the short-range communication unit performs wireless communication with the electronic stethoscope 3 and receives biological sound information obtained by converting the biological sound collected by the electronic stethoscope 3 into a digital signal from the electronic stethoscope 3.
- the short-range communication unit is not particularly limited, but may realize any one of wireless communication means such as infrared communication such as IrDA and IrSS, Bluetooth (registered trademark) communication, WiFi communication, and a non-contact type IC card. However, a plurality of these means may be realized.
- the communication unit 14 includes a remote communication unit that performs data communication with a remote device (such as the management server 4) via the communication network 5 (LAN (Local Area Network), WAN (Wide Area Network), etc.). May be included.
- the remote communication unit can transmit the result of the analysis of the body sound information performed by the information analysis device 100 to the management server 4 via the communication network 5.
- the communication unit 14 has a function of transmitting / receiving voice call data, e-mail data, and the like to / from other devices via a mobile phone network. You may have.
- the storage unit 13 includes (1) a control program executed by the control unit 10 of the information analysis device 100, (2) an OS program executed by the control unit 10, and (3) various functions that the control unit 10 has in the information analysis device 100. And (4) various data to be read when the application program is executed. Alternatively, (5) the control unit 10 stores data used for calculation and calculation results in the course of executing various functions.
- the above data (1) to (4) include ROM (read only memory), flash memory, EPROM (Erasable Programmable ROM), EEPROM (registered trademark) (Electrically EPROM), HDD (Hard Disc Drive), etc. It is stored in a non-volatile storage device.
- the data (5) is stored in a volatile storage device such as a RAM (Random Access Memory).
- the collected body sound information of the patient P is temporarily stored in the RAM and read by the control unit 10 of the information analysis apparatus 100.
- the result of the analysis of the body sound information by the control unit 10 (and the body sound information as necessary) is stored in the storage unit 13 realized by a nonvolatile storage device such as a ROM.
- the control unit 10 performs overall control of each unit included in the information analysis apparatus 100.
- the control unit 10 is realized by, for example, a CPU (central processing unit), and the function of the information analysis apparatus 100 is such that the CPU as the control unit 10 reads a program stored in a ROM or the like into a RAM or the like. It is realized by executing.
- Various functions (particularly, information analysis function) realized by the control unit 10 will be described in detail below with reference to other drawings.
- control unit 10 of the information analysis apparatus 100 includes a biological sound acquisition unit 20, a biological sound processing unit 21, a biological sound analysis unit 22, and a result output unit 23 as functional blocks. is there.
- the body sound acquisition unit 20 acquires the body sound information of the patient P received from the electronic stethoscope 3 by the communication unit 14.
- the biological sound acquisition unit 20 temporarily stores the received biological sound information in the storage unit 13, reads the biological sound information as necessary, and supplies the information to downstream units (the biological sound processing unit 21 and the like).
- the biological sound processing unit 21 processes the sound waveform indicated by the biological sound information acquired by the biological sound acquisition unit 20, and extracts the waveform feature information of the sound waveform.
- Waveform feature information is a plot of a sound waveform included in the body sound information on a two-dimensional or higher-dimensional graph using various information of the sound waveform or each sound component constituting the sound waveform as an index. Shows what The various information included in the sound component includes, but is not limited to, a frequency, an amplitude value, and a generation time.
- the waveform feature information generated by the biological sound processing unit 21 enables the features of the sound waveform to be digitized from various viewpoints and easily understood as feature amounts based on various indexes.
- the extracted waveform feature information and the feature amount calculated from the waveform feature information are used by the body sound analysis unit 22 to analyze the sound waveform.
- the biological sound processing unit 21 is, for example, but not limited to, an autocorrelation analysis unit 211, a Fourier transform unit 212, a time frequency analysis unit 213, an envelope detection unit 214, and an impulse noise detection unit. This is realized by at least one of 215. Each unit of these biological sound processing units 21 derives respective waveform feature information. Details of each part will be described later.
- the body sound analysis unit 22 determines the state of the patient who has made the body sound based on the waveform feature information extracted by the body sound processing unit 21. More specifically, in the present embodiment, the body sound analysis unit 22 includes at least a waveform feature determination unit 30 and a sound type determination unit 40. The biological sound analysis unit 22 preferably further includes an abnormal level determination unit 50.
- the waveform feature determination unit 30 determines whether or not the extracted waveform feature information matches a waveform feature determination criterion. Based on this determination, the waveform features having the waveform feature information are classified and specified. To do.
- the waveform feature determination unit 30 may determine whether one piece of waveform feature information matches each of a plurality of criteria.
- the waveform feature determination unit 30 may determine whether or not each of a plurality of waveform feature information extracted from one sound waveform matches a reference.
- the waveform feature determination criteria are defined in advance and stored in the storage unit 13.
- the waveform feature determination unit 30 reads the waveform feature determination criterion stored in the storage unit 13 and determines whether or not the extracted waveform feature information satisfies the criterion.
- the sound waveform information whose features are classified by the waveform feature determination unit 30 in this way is output to the sound type determination unit 40 as a waveform feature determination result.
- the waveform feature determination result is used by the sound type determination unit 40 to determine the sound type of the sound waveform.
- the waveform feature determination unit 30 is, for example, but not limited to, a periodicity determination unit 31, a spectrum determination unit 32, a spectrogram determination unit 33, an envelope determination unit 34, and an impulse noise determination unit 35. It is implement
- the sound type determination unit 40 determines the sound type of the biological sound information having the sound waveform according to the waveform feature determination result output by the waveform feature determination unit 30.
- the sound type is a classification of sounds included in body sound information collected from a patient based on medical characteristics. That is, the sound type determination unit 40 is a unit that classifies sounds included in the body sound information based on medical characteristics by determining the sound type of the collected body sound information.
- the waveform feature determination unit 30 and the sound type determination unit 40 classify the sound of the patient's respiratory system sound based on the medical characteristics. Therefore, the biological sound analysis unit 22 can determine the patient's condition (medical condition) for the patient who is producing that type of respiratory sound.
- the information analysis apparatus 100 is an apparatus that analyzes respiratory system sounds as biological sounds. Therefore, for example, the sound type determination unit 40 can classify the respiratory system sound into the following sound types based on medical characteristics.
- the sound type determination unit 40 classifies the collected biological sounds into “breathing sounds (sounds associated with exhalation and sounds associated with inspiration)” and “noises (sounds other than exhalation inhalation that occur with a disease)”. May be.
- the sound type determination unit 40 may further classify the “breathing sound” into “normal breathing sound” and “abnormal breathing sound”.
- the sound type determination unit 40 further outputs “abnormal breathing sound”, “breathing sound attenuation (disappearance)”, “breathing sound enhancement”, “exhalation extension”, “bronchial breathing sound”, and “tracheal stenosis sound”. May be classified.
- the sound type determination unit 40 may further classify “noise” into “continuous noise”, “intermittent noise”, “pleural friction sound”, and “pulmonary vascular noise”. Furthermore, the sound type determination unit 40 may classify “continuous noise” into “high-pitched continuity noise” and “low-pitched continuity noise”. Alternatively, the sound type determination unit 40 may classify “intermittent noise” into “fine intermittent noise” and “rough intermittent noise”.
- the sound type determination unit 40 determines whether the “breathing sound” is “normal breathing sound” or “possibly not normal breathing sound”. It may be configured to return in binary whether it is applicable to the type or not.
- the occurrence mechanism of “attenuation of respiratory sounds” is as follows. For example, there may be a case where an obstacle such as pleural effusion is accumulated between the lung and the chest wall. If there is an obstacle before the breathing sound normally generated in the lungs reaches the stethoscope, this obstacle functions as a low-pass filter and cuts high-frequency components. Cases where there is an obstacle between the lung and the chest wall are common in patients with pleural effusion, pneumothorax, atelectasis, emphysema, etc.
- the information analysis apparatus 100 of the present invention can classify the body sound as “attenuation of breathing sound”, the operator is assumed to have pleural effusion, pneumothorax, atelectasis, emphysema, etc. Help U and Physician D judge.
- the generation mechanism of “continuous noise” is as follows. Due to the accumulation of secretions in the trachea, the airflow in the exhaled and inspiratory trachea is disturbed. This makes a noise. This noise will then continue to sound while exhalation and inspiration flow. This accumulation of secretions is common in patients with asthma, obstructive pulmonary disease (emphysema, chronic bronchitis, etc.), and tracheal / bronchial stenosis.
- the information analysis apparatus 100 of the present invention can classify the body sound as “continuous noise”, the disease affecting the patient is asthma, obstructive pulmonary disease (emphysema, chronic bronchitis, etc.), and It is useful for the operator U and the doctor D to determine that the disease is a trachea / bronchial stenosis.
- the frequency of the sound that sounds at a narrow airway becomes high. This sound can be classified as “high-pitched continuous noise”.
- the frequency of the sound produced when the airway is thick (the diameter of the trachea is large), that is, in the upper part of the lung (or in a shallow part where the branch of the trachea is not advanced) is low. This sound can be classified as “basic continuous noise”.
- the information analysis apparatus 100 of the present invention can classify the body sound as “high-pitched continuous noise” or “low-pitched continuous noise”, the region of the lung where the abnormality of continuous noise occurs ( It is useful for the operator U and the doctor D to determine whether the upper part or the lower part).
- the generation mechanism of “intermittent noise” is as follows. Liquid secretions in the trachea can put a thin film of fluid on the trachea and close the airway. When exhaled inspiration flows into the trachea in such a state, a sound of rupturing the membrane sounds. Membranes occur in the trachea, and a bursting sound is produced instantaneously only when the membrane is torn. In this regard, a distinctly different kind of sound is produced from “continuous noise”. The accumulation of liquid secretions is often seen in pneumonia patients. Therefore, if the information analysis apparatus 100 of the present invention can classify the body sound as “intermittent noise”, it is useful for the operator U and the doctor D to determine that the disease the patient is suffering from is pneumonia. .
- the information analysis apparatus 100 can classify the body sound into “fine intermittent noise” or “rough intermittent noise”, the region of the lung where the abnormal noise is generated (the upper part or the It is useful for the operator U and the doctor D to determine whether the lower part).
- the sound type determination unit 40 is, for example, but not limited to, a normal respiratory sound determination unit 41, a respiratory sound attenuation determination unit 42, a continuous noise determination unit 43, and an intermittent noise determination unit 44. It is implement
- the sound type determination result by the sound type determination unit 40 is supplied to the result output unit 23 or stored in the storage unit 13.
- the abnormal level determination unit 50 determines the magnitude (level) of the type of the sound waveform classified into a specific type based on the extracted waveform feature information. In particular, the abnormal level determination unit 50 determines the degree of abnormality (severity, progress, etc.) of the type of abnormal sound.
- the abnormal level determination unit 50 performs level determination by determining whether or not the extracted waveform feature information matches a reference. That is, the abnormal level determination unit 50 compares each of the level determination criteria with different threshold values in stages and the extracted waveform feature information, and according to which level determination criteria the waveform feature information matches. An abnormal level of the body sound is determined.
- the level determination criteria are defined in advance and stored in the storage unit 13.
- the abnormal level determination unit 50 determines that the abnormal level is “high” for a biological sound with a relatively high degree of abnormality (severe), and abnormal for a biological sound with a relatively low degree of abnormality (mild). The level may be determined as “low”. Further, the abnormal level determination unit 50 may determine that the abnormal level of the biological sound is “medium” as long as it is in the meantime.
- the abnormal level determination unit 50 is realized by at least one of, for example, but not limited to, an attenuation level determination unit 51, a continuity level determination unit 52, and an intermittent level determination unit 53. . Details of each part will be described later.
- the level determination result by the abnormal level determination unit 50 is supplied to the result output unit 23 or stored in the storage unit 13.
- the result output unit 23 outputs the sound type determination result output by the sound type determination unit 40 as an analysis result obtained by analyzing the body sound information.
- the control unit 10 includes the abnormal level determination unit 50
- the result output unit 23 includes the level determination result output by the abnormal level determination unit 50 and outputs the result.
- the analysis result output by the result output unit 23 is supplied to the display unit 12 as a video signal, and as a result, the analysis result is displayed on the display unit 12 so that the operator U can visually recognize it.
- the body sound processing unit 21 processes the body sound information to extract the waveform feature information from the sound waveform, and the waveform feature determination unit 30 matches what determination criteria the waveform feature information matches. (Or does not match).
- the sound type determination unit 40 can determine the type of sound according to the waveform feature determination result.
- the sound type determination result performed by the sound type determination unit 40 is displayed on the display unit 12 as an analysis result.
- the sound type determination unit 40 can determine to which sound type the original body sound information has a high correlation depending on whether or not the extracted waveform feature information matches the above-described determination criteria.
- Each functional block of the control unit 10 described above is stored in a storage device (storage unit 13) in which a CPU (central processing unit) or the like is realized by ROM (read only memory), NVRAM (non-Volatile random access memory), or the like. This can be realized by reading the program stored in a RAM (random access memory) or the like and executing it.
- ROM read only memory
- NVRAM non-Volatile random access memory
- Periodity judgment function 3 and 4 are diagrams illustrating a specific example of the body sound information acquired by the body sound acquisition unit 20.
- FIG. 5 (a) and 5 (b) and FIG. 6 are diagrams showing a specific example of the autocorrelation function output from the autocorrelation analysis unit 211.
- the autocorrelation analysis unit 211 of the body sound processing unit 21 analyzes a sound waveform included in the body sound information acquired by the body sound acquisition unit 20 and derives an autocorrelation function.
- the periodicity determination unit 31 of the waveform feature determination unit 30 applies the waveform feature determination criterion to the autocorrelation function output by the autocorrelation analysis unit 211, and the sound wave feature having the autocorrelation function (in particular, periodicity). ).
- the autocorrelation analysis unit 211 is means for analyzing the periodic signal. Autocorrelation is an index for evaluating the degree of correlation between a signal v (t) and a signal v (t + ⁇ ) obtained by time-shifting the signal itself, and is expressed as a function R ( ⁇ ) with time shift ⁇ as a variable. It can be expressed by a formula.
- the autocorrelation analysis unit 211 supplies the derived autocorrelation function to the periodicity determination unit 31 as waveform feature information.
- FIG. 3 is a diagram showing the breathing sound of a healthy person.
- FIG. 5A is a diagram showing an autocorrelation function derived by the autocorrelation analysis unit 211 using the waveform of the respiratory sound shown in FIG. 3 as an input.
- FIG. 5B is a diagram illustrating another example of the autocorrelation function derived by the autocorrelation analysis unit 211 using another respiratory sound waveform as an input.
- the autocorrelation on the vertical axis is normalized by the maximum amplitude.
- the periodicity determination unit 31 determines whether or not the autocorrelation function matches the waveform feature determination criterion.
- the periodicity determination unit 31 determines the strength of the periodicity of the sound waveform and the length (feature value) of one period when there is a period.
- the periodicity determination unit 31 detects peaks with an interval of about 3 seconds from the autocorrelation function shown in FIG. 5A, and finds a periodicity with a period of about 3 seconds. Or the periodicity determination part 31 detects the peak of about 2 second interval from the autocorrelation function shown in FIG.5 (b), and finds the periodicity whose 1 period is about 2 second.
- the periodicity determination unit 31 may also determine the degree of the periodicity according to the ratio of the autocorrelation other than the peak of the autocorrelation (the stronger the periodicity, the more the peak and the other). Ratio increases). For example, the periodicity determination unit 31 may determine what percentage of the respiratory cycle the peak width (period) at 1 ⁇ 4 of the peak amplitude value of the envelope in the autocorrelation function. A smaller value (feature value) means stronger periodicity.
- FIG. 4 is a diagram showing respiratory sounds of a pneumonia patient.
- FIG. 6 is a diagram showing an autocorrelation function derived by the autocorrelation analysis unit 211 using the waveform of the respiratory sound shown in FIG. 4 as an input. In the example shown in FIG. 6, the autocorrelation on the vertical axis is normalized with the maximum amplitude.
- the periodicity determination unit 31 can determine that the periodicity of the sound waveform is weak when such an autocorrelation function is input.
- the periodicity determination unit 31 determines the intensity of periodicity and the details of the sound waveform of the body sound that is the diagnosis target. It is possible to evaluate how strong (weak) it is.
- the periodicity determination unit 31 reads the waveform feature determination criterion stored in the storage unit 13 and applies it to the autocorrelation function. Then, it is determined whether or not the characteristics of the autocorrelation function (here, the strength of periodicity and the length of the period) match the above-described waveform feature determination criteria. Thereby, the periodicity determination part 31 becomes possible [specifying the characteristic which concerns on the periodicity of the sound waveform which has the said autocorrelation function].
- FIG. 7 is a diagram illustrating an example of a waveform feature determination criterion referred to by the periodicity determination unit 31 and an example of a waveform feature determination result output from the periodicity determination unit 31.
- the periodicity determination unit 31 can execute “determination item 1” or “determination item 1 ′” and output a waveform feature determination result in accordance with the waveform feature determination criterion shown in FIG. And the periodicity determination part 31 outputs true or false binary as a waveform feature determination result about each determination item.
- the content shown in FIG. 7 is an example for explaining the function of the periodicity determining unit 31, and is not intended to limit the configuration of the periodicity determining unit 31.
- the threshold value (value between “** _” and “_ **”) defined in the waveform feature criterion shown in FIG. 7 is set by a user (such as operator U) of the information analysis apparatus 100. The configuration may be arbitrarily changed and set. Further, the periodicity determination unit 31 can output the waveform feature determination result with more detailed contents instead of the true / false binary value.
- the periodicity determination unit 31 can determine the strength of the periodicity of the biological sound waveform by executing “determination item 1” shown in FIG. In “determination item 1”, the periodicity determination unit 31 returns “strong periodicity” as true and “false periodicity” as false.
- the periodicity determination unit 31 executes “determination item 1-1” in “determination item 1”. That is, it is determined whether or not the waveform of the autocorrelation function has a peak at intervals of 2 to 5 seconds. The periodicity determination unit 31 returns true if a peak is detected every 2 to 5 seconds, and returns false if no peak is detected.
- the periodicity determination unit 31 executes “determination item 1-2”. That is, whether or not the width of the peak (horizontal axis; time) is 10% or less of the respiratory cycle at an amplitude value that is 1/4 of the peak amplitude value (vertical axis) in the envelope of the autocorrelation function. judge.
- the periodicity determination unit 31 returns true when the peak width is 10% or less (when the periodicity is strong) and returns false when the peak width is greater than 10% (when the periodicity is weak).
- the periodicity determination unit 31 determines that “determination item 1-2” is true if the widths of the envelope peaks at the amplitude value of 0.2 average 0.5 seconds or less.
- the periodicity determination unit 31 integrates the results of “determination item 1-1” and “determination item 1-2” and outputs the waveform feature determination result of “determination item 1”.
- the periodicity determination unit 31 sets “determination item 1” to “true” (that is, the cycle) when “determination item 1-1” is true and “determination item 1-2” is true. Is determined to be strong).
- “judgment item 1” is judged as “false (that is, periodicity is weak)” .
- the periodicity determination unit 31 sets “determination item 1 ′” to “true” (that is, the cycle) when at least one of “determination item 1-1” and “determination item 1-2” is false. Is weak). On the other hand, in other cases, that is, when both “determination item 1-1” and “determination item 1-2” are true, the determination is “false (strong periodicity)”.
- the periodicity determination unit 31 outputs a true or false value of “determination item 1” or “determination item 1 ′” to the sound type determination unit 40 as a waveform feature determination result.
- FIG. 9 is a diagram illustrating another specific example of the biological sound information acquired by the biological sound acquisition unit 20.
- FIG. 8 and 10 are diagrams showing a specific example of a spectrum output from the Fourier transform unit 212.
- FIG. 8 and 10 are diagrams showing a specific example of a spectrum output from the Fourier transform unit 212.
- the Fourier transform unit 212 of the biological sound processing unit 21 analyzes a sound waveform included in the biological sound information acquired by the biological sound acquisition unit 20 and derives a spectrum.
- the spectrum determination unit 32 of the waveform feature determination unit 30 applies a waveform feature determination criterion to the spectrum output by the Fourier transform unit 212, and determines the feature (particularly, the feature related to the frequency component) of the spectrum. More specifically, the spectrum determination unit 32 determines whether the frequency component distribution found in the spectrum shows a normal tendency or an abnormal (with noise) tendency. .
- the frequency component information varies depending on the presence / absence of disease, disease type, disease degree, and the like.
- the Fourier transform unit 212 performs Fourier analysis.
- the Fourier transform unit 212 supplies the spectrum derived from the sound waveform to the spectrum determination unit 32 as waveform feature information.
- FIG. 8 is a diagram illustrating a spectrum derived by Fourier transforming the respiratory sound of the healthy person illustrated in FIG. 3 by the Fourier transform unit 212.
- FIG. 9 is a diagram showing respiratory sounds of asthmatic patients.
- FIG. 10 is a diagram showing a spectrum derived by Fourier transforming the respiratory sound of the asthma patient shown in FIG.
- the sound component of the biological sound waveform collected for a predetermined second for example, 20 seconds is Fourier-transformed.
- the spectrum determination unit 32 can determine the presence or absence of a disease (for example, suspected asthma) or the presence or absence of noise. Become.
- the spectrum determination unit 32 reads the waveform feature determination reference stored in the storage unit 13 and applies it to the spectrum. Then, it is determined whether or not the spectrum meets the waveform feature determination criterion. For example, specifically, the ratio of the signal component amount below 200 Hz from the spectrum to the total signal component amount is calculated as a feature amount, and this feature amount is compared with the threshold included in the waveform feature criterion. To do.
- the spectrum determination unit 32 can classify and specify the characteristics related to the frequency component of the sound waveform having the spectrum.
- FIG. 11 is a diagram illustrating an example of a waveform feature determination criterion referred to by the spectrum determination unit 32 and an example of a waveform feature determination result output from the spectrum determination unit 32.
- the spectrum determination unit 32 executes the determination of “determination item 2-A” or “determination item 2-B” in accordance with the waveform feature determination criterion shown in FIG. Can be output. Then, the spectrum determination unit 32 outputs a true or false binary value as a waveform feature determination result for each determination item.
- the content shown in FIG. 11 is an example for explaining the function of the spectrum determination unit 32, and is not intended to limit the configuration of the spectrum determination unit 32.
- the threshold (defined between “** _” and “_ **”) defined in the waveform feature criterion shown in FIG. 11 is set by a user (such as operator U) of the information analysis apparatus 100. The configuration may be arbitrarily changed and set. Further, the spectrum determination unit 32 can output the waveform feature determination result with more detailed contents instead of the true / false binary value.
- the spectrum determination unit 32 executes “determination item 2-A” shown in FIG.
- the spectrum determination unit 32 determines whether or not the sum of frequency components of 200 Hz or less occupies 80% or more of all components in the spectrum. As shown in FIG. 8, when the sum of frequency components of 200 Hz or less occupies 80% or more of all components, it can be estimated that the spectrum is close to normal.
- the spectrum determination unit 32 is true when the sum of frequency components of 200 Hz or less of the spectrum occupies 80% or more of all components (false), and false when it is less than 80%. Is output to the sound type determination unit 40 as a waveform feature determination result.
- the spectrum determination unit 32 executes “determination item 2-B” shown in FIG.
- the spectrum determination unit 32 determines whether or not the sum of frequency components of 200 Hz or more occupies 30% or more of all components in the spectrum. As shown in FIG. 10, when there are many frequency components of 200 Hz or more, it can be recognized as a sign of abnormality.
- the spectrum determination unit 32 is true when the sum of the frequency components of 200 Hz or higher in the spectrum occupies 30% or more of all components (there is an abnormal sign), and false when it is less than 30%. (No abnormality sign) is output to the sound type determination unit 40 as a waveform feature determination result.
- FIG. 12 to 15 are diagrams illustrating specific examples of spectrograms output from the time-frequency analysis unit 213.
- FIG. 1 A diagram illustrating specific examples of spectrograms output from the time-frequency analysis unit 213.
- the time frequency analysis unit 213 of the body sound processing unit 21 analyzes a sound waveform included in the body sound information acquired by the body sound acquisition unit 20 in a predetermined time unit and derives a spectrogram.
- the spectrogram determination unit 33 of the waveform feature determination unit 30 applies a waveform feature determination criterion to the spectrogram output by the time-frequency analysis unit 213 to determine the feature of the spectrogram. Specifically, the spectrogram determination unit 33 specifies a frequency at which periodicity is recognized (or is not recognized) as a feature amount, or determines the strength of periodicity for each frequency band.
- the spectrum output by the Fourier transform unit 212 is a two-dimensional graph with the frequency component amount (intensity) on the vertical axis and the frequency on the horizontal axis. Since the time information is missing from the spectrum, it is not possible to observe changes with time in the frequency component amount for each frequency band.
- the spectrogram output from the time-frequency analysis unit 213 is a three-dimensional graph taking time information into consideration.
- a spectrogram may be created by plotting a frequency component amount expressed in color on a two-dimensional graph with frequency on the vertical axis and time on the horizontal axis.
- the color is closer to the red direction (the color direction of the uppermost part of the legend), and the darker point (area) has a larger component amount, and the blue direction (the lowermost part of the legend) The closer the color is to (the direction of the color), the darker the dots (regions), the smaller the component amount.
- the time-frequency analysis unit 213 divides a 20-second sound waveform every predetermined seconds (for example, 0.5 seconds), and performs a Fourier transform for each divided 0.5-second interval to derive a spectrogram.
- the time frequency analysis unit 213 supplies the spectrogram derived from the sound waveform to the spectrogram determination unit 33 as waveform feature information.
- the spectrogram determination unit 33 can analyze a change in the component amount for each frequency band with the passage of time. That is, the spectrogram determination unit 33 can determine the presence or absence (or strength) of periodicity for each frequency band.
- FIG. 12 is a diagram showing a spectrogram derived from the breathing sound of a healthy person by the time frequency analysis unit 213 performing a short time frequency analysis.
- FIG. 13 is a diagram showing a spectrogram derived by performing a short-time frequency analysis by the time frequency analysis unit 213 on the respiratory sound in which the respiratory sound is attenuated.
- FIG. 14 is a diagram showing a spectrogram derived by performing a short-time frequency analysis on the breathing sound in which continuity noise is generated by the time-frequency analysis unit 213.
- FIG. 15 is a diagram showing a spectrogram derived by performing a short-time frequency analysis on the respiratory sound in which intermittent noise is generated by the time-frequency analysis unit 213.
- the spectrogram determination unit 33 analyzes the spectrogram of FIG. 12 and recognizes that strong periodicity is recognized even in a band of 400 Hz or higher. That is, the spectrogram determination unit 33 detects that timings (relatively dark portions) at which a certain amount or more of signal components are generated are seen at intervals of about 3 seconds in a band of 400 Hz or higher. Thereby, the spectrogram determination unit 33 can determine that the spectrogram of FIG. 12 is “periodicity is recognized even in a band of 400 Hz or higher”.
- the spectrogram determining unit 33 determines that the spectrogram is “periodicity is finally recognized (intensified) in a band of 200 Hz to 300 Hz (not periodicity is not recognized at 400 Hz)”. Can do.
- the signal component in the high frequency region is not sufficiently observed.
- Abnormal signs of respiratory sound attenuation are common when pleural effusion accumulates between the lungs and the chest wall. This is because there is pleural effusion before the breathing sound normally generated in the lungs reaches the stethoscope, and this pleural effusion functions as a low-pass filter and cuts high-frequency components.
- the periodicity is weak or not recognized in any frequency band.
- the spectrogram determination unit 33 may determine that the characteristic relating to the periodicity of the sound waveform having the spectrogram is “weak periodicity”.
- the periodicity determination unit 31 can determine the periodicity intensity from the autocorrelation function. Therefore, when the waveform feature determination unit 30 includes the periodicity determination unit 31, the spectrogram determination unit 33 does not necessarily determine the periodicity strength.
- the spectrogram determination unit 33 reads the waveform feature determination criteria stored in the storage unit 13 and applies them to the spectrogram. Then, it is determined whether or not the spectrogram meets the waveform feature criterion. Thereby, the spectrogram determination unit 33 can specify the characteristics related to the time-frequency component of the sound waveform having the spectrogram.
- FIG. 16 is a diagram illustrating an example of a waveform feature determination criterion referred to by the spectrogram determination unit 33 and an example of a waveform feature determination result output from the spectrogram determination unit 33.
- the spectrogram determination unit 33 performs the determination of “determination item 3-A” or “determination item 3-B” in accordance with the waveform feature determination criterion shown in FIG. Can be output. Then, the spectrogram determination unit 33 outputs a true or false binary value as a waveform feature determination result for each determination item.
- the content shown in FIG. 16 is an example for explaining the function of the spectrogram determination unit 33, and is not intended to limit the configuration of the spectrogram determination unit 33.
- the threshold value defined in the waveform feature determination standard (the value between “** _” and “_ **”) shown in FIG. 16 is determined by the user (such as operator U) of the information analysis apparatus 100. The configuration may be arbitrarily changed and set.
- the spectrogram determination unit 33 can output the waveform feature determination result with more detailed contents instead of the true / false binary value.
- the spectrogram determination unit 33 executes “determination item 3-A” shown in FIG. In “determination item 3-A”, the spectrogram determination unit 33 determines whether or not periodicity of a frequency component (dark portion) of a certain amount or more is recognized at a spectrogram frequency of 400 Hz (or higher). The spectrogram determination unit 33 determines that “determination item 3-A” is true when periodicity is recognized in a band of 400 Hz or higher, false if the periodicity is not recognized, and the waveform feature determination result as sound. The data is output to the type determination unit 40.
- the sound type determination unit 40 can determine that the sound type determination unit 40 is close to normal when strong periodicity is observed in a band of 400 Hz or higher as shown in FIG. .
- the periodicity is weak or not recognized (false) in a band of 400 Hz or higher, it is determined that there is a possibility of abnormality (particularly, attenuation of respiratory sounds, noise, etc.). be able to.
- the spectrogram determination unit 33 executes “determination item 3-B” shown in FIG.
- the spectrogram determination unit 33 scans the spectrogram so as to shift from the high frequency band (the scan start point may be about 500 to 400 Hz) to the low frequency band. Identify the frequency when it is no longer allowed (or weakened). Then, the spectrogram determination unit 33 determines that “determination item 3-B” is true if the frequency when periodicity is recognized is less than 400 Hz, and false if it is 400 Hz or more.
- the sound type determination unit 40 can determine that there is a low possibility that the biological sound waveform is a respiratory sound attenuation when a false determination result is output for “determination item 3-B”.
- the frequency when strong periodicity comes to be seen is less than 400 Hz, the strong periodicity seen in the low frequency band will be weakened (or not recognized) in the high frequency band. It is. Therefore, when the true determination result is output for “determination item 3-B”, the sound type determination unit 40 can determine that there is a high possibility that the biological sound waveform is the attenuation of the respiratory sound.
- the spectrogram determination unit 33 may specify the frequency when the periodicity is weakened and cannot be seen when scanning the spectrogram so as to shift from the low frequency band (0 Hz) to the high frequency band. Then, the spectrogram determining unit 33 determines that “determination item 3-B” is true if the frequency when the periodicity is weakened and is no longer seen is less than 400 Hz, and false if it is 400 Hz or more.
- the sound type determination unit 40 has a periodicity of less than 400 Hz as shown in FIG. 13 (if “determination item 3-B” is true). It can be determined that the periodicity is strong but the possibility of attenuation is high. On the other hand, as shown in FIG. 12, if the periodicity is weak and cannot be seen 400 Hz or higher (for example, 900 Hz), that is, if “determination item 3-B” is false, there is a possibility of attenuation. Is low and can be determined to be normal.
- the time-frequency analysis unit 213 is configured to perform Fourier transform for each section having a fixed time resolution, that is, a fixed number of seconds (such as 0.5 seconds).
- the configuration of 213 is not limited to this.
- the time frequency analysis unit 213 may perform wavelet transform to derive a time frequency component distribution. In the wavelet transform, the time resolution for a low frequency and the time resolution for a high frequency can be made different, and a more detailed time frequency component distribution can be obtained.
- FIG. 17 is a diagram illustrating a specific example of a biosonic waveform envelope output from the envelope detector 214. Note that the biological sound waveform shown in FIG. 17 is an enlarged part of the sound waveform of the biological sound information shown in FIG.
- the envelope detection unit 214 of the biological sound processing unit 21 detects and outputs a sound wave envelope included in the biological sound information acquired by the biological sound acquisition unit 20.
- the envelope determination unit 34 of the waveform feature determination unit 30 analyzes the envelope of the sound waveform output by the envelope detection unit 214, applies the waveform feature determination criterion, and determines the sound waveform feature based on the envelope. Judgment.
- the biological sound information is converted into a high-frequency signal as in the case of AM modulation or FM modulation in communication technology. It can be seen that it is collected. In such a case, it is desirable to use a technique called envelope detection in order to determine whether or not there is continuous noise of 200 ms or longer.
- the envelope detection performed by the envelope detection unit 214 is a technique used for demodulation of an AM-modulated signal, and is a technique for extracting an envelope of a high-frequency signal.
- the envelope detector 214 detects the envelope from the biological sound wave that is a high-frequency signal, and outputs this to the envelope determination unit 34.
- the envelope determination unit 34 analyzes the waveform of the envelope detected by the envelope detection unit 214, and identifies a sound waveform feature (for example, the length of continuous noise) as a feature amount based on the envelope. It becomes possible to do.
- a sound waveform feature for example, the length of continuous noise
- FIG. 18 is a diagram illustrating an example of a waveform feature determination criterion referred to by the envelope determination unit 34 and an example of a waveform feature determination result output from the envelope determination unit 34.
- FIG. 19A is a diagram illustrating a specific example of an envelope having high continuity
- FIG. 19B is a diagram illustrating a specific example of the envelope having low continuity.
- the envelope determination unit 34 can execute “determination item 4” according to the waveform feature determination criterion shown in FIG. 18 and output the waveform feature determination result. And the envelope determination part 34 outputs true or false binary as a waveform characteristic determination result about the said determination item.
- the content shown in FIG. 18 is an example for explaining the function of the envelope determination unit 34 and is not intended to limit the configuration of the envelope determination unit 34.
- the threshold value (value between “** _” and “_ **”) defined in the waveform feature determination criteria shown in FIG. 18 is determined by the user (such as operator U) of the information analysis apparatus 100. The configuration may be arbitrarily changed and set.
- the envelope determination unit 34 can output the waveform feature determination result with more detailed contents instead of the true binary value.
- the envelope determination unit 34 executes “determination item 4” shown in FIG. In “determination item 4”, the envelope determination unit 34 determines whether or not continuity of sound is recognized for the envelope of the sound waveform.
- the envelope determination unit 34 first executes “determination item 4-1.” In “decision item 4-1,” the envelope determination unit 34 determines whether or not the continuous time exceeding the average amplitude value of the sound waveform envelope is 200 ms or more.
- the envelope shown in FIG. 19A will be described as an example.
- the amplitude average value of the envelope is indicated by a one-dot chain line Avr1.
- the envelope determination unit 34 specifies a section exceeding the amplitude average value Avr1 as Z1.
- the section Z1 continues for 200 ms or more. Therefore, when “determination item 4-1” is executed for the envelope shown in FIG. 19A, the envelope determination unit 34 outputs “true (200 ms or longer)” as the waveform feature determination result.
- the envelope shown in FIG. 19B will be described as an example.
- the amplitude average value of the envelope is indicated by a one-dot chain line Avr2.
- the envelope determination unit 34 specifies sections that exceed the amplitude average value Avr2 as Z2, Z3, and Z4. None of the sections Z2, Z3, and Z4 continues for 200 ms or more. Therefore, when “determination item 4” is executed for the envelope shown in FIG. 19B, the envelope determination unit 34 outputs “false (not longer than 200 ms)” as the waveform feature determination result.
- the envelope determination unit 34 executes “determination item 4-2”.
- “Decision item 4-2” the envelope determination unit 34 determines that the total time exceeding the amplitude average value of the envelope of the sound waveform is 200 ms or more in the sound waveform of one breath period (about 2 to 5 seconds). It is determined whether or not. For example, the envelope determination unit 34 sums the times of the sections exceeding the amplitude average value Avr2 in the envelope for one cycle of breathing. In the example shown in FIG. 19B, the envelope determination unit 34 sums up the times of the sections Z2, Z3, Z4, and the next section. According to “determination item 4-2”, the envelope determination unit 34 returns “true”, unlike “determination item 4-1,” if the total time is 200 ms or more.
- the envelope determination unit 34 then outputs “true (200 ms or more)” or “false (less than 200 ms)” as the waveform feature determination result of “determination item 4-2”.
- the envelope determination unit 34 combines both the “determination item 4-1” and the “determination item 4-2” and outputs the waveform feature determination result for the “determination item 4”. For example, the envelope determination unit 34 determines the waveform feature of “determination item 4” based on the envelope when at least one of “determination item 4-1” and “determination item 4-2” is “true”. The result may be output as “true; continuity of sound is recognized”. Further, the envelope determination unit 34 displays the waveform characteristic determination result of “determination item 4” based on the envelope when both “determination item 4-1” and “determination item 4-2” are “false”. “False: No continuity of sound is recognized”.
- the sound type determination unit 40 determines that the noise continuity is high, that is, there is a possibility of continuity noise when “determination item 4” is “true”. can do. On the other hand, in the case of “false”, it can be determined that there is a possibility that the continuity of noise is low, that is, there is no continuity noise. In this way, the envelope determination unit 34 integrates both “determination item 4-1” and “determination item 4-2” and outputs the waveform feature determination result, so that the sound type determination unit 40 has a higher accuracy. Well, it is possible to execute the type determination for the continuity of sound.
- FIG. 20 is a diagram illustrating a specific example of the impulse noise detection result in which the impulse noise is specified in the biological sound waveform output from the impulse noise detection unit 215.
- the impulse noise detection unit 215 of the biological sound processing unit 21 detects impulse noise included in the sound waveform of the biological sound information acquired by the biological sound acquisition unit 20.
- the impulse noise detection unit 215 supplies an impulse noise detection result indicating the result of detecting the impulse noise to the impulse noise determination unit 35.
- the impulse noise determination unit 35 of the waveform feature determination unit 30 applies a waveform feature determination criterion to the impulse noise detection result supplied from the impulse noise detection unit 215, and generates a sound waveform based on the number (feature amount) of impulse noise. The feature is determined.
- the impulse noise detection result may have a data structure in which the impulse noise in the biological sound waveform is emphasized and can be recognized by the impulse noise determination unit 35 as shown in FIG. It may be information indicating whether the impulse noise is detected.
- Impulse noise is a plosive sound that occurs instantaneously. This burst sound is generated when the liquid film closes the airway and the liquid film ruptures when the airflow passes through. Therefore, it is considered that a patient who emits a breathing sound in which many impulse noises are detected suffers from a disease (pneumonia, sputum retention, etc.) in which the airway is closed by the liquid film.
- FIG. 21 is a diagram illustrating an example of a waveform feature determination criterion referred to by the impulse noise determination unit 35 and an example of a waveform feature determination result output from the impulse noise determination unit 35.
- the impulse noise determination unit 35 can execute “determination item 5” according to the waveform feature determination criterion shown in FIG. 21 and output the waveform feature determination result. And the impulse noise determination part 35 outputs true or false binary as a waveform characteristic determination result about the said determination item.
- the content shown in FIG. 21 is an example for explaining the function of the impulse noise determination unit 35 and is not intended to limit the configuration of the impulse noise determination unit 35.
- the threshold value defined in the waveform feature determination standard (the value between “** _” and “_ **”) shown in FIG. 21 is set by a user (such as operator U) of the information analysis apparatus 100. The configuration may be arbitrarily changed and set. Further, the impulse noise determination unit 35 can output the waveform feature determination result with more detailed contents instead of the true binary value.
- the impulse noise determination unit 35 executes “determination item 5” shown in FIG. In “determination item 5”, the impulse noise determination unit 35 determines whether or not the number of impulse noises included in the sound waveform is 10 or more per cycle.
- the impulse noise determination unit 35 outputs true to the sound type determination unit 40 as a waveform feature determination result when the number of impulse noises per period is 10 or more, and false when the number is less than 10.
- the sound type determination unit 40 can determine that the discontinuity of noise is high when “determination item 5” is “true” based on the waveform feature determination result. On the other hand, in the case of “false”, it can be determined that the discontinuity of noise is low.
- FIG. 22 is a diagram illustrating a specific example of the sound type determination result output from the normal breathing sound determination unit 41 of the sound type determination unit 40 using the waveform feature determination result output by the waveform feature determination unit 30 as an input.
- the normal respiratory sound determination unit 41 of the sound type determination unit 40 determines whether or not the biological sound included in the biological sound information acquired by the biological sound acquisition unit 20 is classified as a “normal respiratory sound”. is there. More specifically, in the present embodiment, the normal breathing sound determination unit 41 represents the biological sound “true: may be a normal breathing sound” or “false: may not be a normal breathing sound” 2 The value information is output as the sound type determination result. The output sound type determination result is supplied to the result output unit 23.
- the normal breathing sound determination unit 41 uses the waveform feature determination results of “determination item 1”, “determination item 2-A”, and “determination item 3-A” to determine true / false. Obtained from the feature determination unit 30.
- the normal breathing sound determination unit 41 acquires the waveform feature determination result of “determination item 1” representing the strength of periodicity from the periodicity determination unit 31.
- the normal breath sound determination unit 41 acquires the waveform feature determination result of “determination item 2-A” indicating the normality of the frequency component distribution from the spectrum determination unit 32. Further, the normal breath sound determination unit 41 acquires from the spectrogram determination unit 33 the waveform feature determination result of “determination item 3-A” indicating the presence or absence (or strength) of periodicity in the high frequency band.
- the true / false patterns are eight patterns (A) to (K) as shown in FIG.
- the normal breathing sound determination unit 41 determines whether the normal breathing sound is true or false for each of the eight patterns.
- the normal breathing sound determination unit 41 is “true: there is a possibility that it is a normal breathing sound” only in the case of the pattern (A) in which all the determination items are true. judge. If any one of the three determination items is false, it is determined that “false: there is a possibility that it is not a normal breathing sound”.
- a body sound (respiratory system sound) with “true” as “judgment item 1” is considered to have “strong periodicity”, and a body sound with “true” as “judgement item 2-A” It is considered that “the frequency component distribution is close to normal”, and the biological sound whose “determination item 3-A” is “true” is considered “having periodicity (or strong) in a high frequency band”. Therefore, in this embodiment, the normal breathing sound determination unit 41 concludes that the biological sound in which all of these are true is “true: there is a possibility that it is a normal breathing sound”.
- a biological sound whose “judgment item 1” is “false” is considered to be “poor periodicity”
- a biological sound whose “judgment item 2-A” is “false” is “normal frequency component distribution.
- a biological sound with “Decision item 3-A” of “Fake” is considered to be “No periodicity (or weak) in the high frequency band”. Therefore, in the present embodiment, the normal breathing sound determination unit 41 determines that a biological sound that is false in any one of these determination items may have some abnormality. It is possible.
- the sound type determination result output by the normal breathing sound determination unit 41 is displayed on the display unit 12 by the result output unit 23.
- the result output unit 23 displays a message such as “There is a possibility of a normal breathing sound” on the display unit 12. Show it.
- the result output unit 23 may display a message such as “There is a possibility that it is not a normal breathing sound” on the display unit 12.
- FIG. 23 is a diagram illustrating a specific example of the sound type determination result output by the respiratory sound attenuation determination unit 42 of the sound type determination unit 40 using the waveform feature determination result output by the waveform feature determination unit 30 as an input.
- the respiratory sound attenuation determination unit 42 of the sound type determination unit 40 determines whether or not the biological sound included in the biological sound information acquired by the biological sound acquisition unit 20 is classified as “respiratory sound attenuation”. is there. More specifically, in the present embodiment, the respiratory sound attenuation determination unit 42 indicates that the biological sound is “true: there is a possibility that the respiratory sound is attenuated” or “false: there is a possibility that the respiratory sound is not attenuated” 2. The value information is output as the sound type determination result. The output sound type determination result is supplied to the result output unit 23.
- the respiratory sound attenuation determination unit 42 uses the waveform feature determination results of “determination item 1”, “determination item 2-A”, and “determination item 3-B” in order to determine authenticity. Obtained from the feature determination unit 30.
- the respiratory sound attenuation determination unit 42 acquires the waveform feature determination result of “determination item 1” representing the strength of periodicity from the periodicity determination unit 31.
- the respiratory sound attenuation determination unit 42 acquires the waveform characteristic determination result of “determination item 2-A” indicating the normality of the frequency component distribution from the spectrum determination unit 32. Further, the respiratory sound attenuation determination unit 42 acquires from the spectrogram determination unit 33 the waveform feature determination result of “determination item 3-B” indicating whether or not the strong periodicity seen in the low frequency band is weakened in the high frequency band. To do.
- the respiratory sound attenuation determination unit 42 determines the authenticity of the respiratory sound attenuation for each of the eight patterns.
- the respiratory sound attenuation determination unit 42 is “true: there is a possibility that the respiratory sound is attenuated” only in the case of the pattern (A) in which all the determination items are true. judge. If any one of the three determination items is false, it is determined that “false: breathing sound may not be attenuated”.
- “not breathing sound attenuation” may indicate that the breathing sound is normal, or may indicate that there is an abnormality other than the attenuation of the breathing sound.
- a body sound (respiratory system sound) with “true” as “judgment item 1” is considered to have “strong periodicity”, and a body sound with “true” as “judgement item 2-A”
- a biological sound whose “frequency component distribution is close to normal” and whose “judgment item 3-B” is “true” is “a periodicity is recognized in the low frequency band but periodic in the high frequency band. Is no longer seen (becomes weak) ”.
- the characteristic that periodicity is recognized in the low frequency band but periodicity is not recognized (becomes weak) in the high frequency band is a typical sign of respiratory sound attenuation. Therefore, in this embodiment, the respiratory sound attenuation determination unit 42 concludes that the biological sound in which all of these are true is “true: there is a possibility that the respiratory sound is attenuated”.
- the respiratory sound attenuation determination unit 42 recognizes that a biological sound that is false in any one of these determination items has characteristics different from the signs of respiratory sound attenuation. It may not be sound attenuation. " In addition, it is thought that the cause different from the sign of respiratory sound attenuation is that the respiratory sound is normal in the first place or there is an abnormality other than the attenuation of respiratory sound.
- the sound type determination result output from the respiratory sound attenuation determination unit 42 is displayed on the display unit 12 by the result output unit 23.
- the result output unit 23 displays a message such as “There is a possibility that the respiratory sound is attenuated” on the display unit 12. Show it.
- the result output unit 23 may display a message such as “There is a possibility that the respiratory sound is not attenuated” on the display unit 12.
- FIG. 24 is a diagram illustrating a specific example of the sound type determination result output by the continuity noise determination unit 43 of the sound type determination unit 40 using the waveform feature determination result output by the waveform feature determination unit 30 as an input.
- the continuity noise determination unit 43 of the sound type determination unit 40 determines whether the biological sound included in the biological sound information acquired by the biological sound acquisition unit 20 is classified as “continuous noise”. is there. More specifically, in this embodiment, the continuity noise determination unit 43 indicates that the biological sound indicates “true: there is a possibility that it is continuity noise” or “false: there is a possibility that it is not continuity noise”. The value information is output as the sound type determination result. The output sound type determination result is supplied to the result output unit 23.
- the continuity noise determination unit 43 uses the waveform feature determination results of “determination item 1 ′”, “determination item 2-B”, and “determination item 4” to determine true / false. Obtained from the determination unit 30.
- the continuity noise determination unit 43 acquires the waveform feature determination result of “determination item 1 ′” indicating whether or not the periodicity is weak from the periodicity determination unit 31.
- the continuity noise determination unit 43 acquires the waveform characteristic determination result of “determination item 2-B” representing the abnormality of the frequency component distribution from the spectrum determination unit 32. Further, the continuity noise determination unit 43 acquires the waveform characteristic determination result of “determination item 4” indicating whether or not noise continuity is recognized from the envelope determination unit 34.
- the continuity noise determination unit 43 determines the authenticity of continuity noise for each of the eight patterns.
- the continuity noise determination unit 43 as shown in FIG. 24, “true: there is a possibility of continuity noise” only in the case of pattern (A) where all the determination items are true. judge. If any one of the three determination items is false, it is determined that “false: there is a possibility that the noise is not continuous noise”.
- “not continuous noise” may indicate that the breathing sound is normal or may indicate that there is an abnormality other than continuous noise.
- a body sound (respiratory system sound) with “true” as “judgment item 1 ′” is considered to have “weak periodicity” and a body sound with “true” as “judgement item 2-B”.
- a biological sound whose “judgment item 4” is “true” is considered to have “continuity in noise”.
- the characteristic that noise is continuous is a typical sign of continuous noise. Therefore, in this embodiment, the continuity noise determination unit 43 concludes that the biological sound in which all of these are true is “true: there is a possibility of continuity noise”.
- a body sound with “false” as “judgment item 1 ′” is considered to have “strong periodicity”, and a body sound with “false” as “judgement item 2-B” has an abnormal frequency component distribution.
- a biological sound whose “judgment item 4” is “false” is considered “no continuity in noise”. Therefore, in the present embodiment, the continuity noise determination unit 43 recognizes that the biological sound that is false in any one of these determination items has characteristics different from the signs of continuity noise. It may not be sexual noise ”. Note that the reason why a characteristic different from the signs of continuous noise is recognized is that the breathing sound is normal in the first place or there is an abnormality other than continuous noise.
- the sound type determination result output by the continuity noise determination unit 43 is displayed on the display unit 12 by the result output unit 23.
- the result output unit 23 displays a message such as “There is a possibility of continuity noise” on the display unit 12. Show it.
- the result output unit 23 may display a message such as “There may be no continuity noise” on the display unit 12.
- FIG. 25 is a diagram illustrating a specific example of the sound type determination result output by the intermittent noise determination unit 44 of the sound type determination unit 40 using the waveform feature determination result output by the waveform feature determination unit 30 as an input.
- the intermittent noise determination unit 44 of the sound type determination unit 40 determines whether or not the biological sound included in the biological sound information acquired by the biological sound acquisition unit 20 is classified as “intermittent noise”. is there. More specifically, in the present embodiment, the intermittent noise determination unit 44 indicates that the biological sound indicates “true: may be intermittent noise” or “false: may not be intermittent noise”. The value information is output as the sound type determination result. The output sound type determination result is supplied to the result output unit 23.
- the intermittent noise determination unit 44 uses the waveform feature determination results of “determination item 1 ′”, “determination item 2-B”, and “determination item 5” to determine whether the feature is true or false. Obtained from the determination unit 30.
- the intermittent noise determination unit 44 acquires from the periodicity determination unit 31 the waveform feature determination result of “determination item 1 ′” indicating whether or not the periodicity is weak.
- the intermittent noise determination unit 44 acquires from the spectrum determination unit 32 the waveform characteristic determination result of “determination item 2-B” that represents the abnormality of the frequency component distribution. Further, the intermittent noise determination unit 44 acquires the waveform feature determination result of “determination item 5” indicating whether or not intermittent noise is recognized from the impulse noise determination unit 35.
- the true / false patterns are eight patterns (A) to (K) as shown in FIG.
- the intermittent noise determination unit 44 determines the authenticity of the intermittent noise for each of the eight patterns.
- the intermittent noise determination unit 44 says “true: there is a possibility of intermittent noise” only in the case of pattern (A) in which all the determination items are true. judge. If any one of the three determination items is false, it is determined that “false: there is a possibility of not being intermittent noise”.
- “not intermittent noise” may indicate that the respiratory sound is normal or may indicate that there is an abnormality other than intermittent noise.
- a body sound (respiratory system sound) with “true” as “judgment item 1 ′” is considered to have “weak periodicity” and a body sound with “true” as “judgement item 2-B”.
- a biological sound whose “judgment item 5” is “true” is considered to be “a lot of intermittent noise is recognized”.
- the characteristic that a lot of intermittent noise (impulse noise) is recognized is a typical sign of intermittent noise. Therefore, in the present embodiment, the intermittent noise determination unit 44 concludes that a biological sound in which all of these are true is “true: there is a possibility of being intermittent noise”.
- the intermittent noise determination unit 44 recognizes that the body sound that is false in any one of these determination items has a different characteristic from the sign of intermittent noise. It may not be sexual noise ”. In addition, it is considered that the reason why the characteristic different from the sign of intermittent noise is recognized is that the breathing sound is normal in the first place or there is an abnormality other than intermittent noise.
- the sound type determination result output by the intermittent noise determination unit 44 is displayed on the display unit 12 by the result output unit 23.
- the result output unit 23 displays a message such as “There is a possibility of intermittent noise” on the display unit 12. Show it.
- the result output unit 23 may display a message such as “There may be no intermittent noise” on the display unit 12.
- FIG. 29 illustrates the case where the result output unit 23 displays all the sound type determination results of the respective units of the sound type determination unit 40, but the configuration of the information analysis apparatus 100 of the present invention is described. Is not limited to this. For example, when the normal breathing sound determination unit 41 classifies that the breathing sound is normal, and the result output unit 23 determines that all other abnormal sound determination results are false (not abnormal sounds), the remaining The sound type determination result of each part of the sound type determination unit 40 may be omitted and the analysis result may be displayed.
- a plurality of abnormal sound determination units other than the normal breath sound determination unit 41 determine that the respiratory sound is abnormal. It is assumed that In this case, the result output unit 23 may be the message “XX sound” at the time of each abnormal sound, regardless of whether or not the normal respiratory sound determination unit 41 determines that the respiratory sound is normal.
- a plurality of abnormal messages such as “There may be multiple diseases” may be displayed separately.
- the result output unit 23 displays a message “There is a possibility that the respiratory sound is attenuated.” And a message “There may be multiple diseases” is displayed at the same time.
- each unit of the sound type determination unit 40 counts the frequency at which abnormality of that type is recognized from all the sound waveforms included in the body sound information, and outputs the result to the result output unit 23.
- the structure which outputs may be sufficient.
- the continuity noise determination unit 43 analyzes a biological sound waveform for 40 seconds (about 10 cycles of respiration) and counts how many waveforms matching the determination pattern (a) shown in FIG. 24 are detected. can do.
- the continuity noise determination unit 43 can supply the result output unit 23 together with the sound type determination result, the number of times the continuity noise is detected in the 10 cycles.
- the information analysis apparatus 100 of the present invention does not necessarily include the abnormal level determination unit 50.
- the sound type determination unit 40 classifies the body sound as an abnormal sound type, it is preferable to include an abnormal level determination unit 50 for determining the degree (level) of the abnormality.
- the attenuation level determination unit 51 determines the attenuation level of the sound waveform of the biological sound when the respiratory sound attenuation determination unit 42 determines that the biological sound is “true: the respiratory sound may be attenuated”. To do.
- FIG. 26 is a diagram illustrating an example of an attenuation level determination criterion referred to by the attenuation level determination unit 51 and an example of an attenuation level determination result output from the attenuation level determination unit 51.
- the attenuation level determination unit 51 executes attenuation level determination. Specifically, the attenuation level determination unit 51 reads the attenuation level determination criterion shown in FIG. 26 stored in the storage unit 13. Then, the read reference is applied to the spectrogram of the biological sound output from the time frequency analysis unit 213. Then, the attenuation level of the body sound is determined according to which criterion the sound waveform matches. In this embodiment, as an example, the attenuation level determination unit 51 outputs the attenuation level determination result in three stages of “low”, “medium”, and “high”.
- Low means a body sound with a relatively low attenuation
- High means a body sound with a relatively heavy attenuation
- Medium in the meantime. Means. It can be considered that the degree of attenuation is heavier as the component of the high frequency band is cut in a wider range.
- the content shown in FIG. 26 is an example for explaining the function of the attenuation level determination unit 51, and is not intended to limit the configuration of the attenuation level determination unit 51.
- the threshold value (value between “** _” and “_ **”) defined in the attenuation level determination criterion shown in FIG. 26 is determined by the user (such as operator U) of the information analysis apparatus 100. The configuration may be arbitrarily changed and set. Further, the attenuation level determination unit 51 can output the level determination result with multi-level numerical values in more detail instead of the three values of low, medium, and high, or simply low (mild), High (severe) binary values can also be output.
- the attenuation level determination unit 51 first specifies a boundary frequency between a frequency band in which periodicity is recognized (strong) and a frequency band in which periodicity is not recognized (weak) in the spectrogram. Similarly to the spectrogram determination unit 33, the attenuation level determination unit 51 may scan the spectrogram to detect the boundary. If the spectrogram determination unit 33 has already specified the boundary, the attenuation level determination unit 51 may The frequency value may be acquired from the spectrogram determination unit 33. For example, in the example illustrated in FIG. 13, the attenuation level determination unit 51 specifies the boundary frequency as approximately 330 Hz.
- the attenuation level determination unit 51 reads the attenuation level determination criterion shown in FIG. 26, and determines to which criterion the spectrogram having the above-mentioned boundary matches. According to the example shown in FIG. 13 and FIG. 26, the attenuation level determination unit 51 determines that the boundary (the frequency when the periodicity is weakened from the state where strong periodicity is recognized) is 300 Hz. It is determined that the frequency band is 400 Hz or less.
- the attenuation level determination unit 51 outputs the attenuation level “low” corresponding to the determination result to the result output unit 23 as the attenuation level determination result.
- the attenuation level determination result output by the attenuation level determination unit 51 is displayed on the display unit 12 by the result output unit 23.
- a message such as “• attenuation level: low” may be displayed on the display unit 12 in the area for displaying the level determination result of FIG.
- the continuity level determination unit 52 when the continuity noise determination unit 43 determines that the biological sound is “true: possibility of continuity noise”, the continuity level of the sound waveform of the biological sound Is determined.
- FIG. 27 is a diagram illustrating an example of a continuity level determination criterion referred to by the continuity level determination unit 52 and an example of a continuity level determination result output from the continuity level determination unit 52.
- the continuity level determination unit 52 performs continuity level determination. Specifically, the continuity level determination unit 52 reads the continuity level determination standard shown in FIG. 27 stored in the storage unit 13. Then, the read reference is applied to the envelope of the body sound output from the envelope detector 214. Then, the level of continuity of the biological sound is determined according to which criterion the sound waveform matches. In the present embodiment, as an example, the continuity level determination unit 52 outputs the continuity level determination result in three stages of “low”, “medium”, and “high”.
- Low means a body sound with a relatively low degree of continuity
- High means a body sound with a relatively heavy degree of continuity
- Medium means a body sound between them. It means that there is. It can be considered that the degree of continuity becomes heavier as the waveform having a larger amplitude value continues in the envelope.
- a threshold value (value between “** _” and “_ **”) defined in the continuity level determination criterion shown in FIG. 27 is a user of the information analysis apparatus 100 (operator U or the like). May be arbitrarily changed and settable.
- the continuity level determination unit 52 can output the level determination result with multi-level numerical values in more detail instead of the three values of low, medium, and high, or simply low (mild) High (severe) values can also be output.
- the continuity level determination unit 52 specifies how long a continuous section (time) exceeding the average amplitude value is in the detected envelope. Similar to the envelope determination unit 34, the continuity level determination unit 52 may specify a section Z that exceeds the amplitude average value of the envelope and may specify the length of time of the section Z. Alternatively, when the envelope determination unit 34 has already specified the length of time of the section, it may be acquired from the envelope determination unit 34. For example, in the example shown in FIG. 19A, the time length of the section Z1 is specified as 250 ms. Further, for example, when there are a plurality of sections exceeding the amplitude average value Avr as in the example shown in FIG. 19B, the continuity level determination unit 52 determines the average time of each section 2-4. One length may be specified. Alternatively, one of the longest times in the sections 2 to 4 may be specified.
- the continuity level determination unit 52 reads out the continuity level determination criteria shown in FIG. 27 and determines which criteria the specified length of time matches. According to the example shown in (a) of FIG. 19 and FIG. 27, the continuity level determination unit 52 determines that it corresponds to 200 ms or more and less than 600 ms since the length of the time specified above is 250 ms.
- the continuity level determination unit 52 outputs the continuity level “low” corresponding to the determination result to the result output unit 23 as the continuity level determination result.
- the continuity level determination result output by the continuity level determination unit 52 is displayed on the display unit 12 by the result output unit 23.
- a message such as “.continuity level: low” may be displayed on the display unit 12 in the area for displaying the level determination result.
- the intermittent level determination unit 53 when the intermittent noise determination unit 44 determines that the biological sound is “true: there is a possibility of intermittent noise”, the intermittent level of the sound waveform of the biological sound. Is determined.
- FIG. 28 is a diagram illustrating an example of an intermittent level determination criterion referred to by the intermittent level determination unit 53 and an exemplary intermittent level determination result output by the intermittent level determination unit 53.
- the intermittent level determination unit 53 performs the intermittent level determination. Specifically, the continuity level determination unit 53 reads the continuity level determination standard shown in FIG. 28 stored in the storage unit 13. Then, the read reference is applied to the impulse noise detection result of the biological sound output from the impulse noise detection unit 215. Then, the level of discontinuity of the body sound is determined according to which criterion the sound waveform matches. In the present embodiment, as an example, the continuity level determination unit 53 outputs the continuity level determination result in three stages of “low”, “medium”, and “high”.
- Low means a body sound with a relatively low degree of intermittentness
- High means a body sound with a relatively heavy degree of intermittentness
- Medium means a body sound between them. It means that there is.
- the impulse noise detection result it can be considered that the more the impulse noise is detected, the greater the degree of intermittentness.
- the content shown in FIG. 28 is an example for explaining the function of the intermittent level determining unit 53, and there is no intention to limit the configuration of the intermittent level determining unit 53.
- the threshold value (value between “** _” and “_ **”) defined in the intermittent level determination criterion shown in FIG. 28 is the user of the information analysis apparatus 100 (operator U or the like). May be arbitrarily changed and settable.
- the intermittent level determination unit 53 can output the level determination result with multi-level numerical values in more detail instead of the three values of low, medium and high, or simply low (mild) High (severe) values can also be output.
- the discontinuity level determination unit 53 identifies how many impulse noises are detected per cycle in the impulse noise detection result. Similar to the impulse noise determination unit 35, the intermittent level determination unit 53 may specify the number of impulse noises per cycle from the impulse noise detection result. Alternatively, when the impulse noise determination unit 35 has already specified the number of impulse noises per cycle, it may be acquired from the impulse noise determination unit 35.
- the intermittent level determination unit 53 may specify the number of impulse noises per cycle of the biological sound as 50.
- the continuity level determination unit 53 reads the continuity level determination criterion shown in FIG. 28 and determines which criterion the specified number of impulse noises matches. According to the example shown in FIGS. 20 and 28, the intermittent level determination unit 53 determines that the number of impulse noises identified earlier is 50, and thus corresponds to 30 or more.
- the continuity level determination unit 53 outputs the continuity level “high” corresponding to the determination result to the result output unit 23 as the continuity level determination result.
- the intermittent level determination result output by the intermittent level determination unit 53 is displayed on the display unit 12 by the result output unit 23.
- a message such as “.intermittentity level: high” may be displayed on the display unit 12 in the area for displaying the level determination result in FIG.
- FIG. 30 is a flowchart showing a flow of information analysis processing of the information analysis apparatus 100 in the present embodiment.
- the body sound acquisition unit 20 acquires body sound information to be subjected to information analysis processing from the electronic stethoscope 3 via the communication unit 14 (S1).
- the body sound processing unit 21 processes the sound waveform included in the body sound information acquired by the body sound acquisition unit 20 to generate waveform feature information (S2).
- the body sound processing unit 21 generates the waveform feature information.
- the autocorrelation analysis unit 211 derives an autocorrelation function (waveform feature information) from the sound waveform;
- the Fourier transform unit 212 derives a spectrum (waveform feature information) from the sound waveform;
- the time frequency analysis unit 213 derives a spectrogram (waveform feature information) from the sound waveform;
- the envelope detector 214 detects the envelope of the sound waveform (waveform feature information); and
- the impulse noise detection unit 215 identifies the impulse noise of the sound waveform and outputs an impulse noise detection result (waveform feature information);
- the biological sound processing unit 21 may be configured to generate all the above-described waveform feature information, or may be configured to generate a part thereof.
- the waveform feature determination unit 30 analyzes the waveform feature information generated by the biological sound processing unit 21, determines the characteristics of the sound waveform, and generates a waveform feature determination result reflecting the determination result (S3). ).
- the waveform feature determination unit 30 generates a waveform feature determination result.
- the periodicity determination unit 31 executes “determination item 1” or “determination item 1 ′” to determine a feature related to the periodicity of the body sound;
- the spectrum determining unit 32 executes “determination item 2-A” or “determination item 2-B” to determine a feature relating to the frequency component distribution of the body sound;
- the spectrogram determination unit 33 executes “determination item 3-A” or “determination item 3-B” to determine a feature related to periodicity in the time-frequency component distribution of the body sound,
- the envelope determination unit 34 executes “determination item 4” to determine characteristics relating to the continuity of noise included in the body sound; and
- the impulse noise determination unit 35 executes “determination item 5” to determine a feature related to the intermittentness of noise included in the body sound,
- the waveform feature determination unit 30 may execute all the determination items described above, or may execute some determination items.
- the fact that the waveform characteristic determination result of “determination item 4” executed by the envelope determination unit 34 is “true” indicates that the continuous noise determination unit 43 is “continuous noise” for the target breathing sound. It is a sufficient basis to determine that there is a possibility.
- the envelope determination unit 34 of the waveform feature determination unit 30 executes “determination item 4”, and the continuity noise determination unit 43 of the sound type determination unit 40 receives only the waveform feature determination result of “determination item 4”.
- a configuration in which the presence / absence of the possibility of continuity noise is also included in the category of the present invention.
- the fact that the waveform characteristic determination result of “determination item 5” executed by the impulse noise determination unit 35 is “true” indicates that the intermittent noise determination unit 44 performs “intermittent noise” on the target breathing sound. It is a sufficient basis to determine that it may be
- the impulse noise determination unit 35 of the waveform feature determination unit 30 executes “determination item 5”, and the intermittent noise determination unit 44 of the sound type determination unit 40 receives only the waveform feature determination result of “determination item 5”.
- a configuration in which the presence / absence of the possibility of intermittent noise is also included in the category of the present invention.
- the sound type determination unit 40 determines the sound type of the sound waveform based on the waveform feature determination result generated by the waveform feature determination unit 30, and generates a sound type determination result reflecting the determination result. (S4).
- the sound type determination unit 40 generates a sound type determination result.
- the normal breath sound determination unit 41 determines whether or not the biological sound may be a “normal breath sound”;
- the respiratory sound attenuation determination unit 42 determines whether or not the biological sound may be “respiratory sound attenuation”;
- the continuity noise determining unit 43 determines whether or not the biological sound may be “continuous noise”;
- the intermittent noise determination unit 44 determines whether or not the biological sound may be “intermittent noise”; However, it is not limited to this.
- the sound type determination unit 40 may perform the determination related to all the above-described sound types, or may perform the determination only for some types.
- the body sound analysis unit 22 does not include the abnormal level determination unit 50, or when the sound type determination unit 40 does not classify the sound type of the body sound as an abnormal sound (1 in S5).
- S6 is executed, and the information analysis apparatus 100 ends the series of information analysis processing. That is, the result output unit 23 displays the sound type determination result output from the sound type determination unit 40 on the display unit 12 (S6).
- the result output unit 23 displays the sound type determination result output by each unit of the sound type determination unit 40 in the area for displaying the analysis result, as shown in FIG.
- the sound type determination unit 40 determines that there is a possibility of an abnormal sound (here, respiratory sound attenuation, continuity noise, and intermittent noise), the abnormal sound is converted into the biological sound.
- the structure which counts the frequency which appears may be sufficient. Therefore, the result output unit 23 may further display the determined appearance frequency of the abnormal sound in the area for displaying the analysis result.
- the body sound analysis unit 22 includes the abnormal level determination unit 50 and the sound type determination unit 40 classifies the sound type of the body sound as an abnormal sound (2 in S5), an abnormality occurs.
- the level determination unit 50 performs abnormal level determination.
- the abnormal level determination unit 50 determines the degree of abnormality of the classified sound type and generates an abnormal level determination result (S7).
- the abnormal level determination unit 50 generates an abnormal level determination result.
- the attenuation level determination unit 51 performs an attenuation level determination on the spectrogram to generate an attenuation level determination result;
- the continuity level determination unit 52 performs continuity level determination on the envelope to generate a continuity level determination result;
- the intermittent level determination unit 53 performs an intermittent level determination on the impulse noise detection result to generate an intermittent level determination result;
- the abnormal level determination unit 50 may perform level determination related to all the above-described abnormal sound types, or may perform level determination only for some abnormal sound types.
- the result output unit 23 displays the sound type determination result output from the sound type determination unit 40 and the abnormal level determination result output from the abnormal level determination unit 50 on the display unit 12 (S8). For example, as shown in FIG. 29, for each type of abnormal sound, values such as “low”, “medium”, and “high” indicating the level of abnormality are displayed in an area for displaying an abnormality level determination result.
- the normal breathing sound determination unit 41 outputs “determination item 1”, “determination item 2-A”, and “determination item 3-
- the configuration for determining whether or not the breathing sound is normal based on the waveform feature determination result of “A” has been described.
- the configuration of the normal breathing sound determination unit 41 of the present invention is not limited to this.
- each of the respiratory sound attenuation determination unit 42, the continuous noise determination unit 43, and the intermittent noise determination unit 44 of the sound type determination unit 40 Determining if there is a possibility of “attenuation of breathing sound”; Determining if it may be “continuous noise”; and Determining whether there may be “intermittent noise”;
- the normal breathing sound determination unit 41 may determine that the breathing sound is normal (possibly).
- Embodiment 2 Another embodiment relating to the information analysis apparatus of the present invention will be described below with reference to FIGS. For convenience of explanation, members having the same functions as those in the drawings described in the first embodiment are given the same reference numerals, and descriptions thereof are omitted.
- the sound type determination unit 40 is configured to include a determination unit for determining whether or not the sound type to be classified is the sound type.
- the configuration of the information analysis apparatus 100 of the present invention is not limited to the above.
- the sound type determination unit 40 performs a comprehensive determination based on all characteristics of the body sound so that the body sound is finally classified into one sound type instead of including a determination unit for each sound type.
- the comprehensive determination unit 45 may be provided.
- FIG. 31 is a functional block diagram showing a main configuration of the information analysis apparatus 100 according to this embodiment.
- the sound type determination unit 40 includes a normal respiratory sound determination unit 41, a respiratory sound attenuation determination unit 42, and a continuous noise determination unit 43.
- a comprehensive determination unit 45 is provided instead of not including the intermittent noise determination unit 44.
- the comprehensive judgment unit 45 comprehensively uses the waveform feature judgment results output by each part of the waveform feature judgment unit 30 and identifies the sound type of the target biological sound.
- Each functional block of the control unit 10 described above, in particular, the overall determination unit 45 is a storage device (CPU (central processing unit) or the like realized by ROM (read only memory), NVRAM (non-Volatile random access memory), etc. This can be realized by reading a program stored in the storage unit 13) into a RAM (random access memory) or the like and executing it.
- CPU central processing unit
- ROM read only memory
- NVRAM non-Volatile random access memory
- FIG. 32 is a diagram illustrating a classification system used when the comprehensive determination unit 45 in the present embodiment classifies respiratory system sounds acquired from the patient P into predetermined sound types.
- the comprehensive determination unit 45 selects “respiratory system sound” as “normal respiratory sound”, “reducing respiratory sound”, “other abnormal sounds”, “high-pitched continuity noise”. ",” Low-pitched continuous noise “,” fine intermittent noise “,” rough intermittent noise “, and” other noise ". Then, the specified sound type is output to the result output unit 23 as a comprehensive determination result.
- the comprehensive judgment unit 45 first classifies “respiratory system sounds” collected from the patient P into “respiratory sounds” and “noises”. The overall determination unit 45 performs this classification based on the waveform feature determination results of “determination item 1-1” and “determination item 1-2” of FIG. 7 output by the periodicity determination unit 31.
- the comprehensive judgment unit 45 classifies the “breathing sound” into “breathing sound (normal or attenuated)” and “other abnormal sounds”.
- the overall determination unit 45 performs this classification based on the waveform feature determination result of “determination item 2-A” of FIG. 11 output by the spectrum determination unit 32.
- the comprehensive judgment unit 45 classifies the “breathing sound (normal or attenuated)” into “normal breathing sound” and “breathing sound attenuation”.
- the overall determination unit 45 performs this classification based on the waveform feature determination result of “determination item 3-A” of FIG. 16 output by the spectrogram determination unit 33.
- the comprehensive judgment unit 45 classifies “noise” into “continuous noise” and “noise other than continuous noise”.
- the overall determination unit 45 performs this classification based on the waveform feature determination result of “determination item 4” in FIG. 18 output by the envelope determination unit 34.
- the comprehensive judgment unit 45 classifies “continuous noise” into “high-pitched continuous noise” and “low-pitched continuous noise”.
- the overall determination unit 45 performs this classification based on the waveform feature determination result of “determination item 2-B” of FIG. 11 output by the spectrum determination unit 32.
- the comprehensive judgment unit 45 classifies “noise other than continuous noise” into “intermittent noise” and “other noises”.
- the comprehensive determination unit 45 performs this classification based on the waveform feature determination result of “determination item 5” in FIG. 21 output by the impulse noise determination unit 35.
- the comprehensive judgment unit 45 classifies “intermittent noise” into “fine intermittent noise” and “rough intermittent noise”.
- the overall determination unit 45 performs this classification based on the waveform feature determination result of “determination item 2-B” of FIG. 11 output by the spectrum determination unit 32.
- FIG. 33A and 33B are flowcharts showing a flow of information analysis processing of the information analysis apparatus 100 in the present embodiment. In this embodiment, it is assumed that S1 and S2 shown in FIG. 30 have been executed prior to S101 in FIG. 33A.
- the periodicity determination unit 31 executes “determination item 1-1” (S101). That is, it is determined whether or not the waveform of the autocorrelation function has a peak at a period of 2 to 5 seconds. Further, the periodicity determination unit 31 executes “determination item 1-2” (S102). That is, it is determined whether or not the width (period) of the peak at 1 ⁇ 4 of the peak amplitude value of the envelope in the autocorrelation function is 10% or less of the respiratory cycle. Note that the periodicity determination unit 31 may execute either S101 or S102 first.
- the comprehensive judgment unit 45 converts the body sound into “breathing sound (no noise) ) ”(S104).
- the comprehensive determination unit 45 determines that the biological determination unit 45 The sound is classified as “noise (present)” (S105).
- the spectrum determination unit 32 executes “determination item 2-A” on the biological sound classified as “breathing sound” (S106). That is, it is determined whether the sum of frequency components of 200 Hz or less occupies 80% or more of the whole.
- the spectrogram determination unit 33 executes “determination item 3-A” on the biological sound classified as “either normal breathing sound or respiratory sound attenuation” (S110). That is, it is determined whether the periodicity of the frequency component is recognized at 400 Hz (or higher).
- the comprehensive judgment unit 45 converts the body sound into the “normal breathing sound”. (S112).
- the comprehensive determination unit 45 converts the biological sound into “ It classify
- the body sound analysis unit 22 includes the attenuation level determination unit 51, the attenuation level determination unit 51 determines the attenuation level of the body sound (S114).
- the envelope determination unit 34 determines whether the biological sound is classified as “noise”. “Decision item 4” is executed (S115). That is, it is determined whether or not continuity is recognized in the envelope (noise).
- the comprehensive determination unit 45 classifies the body sound as “continuity noise” (S117). .
- the spectrum determination unit 32 executes “determination item 2-B” on the biological sound classified as “continuous noise” (S118). That is, it is determined whether the sum of frequency components of 200 Hz or more occupies 30% or more of the whole.
- the comprehensive determination unit 45 converts the body sound into “continuity noise”. Is classified as “noise other than” (S123).
- the impulse noise determination unit 35 executes “determination item 5” on the biological sound classified as “noise other than continuous noise” (S124). That is, it is determined whether there are 10 or more impulse noises in one cycle.
- the overall determination unit 45 classifies the body sound as “intermittent noise” (S126). .
- the comprehensive determination unit 45 classifies the body sound as “other noise”. (S127).
- the spectrum determination unit 32 executes “determination item 2-B” on the biological sound classified as “intermittent noise” (S128). That is, it is determined whether the sum of frequency components of 200 Hz or more occupies 30% or more of the whole.
- the result output unit 23 displays on the display unit 12 the comprehensive determination result output by the comprehensive determination unit 45, in which the biological sound is classified into one of the above sound types (S133). . Further, when the abnormal level determination unit 50 outputs the abnormal level determination result, the abnormal level determination result is also displayed on the display unit 12.
- FIG. 34 is a flowchart showing the flow of attenuation level determination processing executed by the attenuation level determination unit 51.
- the attenuation level determination unit 51 When the attenuation level determination process is started in S7 of FIG. 30 or S114 of FIG. 33A, the attenuation level determination unit 51 first scans the spectrogram of the biological sound to be processed and has a strong periodicity (The frequency at the boundary between the recognized frequency band and the frequency band having a weak periodicity (not recognized) is specified (S201). For example, the attenuation level determination unit 51 refers to the attenuation level determination standard illustrated in FIG. 26 from the storage unit 13.
- the attenuation level determination unit 51 determines the attenuation level to be “low” (S203).
- the attenuation level determination unit 51 determines whether or not the boundary frequency corresponds to 200 Hz or more and less than 300 Hz. Determine (S204). If the boundary frequency falls within the range of 200 Hz to less than 300 Hz (YES in S204), the attenuation level determination unit 51 determines that the attenuation level is “medium” (S205).
- the boundary frequency does not correspond to 200 Hz or more and less than 300 Hz (NO in S204)
- the boundary frequency corresponds to less than 200 Hz.
- the attenuation level determination unit 51 determines that the attenuation level is “high” (S206).
- the attenuation level determination result output by the attenuation level determination unit 51 is output to the result output unit 23.
- FIG. 35 is a flowchart showing the flow of continuity level determination processing executed by the continuity level determination unit 52.
- the continuity level determination unit 52 calculates the amplitude average value in the envelope of the biological sound wave waveform to be processed.
- the continuous continuous time is specified (S301).
- the continuity level determination unit 52 refers to the continuity level determination standard illustrated in FIG. 27 from the storage unit 13.
- the continuity level determination unit 52 determines that the continuity level is “low” (S303).
- continuity level determination unit 52 determines whether or not the continuous time corresponds to 600 ms or more and less than 1000 ms. Is determined (S304). If the continuous time corresponds to 600 ms or more and less than 1000 ms (YES in S304), the continuity level determination unit 52 determines that the continuity level is “medium” (S305).
- the continuity level determination unit 52 determines that the continuity level is “high” (S306).
- the continuity level determination result output by the continuity level determination unit 52 is output to the result output unit 23.
- the configuration for obtaining the continuity level using the continuous time exceeding the amplitude average value as an index has been described.
- the configuration of the continuity level determination unit 52 is not limited to this.
- the continuity level determination unit 52 may be configured to obtain the continuity level using the total time exceeding the amplitude average value of the envelope per cycle as an index.
- FIG. 36 is a flowchart showing the flow of the continuity level determination process executed by the continuity level determination unit 53.
- the discontinuity level determination unit 53 first determines the number of impulse noises per cycle in the biological sound waveform to be processed. Is identified (S401).
- the continuity level determination unit 53 refers to the continuity level determination standard illustrated in FIG. 28 from the storage unit 13.
- the continuity level determination unit 53 determines the continuity level to be “low” (S403).
- the intermittent level determination unit 53 further applies the number of impulse noises to 20 or more and less than 30. It is determined whether or not to perform (S404). If the number of impulse noises falls between 20 and less than 30 (YES in S404), the continuity level determination unit 53 determines that the continuity level is “medium” (S405).
- the continuity level determination unit 53 determines that the continuity level is “high” (S406).
- the intermittent level determination result output by the intermittent level determination unit 53 is output to the result output unit 23.
- the result output unit 23 displays, on the display unit 12, the comprehensive determination result that is output from the comprehensive determination unit 45 and classifies the biological sound into any one of the above sound types.
- the comprehensive determination result is displayed in the area where the analysis result is displayed.
- FIG. 37 shows an example of the comprehensive determination result when the comprehensive determination unit 45 classifies the body sound as “high-pitched continuity noise”.
- the result output unit 23 displays the appearance frequency acquired from the comprehensive determination unit 45 on the display unit 12 together. May be.
- the result output unit 23 may display the abnormal level determination result together on the display unit 12.
- the biological sound is classified as “high-pitched continuity noise”. Accordingly, the result output unit 23 displays the continuity level determination result determined by the continuity level determination unit 52 in an area for displaying the level determination result.
- the result output unit 23 may receive an instruction from the operator U to reproduce the biological sound subjected to the analysis processing by displaying a “reproduce sound” button as shown in FIGS. 29 and 37.
- the result output unit 23 reproduces the biological sound information acquired by the biological sound acquisition unit 20 and sends the audio signal to an audio output unit (not shown). It may be output. Further, for example, when the “play sound” button is double-tapped, the result output unit 23 controls the sound output unit so that the sound is played back from the place where the abnormality appears in the body sound. May be.
- the result output unit 23 associates the biological sound information, the determination results, and necessary patient information with each other, 13.
- the result output unit 23 may store the biological sound information associated with the determination result in a database (not shown) of the external device. Specifically, the result output unit 23 may transmit various determination results received from the body sound analysis unit 22 to an external apparatus together with the collected body sound information via the communication unit 14. For example, the communication unit 14 of the information analysis apparatus 100 can transmit each determination result and biological sound information to the management server 4 via the communication network 5.
- the management server 4 can display the determination result shown in FIG. 29 or FIG. 37 on the display unit of its own device, and present the determination result of the body sound of the patient P to the doctor D in the remote place. It becomes possible.
- the management server 4 can reproduce the body sound information desired by the doctor D according to the operation of the doctor D, and let the doctor D hear it.
- the body sound processing unit 21 processes the body sound information to extract the waveform feature information from the sound waveform, and the waveform feature determination unit 30 determines what criterion the waveform feature information is based on. It is determined whether they match (or do not match).
- the comprehensive determination unit 45 can specify the type of the biological sound according to the waveform feature determination result. Specifically, the biological sound can be classified into one type having a high possibility from a plurality of types defined in advance based on the medical characteristics of the sound.
- the comprehensive determination result made by the comprehensive determination unit 45 is displayed on the display unit 12 as an analysis result.
- the comprehensive determination unit 45 determines which sound type the original biological sound information has a high correlation (or low) depending on whether or not the extracted waveform feature information matches the above-described determination criteria. Can do.
- the function of analyzing information such as respiratory sounds is realized by the information analysis apparatus 100 as a terminal apparatus operated by the operator U in the auscultation system 200 of the present invention.
- the information analysis apparatus 100 is configured to communicate with the electronic stethoscope 3 and the management server 4 of the support center 2 in the auscultation system 200.
- the configuration of the auscultation system 200 of the present invention is not limited to this.
- the function of analyzing information such as respiratory sounds performed by the information analysis apparatus 100 of the present invention may be installed in the electronic stethoscope 3 and / or the management server 4 of the support center 2.
- the electronic stethoscope 3 and / or the management server 4 function as the information analysis apparatus of the present invention.
- Embodiment 3 Another embodiment of the present invention will be described below with reference to FIG.
- members having the same functions as those in the drawings described in the first and second embodiments are given the same reference numerals, and descriptions thereof are omitted.
- Patent Literature 3 when creating a medical image, after acquiring image data in which a predetermined part of a living body is imaged, body sound measurement is performed on the imaged part, and the results are associated with each other. A medical image display system for displaying medical images is described.
- a measurement system for performing medical image photographing will be described in consideration of the biological sound measurement result obtained by performing the biological sound measurement.
- FIG. 38 is a block diagram illustrating an outline of a measurement system 3600 according to the third embodiment and a configuration of main parts of an image capturing device 3006 constituting the measurement system 3600.
- the measurement system 3600 includes at least an electronic stethoscope 3 and an image capturing device 3006. Furthermore, the measurement system 3600 may include the above-described auscultation system 200 (FIG. 2) as necessary. That is, the electronic stethoscope 3 and the image capturing device 3006 in the third embodiment are connected to the various devices in the auscultation system 200 in the above-described first and second embodiments as needed, and operate in cooperation with each other. It is possible.
- the electronic stethoscope 3 collects the body sound information of the patient P.
- the electronic stethoscope 3 is the electronic stethoscope 3 that functions as a part of the auscultation system 200 shown in FIG.
- the image imaging device 3006 captures the patient P using an appropriate imaging unit and acquires image data.
- the image data acquired by the image capturing device 3006 is used as a medical image by the operator U or the doctor D.
- the image capturing apparatus 3006 is linked to the auscultation system 200 shown in FIG.
- the image capturing apparatus 3006 can select and execute an image capturing process optimal for the patient P in consideration of the auscultation result of the patient P derived by the auscultation system 200.
- the image capturing device 3006 includes a communication unit 3011 that transmits and receives information to and from each device of the auscultation system 200, a storage unit 3012 that stores various types of information processed by the image capturing device 3006, and a patient.
- the image capturing unit 3013 that performs the above shooting and the control unit 3010 that controls each unit of the image capturing apparatus 3006 are configured.
- the communication unit 3011 communicates with each device of the auscultation system 200 and receives the auscultation result of the patient P derived by the auscultation system 200.
- the storage unit 3012 stores, for example, image data captured by the imaging unit 3013, and stores analysis result information d1 and part information d2 acquired by the communication unit 3011.
- the imaging unit 3013 is not limited to this.
- the imaging unit 3013 is a living body using appropriate means such as X-ray, CT (Computed Tomography), MRI (Magnetic Resonance Imaging), magnetic measurement, bioelectric signal, ultrasound, or light.
- Image The imaging unit 3013 may include a positioning mechanism that arranges the image sensor unit at an appropriate site in order to image a desired site of the patient P.
- the control unit 3010 includes an auscultation result acquisition unit 3020, an imaging site identification unit 3021, and an imaging control unit 3022 as functional blocks.
- the auscultation result acquisition unit 3020 controls the communication unit 3011 to acquire the auscultation result from the information analysis apparatus 100.
- the auscultation result acquired by the auscultation result acquisition unit 3020 includes at least two pieces of information.
- the first is analysis result information d1 indicating the analysis result of the body sound information collected by the electronic stethoscope 3.
- the second part is part information d2 indicating the part from which the body sound information is collected.
- the auscultation result including at least the presence / absence of abnormality determined by the information analysis apparatus 100 based on the body sound information of the patient P and the part from which the body sound information is collected is acquired.
- the image capturing apparatus 3006 is connected to the information analysis apparatus 100 in the first or second embodiment so as to be communicable.
- the auscultation result acquisition unit 3020 receives the sound type determination result determined by the sound type determination unit 40 from the information analysis apparatus 100 of the first embodiment and, depending on circumstances, the abnormal level determination unit 50.
- the determined level determination result is acquired as analysis result information d1.
- the auscultation result acquisition unit 3020 obtains the comprehensive determination result determined by the comprehensive determination unit 45 from the information analysis apparatus 100 according to the second embodiment, and further the level determination result determined by the abnormal level determination unit 50 in some cases. And received as analysis result information d1.
- the information analysis apparatus 100 may accept the input of the part information immediately before the operator U collects the body sound information from the patient P using the electronic stethoscope 3.
- the operator U may roughly input “lung”, or more specifically “right lung” or “left lung”, or more specifically, “upper right lobe”, “Right middle leaf”, “Right lower leaf”, “Left upper leaf”, “Left lower leaf” or the like may be input.
- the lung is defined by being divided into several parts as shown in FIG. 39, and a “shallow branch point” indicating a shallow part A where the branch of the trachea (airway) is not advanced, or Further, a place where the airway is relatively thin (the diameter of the trachea is small), that is, a “deep branch point” indicating a deep part B where the branch of the trachea (airway) is advanced may be input.
- the part designated by the operator U is stored in the information analysis apparatus 100 in association with the body sound information.
- a part linking unit (not shown) of the information analysis apparatus 100 links the part information input from the input unit 11 to the analysis result output from the biological sound analysis unit 22.
- the result output unit 23 of the information analysis apparatus 100 transmits the part associated with the body sound information as the part information d2 to the image capturing apparatus 3006 together with the analysis result information d1 of the body sound information.
- the auscultation result acquisition unit 3020 acquires the auscultation result transmitted as described above, that is, the analysis result information d1 and the part information d2.
- the auscultation result acquired by the auscultation result acquisition unit 3020 is used by the imaging region specifying unit 3021 to specify the imaging region.
- the imaging part specifying unit 3021 specifies a part of a living body to be imaged by the imaging unit 3013.
- the imaging part specifying unit 3021 specifies the position where the body sound information, which is suggested to be abnormal or possibly abnormal by the analysis result information d1, is taken as a part to be imaged.
- the imaging part specifying unit 3021 can specify the part to be imaged by the part information d2 acquired together with the analysis result information d1.
- the analysis result information d1 acquired from the information analysis apparatus 100 according to the first embodiment may be “not a normal breathing sound”, “may be a respiratory sound attenuation”, or “continuous noise”. It is assumed that it includes at least one sound type determination result of “There is a possibility” and “It may be intermittent noise”.
- the imaging part specifying unit 3021 specifies a part to be imaged with reference to the part information d2 acquired together with the analysis result information d1. For example, when the part information d2 is “left lower lobe”, the imaging part specifying unit 3021 specifies an imaging target part as “left lower lobe” because there is an abnormality sign in “left lower lobe”.
- the analysis result information d1 acquired from the information analysis apparatus 100 according to the second embodiment includes a comprehensive determination result regarding some abnormal sound other than “highly likely to be a normal breathing sound”. Also in this case, the imaging part specifying unit 3021 specifies a part to be imaged with reference to the part information d2 acquired together with the analysis result information d1.
- the imaging part specifying unit 3021 may function not only to select whether or not imaging is necessary, but also to narrow down parts that should be subjected to precise imaging with higher resolution.
- the imaging region specifying unit 3021 may determine that only the “left lower lobe” in which an abnormality sign is recognized is imaged with a setting different from the imaging method of other regions (for example, with high resolution).
- the imaging control unit 3022 performs various settings for the imaging unit 3013 based on the site specified by the imaging site specifying unit 3021 and then controls the imaging unit 3013 to image the living body. In other words, the imaging control unit 3022 executes image imaging processing with different settings (in the manner of imaging) for the site specified by the imaging site specifying unit 3021 and the other site.
- imaging is performed by controlling the positioning mechanism of the imaging unit 3013 so that the lower left lobe of the patient P is appropriately captured.
- the imaging control unit 3022 may set the imaging unit 3013 so as to obtain a high resolution only when imaging the lower left lobe, and perform a series of imaging including other parts.
- Image data acquired by the imaging unit 3013 according to the control of the imaging control unit 3022 is stored in the storage unit 3012.
- the imaging control unit 3022 preferably stores the acquired image data in association with the corresponding analysis result information d1 and part information d2.
- the imaging control unit 3022 includes, in the image data obtained by the imaging unit 3013 imaging the lower left leaf, analysis result information d1 indicating “possibility of continuity noise” and a part indicating “lower left leaf”.
- the information d2 is associated and stored in the storage unit 3012.
- the analysis result information d1 includes disease name information as necessary. It doesn't matter.
- the image capturing control unit 3022 can associate the suspected disease name with the acquired image data and store the image data in the storage unit 3012. . If such image data is displayed on a display unit (not shown) together with the disease name and sound type determination result, more detailed information can be provided to the doctor D.
- the imaging region specifying unit 3021 selects the imaging region according to the abnormality level. This is because it can be specified in more detail. Specifically, the imaging region specifying unit 3021 can specify the imaging target area size according to the level of abnormality.
- abnormal part information d2 as an auscultation result
- analysis result information d1 including an abnormal level to the image capturing apparatus 3006.
- specification part 3021 of the image imaging device 3006 designates an imaging object area size according to the level of abnormality.
- the imaging control unit 3022 controls the imaging unit 3013 according to the size designated by the imaging site specifying unit 3021 and acquires a medical image of an appropriate size for an appropriate site.
- the imaging control unit 3022 allows the imaging unit 3013 to be measured based on the imaging target region specified by the imaging region specifying unit 3021 and the determined imaging target area size.
- the medical image can be taken by arranging it at an appropriate place.
- the acquired image data is associated with the part information d2 and the analysis result information d1 (the content of the abnormality, the level of the abnormality) and is stored in the storage unit 3012, and is used as a medical image for diagnosis by the doctor D. Be utilized. Furthermore, since the above-mentioned attached information associated with the image data is managed, it can be used as reference information when the same subject needs to be photographed again after the first diagnosis. Is possible. Thereby, it is possible to improve the measurement accuracy in the subsequent image capturing process. For example, there is a case where the medical image measured for the first time does not have information expected by the doctor D (eg, the resolution is low, the imaging area is narrow, or the abnormal part is photographed with a shift). In that case, the imaging part specifying unit 3021 changes the measurement part, the resolution, and the imaging target area size specified last time, and corrects the image data including the information requested by the doctor D to be captured. Is the method.
- the medical image measured for the first time does not have information expected by the doctor D (e
- the image capturing apparatus 3006 takes the auscultation result output by the auscultation system 200 into consideration, and minimizes the part to be subjected to the image capturing process on the patient P. It is possible to narrow down. That is, the image capturing apparatus 3006 and the image capturing method that are appropriate to the extent that the doctor D can diagnose and that can minimize the burden on the patient P are realized. Specifically, the imaging region specifying unit 3021 determines to capture only the region where abnormality (or the possibility) is recognized based on the auscultation result, or only the region with high resolution. Or decide to shoot. For example, when the imaging unit 3013 is a mechanism that performs imaging with X-rays, the radiation dose received by the patient P can be reduced.
- the information analysis apparatus 100 of the present invention can be realized by various information processing apparatuses.
- the information analysis apparatus 100 of the present invention can be applied to personal computers (PCs), AV devices such as digital televisions, notebook computers, tablet PCs, mobile phones, PDAs (Personal Digital Assistants), and the like. It is.
- the information analysis apparatus 100 may be mounted on the electronic stethoscope 3.
- each block of the information analysis apparatus 100 in particular, the body sound acquisition unit 20, the body sound processing unit 21, the body sound analysis unit 22 and the result output unit 23, and each block of the body sound processing unit 21 and body sound analysis
- Each block of the unit 22 may be configured by hardware logic, or may be realized by software using a CPU as follows.
- each block of the image capturing apparatus 3006, in particular, the auscultation result acquisition unit 3020, the imaging site specifying unit 3021, and the imaging control unit 3022 may be configured by hardware logic, and uses a CPU as follows. It may be realized by software.
- the information analysis apparatus 100 and the image capturing apparatus 3006 are configured such that a CPU (central processing unit) that executes instructions of a control program that realizes each function, a ROM (read memory only) that stores the program, and a RAM that expands the program (Random access memory), a storage device (recording medium) such as a memory for storing the program and various data.
- An object of the present invention is to read the program codes (execution format program, intermediate code program, source program) of the control programs of the information analysis apparatus 100 and the image capturing apparatus 3006, which are software for realizing the functions described above, by a computer.
- the recording medium recorded as possible is supplied to each of the information analysis apparatus 100 and the image capturing apparatus 3006, and the computer (or CPU or MPU) reads and executes the program code recorded on the recording medium. Achievable.
- Examples of the recording medium include tapes such as magnetic tapes and cassette tapes, magnetic disks such as floppy (registered trademark) disks / hard disks, and disks including optical disks such as CD-ROM / MO / MD / DVD / CD-R.
- Card system such as IC card, IC card (including memory card) / optical card, or semiconductor memory system such as mask ROM / EPROM / EEPROM (registered trademark) / flash ROM.
- each of the information analysis apparatus 100 and the image capturing apparatus 3006 may be configured to be connectable to a communication network, and the program code may be supplied via the communication network.
- the communication network is not particularly limited.
- the Internet intranet, extranet, LAN, ISDN, VAN, CATV communication network, virtual private network, telephone line network, mobile communication network, satellite communication. A net or the like is available.
- the transmission medium constituting the communication network is not particularly limited.
- infrared rays such as IrDA and remote control, Bluetooth ( (Registered Trademark), 802.11 wireless, HDR (High Data Rate), mobile phone network, satellite line, terrestrial digital network, and the like can also be used.
- the present invention can also be realized in the form of a computer data signal embedded in a carrier wave in which the program code is embodied by electronic transmission.
- the information analysis apparatus is a waveform feature that indicates a reference for distinguishing a feature of a sound waveform from a sound waveform included in biological sound information collected by a stethoscope. Applying the determination criteria, a waveform feature determining means for specifying the characteristics of the sound waveform, and a type of sound to which the biological sound information belongs based on the characteristics of the sound waveform specified by the waveform feature determining means. And a sound type determining means for determining.
- the waveform feature determination means can apply the waveform feature determination standard to the sound waveform included in the body sound information and specify the feature of the sound waveform. Since the waveform feature determination criterion has a criterion for classifying the characteristics of the sound waveform, the waveform feature determination means can be used for any sound waveform by following the waveform feature determination criterion. It is always possible to implement an objective feature classification.
- the sound type determination unit can determine the type of sound included in the biological sound information according to the determination result by the waveform feature determination unit, that is, the identified feature classification.
- the sound type determination means can accurately determine to which sound type the original body sound information has a high correlation according to an objective classification according to the waveform feature determination criteria.
- the waveform feature determination reference referred to by the waveform feature determination further includes a threshold value to be compared with a feature amount derived from the sound waveform and a condition defined by the threshold value, and the waveform feature It is preferable that the determination unit specifies the feature of the sound waveform by determining whether or not the feature value of the sound waveform satisfies the condition.
- a threshold value (quantitative value) is defined in advance based on features deeply related to the type of sound.
- the waveform feature determination unit can determine whether or not the feature amount extracted from the sound waveform matches the condition defined by the threshold value, and can supply the determination result to the sound type determination unit.
- the sound type determination means can accurately determine to which type the biological sound information including the sound waveform is highly correlated based on the determination result.
- the sound type determined by the sound type determining means is “normal breathing sound” indicating that the breathing sound emitted by the living body is normal, and until the breathing sound emitted by the living body is collected by the stethoscope. Included are “attenuation of respiratory sound” indicating that the sound is attenuated, “continuous noise” indicating that the breathing sound emitted by the living body contains continuous noise, and noise that is intermittent to the breathing sound emitted by the living body It may be at least one of “intermittent noise” indicating that the
- the information analysis apparatus can clarify to the user whether the body sound information (breathing sound) collected by the stethoscope belongs to the “normal breathing sound”.
- the information analysis device can clarify to the user whether or not the body sound information (breathing sound) collected by the stethoscope belongs to “attenuation of breathing sound”.
- the information analysis apparatus can clarify to the user whether or not the body sound information (breathing sound) collected by the stethoscope belongs to “continuous noise”.
- the information analysis apparatus can clarify to the user whether or not the body sound information (breathing sound) collected by the stethoscope belongs to “intermittent noise”.
- the waveform feature determination means determines whether the envelope of the sound waveform continues for a certain period or more with a certain amplitude value or more according to the waveform feature determination standard related to the envelope, and the sound type determination means When it is determined that the envelope is continuous, it can be determined that the biological sound information may belong to “continuity noise”.
- the sound type determining means based on the characteristics relating to the continuity of the envelope of the sound waveform, when the envelope is continuous for a certain period or more with a certain amplitude value or more, a biological body having such an envelope Sound information can be classified as “continuous noise”.
- the information analysis apparatus further determines whether or not it corresponds to “continuous noise” for the body sound information that is determined to have a low periodicity (a type that includes noise) based on the envelope. Can be made clear to the user.
- the waveform feature determining means determines whether or not the sound waveform includes a predetermined number or more of impulse noises according to a waveform feature determination standard related to the number of impulse noises, and the sound type determining means is configured to determine whether the sound waveform is When it is determined that the impulse noise includes a certain number or more, it can be determined that the biological sound information may belong to “intermittent noise”.
- the sound type determination means can classify biological sound information including a certain number of impulse noises as “intermittent noise” based on characteristics relating to the number of impulsive sounds (number of impulse noises). .
- the information analysis apparatus further determines whether or not it corresponds to “intermittent noise” for the body sound information determined to have a low periodicity (a type including noise) based on the number of impulse noises. Can be made clear to the user.
- the waveform feature determination means may determine that the envelope is continuous when the continuous time exceeding the average amplitude value of the envelope is 200 ms or more according to the waveform feature determination criterion.
- the waveform feature determination unit may determine that the envelope is continuous when the total time exceeding the amplitude average value in the envelope within a predetermined period is 200 ms or more according to the waveform feature determination criterion. .
- the waveform feature determining means determines that the sound waveform includes a certain number or more of impulse noises when the sound waveform includes 10 or more impulse noises per period according to the waveform feature determination criterion. May be.
- the waveform feature determination means determines whether the frequency component distribution of the sound waveform shows a normal tendency or an abnormal tendency according to a waveform feature determination criterion related to the frequency component distribution.
- the sound type determining means determines that the body sound information is at least one of “normal respiratory sound” and “respiratory sound attenuation” when it is determined that the frequency component distribution of the sound waveform shows a normal tendency.
- the biological sound information includes “continuous noise” and “intermittent noise”. It can be determined that there is a possibility of belonging to at least one of them.
- the sound type determination means determines the type of the body sound information based on the characteristics related to the frequency component distribution of the sound wave, first, the type not including noise (“normal respiratory sound”, “respiratory sound attenuation”, etc. ).
- the sound type determination means determines the type of the body sound information based on the characteristics relating to the frequency component distribution of the sound wave, and first, the type including noise (such as “continuous noise” and “intermittent noise”). Can be classified.
- the waveform feature determination means when the sum of the frequency components of 200 Hz or less occupies 80% or more in the frequency component distribution of the sound waveform, When the sum of the frequency components of 200 Hz or more occupies 30% or more of the entire frequency component distribution of the sound waveform, it is determined that the frequency component distribution exhibits an abnormal tendency. May be.
- the waveform feature determination means determines whether or not the sound waveform has a strong periodicity according to a waveform feature determination criterion for distinguishing whether or not the sound waveform has a strong periodicity, and determines the sound type determination.
- the means determines that the biological sound information may belong to at least one of “normal respiratory sound” and “respiratory sound attenuation”.
- the periodicity of the sound waveform is weak, it can be determined that the biological sound information may belong to at least one of “continuous noise” and “intermittent noise”.
- the sound type determination means sets the type of the body sound information based on the characteristics related to the periodicity of the sound waveform, first, the type not including noise (such as “normal respiratory sound” and “respiratory sound attenuation”). And types including noise (such as “continuous noise” and “intermittent noise”).
- the waveform feature determination means determines the presence or absence of periodicity for each frequency band of the sound wave according to a waveform feature determination standard related to a frequency component distribution based on time-frequency analysis
- the sound type determination means includes: In the frequency component distribution based on the time frequency analysis, when it is determined that there is periodicity in a high frequency band, it is determined that the biological sound information may belong to a “normal breathing sound”, and the time frequency analysis is performed. In the frequency component distribution based on the above, when it is determined that there is periodicity in the low frequency band and no periodicity in the high frequency band, the biological sound information may belong to “respiratory sound attenuation” Can be determined.
- the sound type determination means for the biological sound information determined to have a strong periodicity (type that does not include noise) based on the characteristics relating to the frequency component distribution based on the time-frequency analysis of the sound waveform, It can be classified as a “normal breathing sound”.
- the sound type determination means this time for the body sound information determined to have a strong periodicity (type that does not include noise) based on the characteristics relating to the frequency component distribution based on the time-frequency analysis of the sound waveform. , It can be classified as “attenuation of respiratory sounds”.
- the information analysis apparatus further selects “normal breathing sound” for biological sound information classified as having the same strong periodicity (type not including noise). It can be clarified to the user whether it falls under "Respiration sound attenuation”.
- the waveform feature determining means is configured to determine a predetermined amplitude value in an envelope of the autocorrelation function when the autocorrelation function of the sound waveform has a peak at intervals of 2 to 5 seconds according to the waveform feature determination criterion.
- the envelope peak period is 10% or less of the respiratory cycle, it may be determined that the periodicity of the sound waveform is strong.
- the waveform feature determination means is configured to perform periodicity in a high frequency band when periodicity is recognized in a band of 400 Hz or higher in the frequency component distribution based on the time-frequency analysis of the sound waveform according to the waveform feature determination criterion.
- periodicity when periodicity is recognized in a band of less than 400 Hz, there is periodicity in the low frequency band, and in the high frequency band. It may be determined that there is no periodicity.
- the sound type determination means determines that the biological sound information may belong to an abnormal sound
- the degree of abnormality of the abnormal sound is determined by the waveform feature determination means. It is preferable to include an abnormal level determination unit that determines based on the feature.
- the information analysis apparatus not only clarifies the sound type of the body sound information to the user, but also if the body sound information is abnormal sound, the degree of abnormality (abnormal level). ) Can be made clear to the user.
- the sound type determination unit may determine whether the biological sound information corresponds to the type of sound for each type of sound defined in advance.
- the information analysis apparatus determines whether the body sound information corresponds to “normal breathing sound”, whether it corresponds to “respiration sound attenuation”, whether it corresponds to “continuous noise”, and “ Whether or not it falls under “intermittent noise” can be made clear to the user.
- the sound type determination unit may specify which of the plurality of predefined sound types the biological sound information corresponds to.
- the predefined sound types are the above-mentioned “normal breathing sound”, “breathing sound attenuation”, “continuous noise”, and “intermittent noise”.
- the information analysis device clarifies to the user whether the body sound information corresponds to “normal respiratory sound”, “respiratory sound attenuation”, “continuous noise”, or “intermittent noise”. It becomes possible.
- the apparatus further comprises a result output means for outputting a sound type determination result generated by the sound type determination means and indicating a sound type to which the biological sound information belongs to a display unit.
- the result output means preferably stores the sound type determination result in the storage unit in association with the biological sound information.
- the information analysis apparatus of the present invention described above may be mounted on an electronic stethoscope.
- the electronic stethoscope functions as the information analysis apparatus of the present invention.
- the information analysis method provides a waveform feature that indicates a reference for distinguishing a feature of a sound waveform from a sound waveform included in biological sound information collected by a stethoscope.
- a waveform feature determination step for specifying a characteristic of the sound waveform by applying a determination criterion, and a type of sound to which the biological sound information belongs based on the characteristic of the sound waveform specified in the waveform feature determination step
- a sound type determination step for determining the sound quality.
- a measurement system analyzes an electronic stethoscope for performing auscultation on a person to be measured and biological sound information collected by the electronic stethoscope. Based on one of the information analysis devices described above and an auscultation result that is output by the information analysis device and indicates the result of the auscultation performed using the electronic stethoscope, the image capturing process is performed on the subject.
- An auscultation result including at least the presence / absence of an abnormality determined by the information analysis device based on the body sound information and a part from which the body sound information is collected.
- An auscultation result acquisition means for acquiring a position, a part specifying means for specifying a part determined to be abnormal based on the auscultation result acquired by the auscultation result acquisition means, and the part specifying means against sites identified Te, captures an image differently than the imaging for a site other than it, is characterized in that an imaging control unit that acquires image data of the measured person.
- the image capturing apparatus can perform the image capturing process by utilizing the auscultation result output by the information analyzing apparatus. That is, a link between the measurement related to the auscultatory sound and the measurement for image capturing is realized. For example, it is possible to perform image capturing focusing on a specific part where abnormality is recognized in the body sound information. Alternatively, for example, when there is no particular problem with auscultation, the inconvenience that unnecessary image capturing is performed on the part is solved.
- the information analysis apparatus may be realized by a computer, and in this case, a control program for the information analysis apparatus that causes the information analysis apparatus to be realized by the computer by causing the computer to operate as the respective means, and A computer-readable recording medium on which it is recorded also falls within the scope of the present invention.
- the information analysis apparatus of the present invention can process body sound information collected and collected by a stethoscope, and determine the sound type of the body sound based on the characteristics of the sound. Therefore, it can be widely used in a system for grasping the state of a living body that emits the body sound using the body sound information. In particular, it is suitably used for an auscultation system that uses the collected body sound information to grasp the patient's condition and perform medical care.
Abstract
Description
本発明の情報解析装置に関する実施形態について、図1~図30に基づいて説明すると以下の通りである。
図2は、本発明の実施形態における聴診システムの概要を示す図である。図2に示すとおり、聴診システム200は、少なくとも、操作者Uが患者Pの生体音を採取する(すなわち聴診する)ための電子聴診器3と、操作者Uが聴診時に使用する情報解析装置100とを含んで構築される。
図1は、本実施形態における情報解析装置100の要部構成を示す機能ブロック図である。
図1に示すとおり、情報解析装置100の制御部10は、機能ブロックとして、生体音取得部20、生体音処理部21、生体音解析部22、および、結果出力部23を備えている構成である。
まず、生体音処理部21および波形特徴判定部30の各部について詳細に説明する。
図3および図4は、生体音取得部20によって取得される生体音情報の一具体例を示す図である。
周期性判定部31は、図7に示す「判定項目1」を実行することにより、生体音波形の周期性強弱を判定することができる。「判定項目1」では、周期性判定部31は、“周期性が強い”を真、“周期性が弱い”を偽で返す。
周期性判定部31は、「判定項目1’」を実行するときも、上述の「判定項目1」と同様に、「判定項目1-1」および「判定項目1-2」を実行する。ただし、「判定項目1’」では、「判定項目1-1」および「判定項目1-2」の結果を統合する方法が「判定項目1」と異なる。
図9は、生体音取得部20によって取得される生体音情報の他の具体例を示す図である。
スペクトラム判定部32は、図11に示す「判定項目2-A」を実行する。「判定項目2-A」では、スペクトラム判定部32は、スペクトラムにおいて、200Hz以下の周波数成分の和が、全成分の80%以上を占めるか否かを判定する。図8に示すとおり、200Hz以下の周波数成分の和が、全成分の80%以上を占める場合、スペクトラムは正常に近いと推定できる。「判定項目2-A」では、スペクトラム判定部32は、スペクトラムの200Hz以下の周波数成分の和が、全成分の80%以上を占める場合に真(正常に近い)、80%未満の場合に偽(正常ではない可能性がある)を、波形特徴判定結果として音種別判定部40に出力する。
スペクトラム判定部32は、図11に示す「判定項目2-B」を実行する。「判定項目2-B」では、スペクトラム判定部32は、スペクトラムにおいて、200Hz以上の周波数成分の和が、全成分の30%以上を占めるか否かを判定する。図10に示すとおり、200Hz以上の周波数成分が多いと、それを異常の兆候と認めることができる。「判定項目2-B」では、スペクトラム判定部32は、スペクトラムの200Hz以上の周波数成分の和が、全成分の30%以上を占める場合に真(異常兆候あり)、30%未満の場合に偽(異常兆候なし)を、波形特徴判定結果として音種別判定部40に出力する。
図12~図15は、時間周波数解析部213が出力するスペクトログラムの一具体例を示す図である。
スペクトログラム判定部33は、図16に示す「判定項目3-A」を実行する。「判定項目3-A」では、スペクトログラム判定部33は、スペクトログラムの周波数400Hz(もしくはそれ以上)において、一定量以上の周波数成分(濃い部分)の周期性が認められるか否かを判定する。スペクトログラム判定部33は、400Hz以上の帯域で周期性が認められる場合に、「判定項目3-A」を真と判定し、周期性が認められない場合に偽を、波形特徴判定結果をとして音種別判定部40に出力する。
スペクトログラム判定部33は、図16に示す「判定項目3-B」を実行する。「判定項目3-B」では、スペクトログラム判定部33は、スペクトログラムを、高周波帯域(スキャン開始点は、500~400Hzくらいでよい)から低周波帯域へと移行するようにスキャンして、周期性が認められなくなった(または、弱まった)ときの周波数を特定する。そして、スペクトログラム判定部33は、周期性が認められるときの周波数が400Hz未満であれば、「判定項目3-B」を真、400Hz以上であれば偽と判定する。
図17は、包絡線検波部214が出力する、生体音波形の包絡線の一具体例を示す図である。なお、図17に示す生体音波形は、図9に示す生体音情報の音波形の一部を拡大したものである。
包絡線判定部34は、図18に示す「判定項目4」を実行する。「判定項目4」では、包絡線判定部34は、音波形の包絡線について、音の連続性が認められるか否か判定する。
図20は、インパルスノイズ検出部215が出力する、生体音波形においてインパルスノイズが特定されたインパルスノイズ検出結果の一具体例を示す図である。
インパルスノイズ判定部35は、図21に示す「判定項目5」を実行する。「判定項目5」では、インパルスノイズ判定部35は、音波形に含まれるインパルスノイズの数が、1周期あたりに10個以上存在したか否かを判定する。
図22は、波形特徴判定部30によって出力された波形特徴判定結果を入力として、音種別判定部40の正常呼吸音判定部41が出力する音種別判定結果の一具体例を示す図である。
図23は、波形特徴判定部30によって出力された波形特徴判定結果を入力として、音種別判定部40の呼吸音減弱判定部42が出力する音種別判定結果の一具体例を示す図である。
図24は、波形特徴判定部30によって出力された波形特徴判定結果を入力として、音種別判定部40の連続性雑音判定部43が出力する音種別判定結果の一具体例を示す図である。
図25は、波形特徴判定部30によって出力された波形特徴判定結果を入力として、音種別判定部40の断続性雑音判定部44が出力する音種別判定結果の一具体例を示す図である。
また、図29に示すとおり、音種別判定部40の各部は、生体音情報に含まれるすべての音波形の中から、その種別の異常が認められた頻度をカウントして、結果出力部23に出力する構成であってもよい。例えば、連続性雑音判定部43は、40秒間(呼吸の約10周期分)の生体音波形を解析し、図24に示す、判定パターン(ア)に合致する波形がいくつ検出されたのかをカウントすることができる。そして、連続性雑音判定部43は、10周期のうち、連続性雑音が検出された回数を、音種別判定結果とともに、結果出力部23に供給することができる。
減弱レベル判定部51は、呼吸音減弱判定部42によって、生体音が「真:呼吸音減弱である可能性がある」と判定された場合に、当該生体音の音波形の減弱のレベルを判定するものである。
連続性レベル判定部52は、連続性雑音判定部43によって、生体音が「真:連続性雑音である可能性がある」と判定された場合に、当該生体音の音波形の連続性のレベルを判定するものである。
断続性レベル判定部53は、断続性雑音判定部44によって、生体音が「真:断続性雑音である可能性がある」と判定された場合に、当該生体音の音波形の断続性のレベルを判定するものである。
図30は、本実施形態における情報解析装置100の情報解析処理の流れを示すフローチャートである。
自己相関解析部211が、音波形から自己相関関数(波形特徴情報)を導出すること、
フーリエ変換部212が、音波形からスペクトラム(波形特徴情報)を導出すること、
時間周波数解析部213が、音波形からペクトログラム(波形特徴情報)を導出すること、
包絡線検波部214が、音波形の包絡線(波形特徴情報)を検波すること、および、
インパルスノイズ検出部215が、音波形のインパルスノイズを特定して、インパルスノイズ検出結果(波形特徴情報)を出力すること、
などが含まれるが、これに限定されない。また、生体音処理部21は、上述のすべての波形特徴情報を生成する構成であってもよいし、一部を生成する構成であってもよい。
周期性判定部31が、「判定項目1」または「判定項目1’」を実行して、生体音の周期性に係る特徴を判定すること、
スペクトラム判定部32が、「判定項目2-A」または「判定項目2-B」を実行して、生体音の周波数成分分布に係る特徴を判定すること、
スペクトログラム判定部33が、「判定項目3-A」または「判定項目3-B」を実行して、生体音の時間周波数成分分布における周期性に係る特徴を判定すること、
包絡線判定部34が、「判定項目4」を実行して、生体音に含まれる雑音の連続性に係る特徴を判定すること、および、
インパルスノイズ判定部35が、「判定項目5」を実行して、生体音に含まれる雑音の断続性に係る特徴を判定すること、
などが含まれるが、これに限定されない。また、波形特徴判定部30は、上述のすべての判定項目を実施してもよいし、一部の判定項目を実施してもよい。
正常呼吸音判定部41が、上記生体音が「正常呼吸音」である可能性があるか否かを判定すること、
呼吸音減弱判定部42が、上記生体音が「呼吸音減弱」である可能性があるか否かを判定すること、
連続性雑音判定部43が、上記生体音が「連続性雑音」である可能性があるか否かを判定すること、および、
断続性雑音判定部44が、上記生体音が「断続性雑音」である可能性があるか否かを判定すること、
などが含まれるが、これに限定されない。また、音種別判定部40は、上述のすべての音種別に係る判定を実施してもよいし、一部の種別についてのみ判定を実施してもよい。
減弱レベル判定部51が、スペクトログラムに対して減弱レベル判定を行って、減弱レベル判定結果を生成すること、
連続性レベル判定部52が、包絡線に対して連続性レベル判定を行って、連続性レベル判定結果を生成すること、および、
断続性レベル判定部53が、インパルスノイズ検出結果に対して断続性レベル判定を行って、断続性レベル判定結果を生成すること、
などが含まれるが、これに限定されない。また、異常レベル判定部50は、上述のすべての異常音の音種別に係るレベル判定を実施してもよいし、一部の異常音種別についてのみレベル判定を実施してもよい。
「呼吸音減弱」である可能性があるか否かを判定すること、
「連続性雑音」である可能性があるか否かを判定すること、および、
「断続性雑音」である可能性があるか否かを判定すること、
を実行し、対象の呼吸音が、いずれの異常音でもないと判定された場合に、正常呼吸音判定部41が、当該呼吸音を正常(の可能性がある)と判定してもよい。
本発明の情報解析装置に関する他の実施形態について、図31~図37に基づいて説明すれば、以下のとおりである。なお、説明の便宜上、上述の実施形態1にて説明した図面と同じ機能を有する部材については、同じ符号を付記し、その説明を省略する。
図31は、本実施形態における情報解析装置100の要部構成を示す機能ブロック図である。
図32は、本実施形態における総合判定部45が、患者Pから取得された呼吸器系音を所定の音種別に分類する際の種別体系を示す図である。図32に示すとおり、総合判定部45は、本実施形態では、「呼吸器系音」を、『正常呼吸音』、『呼吸音減弱』、『その他の異常音』、『高音性連続性雑音』、『低音性連続性雑音』、『細かい断続性雑音』、『荒い断続性雑音』、および、『その他の雑音』のいずれかの音種別に分類する。そして、特定した音種別を、総合判定結果として結果出力部23に出力する。
図33Aおよび図33Bは、本実施形態における情報解析装置100の情報解析処理の流れを示すフローチャートである。なお、本実施形態では、図30に示すS1およびS2は、図33AのS101に先行して実行済みであるものとする。
次に、図34~図36を参照しながら、異常レベル判定部50の各部が実行する異常レベル判定処理の流れを説明する。なお、図34~図36に示す異常レベル判定部50の各部の処理の流れは、実施形態1および実施形態2に共通するものである。
上述したとおり、結果出力部23は、総合判定部45が出力した、上記生体音を上述のいずれかの音種別に分類した総合判定結果を表示部12に表示する。例えば、図37に示すとおり、解析結果を表示する領域に、上記総合判定結果を表示する。図37には、総合判定部45が、生体音を『高音性連続性雑音』に分類したときの総合判定結果の例を示している。さらに、総合判定部45が、上記生体音の波形における異常の出現頻度をカウントしている場合には、結果出力部23は、総合判定部45から取得した出現頻度を併せて表示部12に表示してもよい。
上述の実施形態1および2では、本発明の聴診システム200において、呼吸音などの情報を解析する機能を、操作者Uが操作する端末装置としての情報解析装置100によって実現する場合について説明した。そして、上述の実施形態1および2では、聴診システム200において、情報解析装置100が、電子聴診器3とサポートセンター2の管理サーバ4と通信する構成であった。
本発明の他の実施形態について、図38に基づいて説明すれば、以下のとおりである。なお、説明の便宜上、上述の実施形態1および2にて説明した図面と同じ機能を有する部材については、同じ符号を付記し、その説明を省略する。
特許文献3には、医用画像を作成する際に、生体の所定部位が撮影された画像データを取得した後、撮影された部位に対して生体音測定を行い、それらの結果を関連付けて、当該医用画像を表示する医用画像表示システムが記載されている。
図38は、実施形態3に係る測定システム3600の概要、および、測定システム3600を構成する画像撮像装置3006の要部構成を示すブロック図である。
画像撮像装置3006は、図38に示すとおり、聴診システム200の各装置との間で情報の送受信を行う通信部3011、画像撮像装置3006が処理する各種情報を記憶するための記憶部3012、患者の撮影を実施する撮像部3013、および、画像撮像装置3006の各部を統括して制御する制御部3010を含む構成となっている。
最後に、情報解析装置100の各ブロック、特に、生体音取得部20、生体音処理部21、生体音解析部22および結果出力部23、ならびに、生体音処理部21の各ブロックおよび生体音解析部22の各ブロックは、ハードウェアロジックによって構成してもよいし、次のようにCPUを用いてソフトウェアによって実現してもよい。
本発明に係る情報解析装置は、上記課題を解決するために、聴診器によって採取された生体音情報に含まれる音波形に対して、音波形の特徴を区分するための基準を示した波形特徴判定基準を適用して、上記音波形の特徴を特定する波形特徴判定手段と、上記波形特徴判定手段によって特定された、上記音波形の特徴に基づいて、上記生体音情報が属する音の種別を判定する音種別判定手段とを備えていることを特徴としている。
2 サポートセンター(遠隔地)
3 電子聴診器(聴診器)
4 管理サーバ
5 通信網
10 制御部
11 入力部
12 表示部
13 記憶部
14 通信部
20 生体音取得部(生体音取得手段)
21 生体音処理部(生体音処理手段)
22 生体音解析部(生体音解析手段)
23 結果出力部(結果出力手段)
30 波形特徴判定部(波形特徴判定手段)
31 周期性判定部(波形特徴判定手段/周期性判定手段)
32 スペクトラム判定部(波形特徴判定手段/周波数成分分布判定手段)
33 スペクトログラム判定部(波形特徴判定手段/周波数帯域別周期性判定手段)
34 包絡線判定部(波形特徴判定手段/包絡線判定手段)
35 インパルスノイズ判定部(波形特徴判定手段/インパルスノイズ判定手段)
40 音種別判定部(音種別判定手段)
41 正常呼吸音判定部(音種別判定手段/正常呼吸音判定手段)
42 呼吸音減弱判定部(音種別判定手段/呼吸音減弱判定手段)
43 連続性雑音判定部(音種別判定手段/連続性雑音判定手段)
44 断続性雑音判定部(音種別判定手段/断続性雑音判定手段)
45 総合判定部(音種別判定手段/総合判定手段)
50 異常レベル判定部(異常レベル判定手段)
51 減弱レベル判定部(異常レベル判定手段/減弱レベル判定手段)
52 連続性レベル判定部(異常レベル判定手段/連続性レベル判定手段)
53 断続性レベル判定部(異常レベル判定手段/断続性レベル判定手段)
100 情報解析装置
200 聴診システム
211 自己相関解析部(生体音処理手段)
212 フーリエ変換部(生体音処理手段)
213 時間周波数解析部(生体音処理手段)
214 包絡線検波部(生体音処理手段)
215 インパルスノイズ検出部(生体音処理手段)
3006 画像撮像装置
3010 制御部
3011 通信部
3012 記憶部
3013 撮像部
3020 聴診結果取得部(聴診結果取手段)
3021 撮像部位特定部(部位特定手段)
3022 撮像制御部(撮像制御手段)
3600 測定システム
Claims (24)
- 聴診器によって採取された生体音情報に含まれる音波形に対して、音波形の特徴を区分するための基準を示した波形特徴判定基準を適用して、上記音波形の特徴を特定する波形特徴判定手段と、
上記波形特徴判定手段によって特定された、上記音波形の特徴に基づいて、上記生体音情報が属する音の種別を判定する音種別判定手段とを備えていることを特徴とする情報解析装置。 - 上記波形特徴判定基準は、上記音波形より導出された特徴量と比較する閾値と、該閾値によって定められた条件とを含み、
上記波形特徴判定手段は、上記音波形の特徴量が上記条件に合致するか否かを判定することにより、上記音波形の特徴を特定することを特徴とする請求項1に記載の情報解析装置。 - 上記音種別判定手段が判定する音の種別は、
生体が発する呼吸音が正常であることを示す「正常呼吸音」
生体が発する呼吸音が聴診器によって採取されるまでに減衰していることを示す「呼吸音減弱」
生体が発する呼吸音に連続する雑音が含まれていることを示す「連続性雑音」、および、
生体が発する呼吸音に断続する雑音が含まれていることを示す「断続性雑音」、
の少なくともいずれか1つであることを特徴とする請求項1または2に記載の情報解析装置。 - 上記波形特徴判定手段は、
包絡線に係る波形特徴判定基準にしたがって、音波形の包絡線が、一定以上の振幅値で連続するか否かを判定し、
上記音種別判定手段は、
上記包絡線が連続すると判定された場合に、上記生体音情報が「連続性雑音」に属する可能性があると判定することを特徴とする請求項3に記載の情報解析装置。 - 上記波形特徴判定手段は、
インパルスノイズ数に係る波形特徴判定基準にしたがって、音波形が一定数以上のインパルスノイズを含むか否かを判定し、
上記音種別判定手段は、
上記音波形が一定数以上のインパルスノイズを含むと判定された場合に、上記生体音情報が「断続性雑音」に属する可能性があると判定することを特徴とする請求項3または4に記載の情報解析装置。 - 上記波形特徴判定手段は、上記波形特徴判定基準にしたがって、
包絡線の振幅平均値を超える連続する時間が200ms以上ある場合に、上記包絡線が連続すると判定することを特徴とする請求項4に記載の情報解析装置。 - 上記波形特徴判定手段は、上記波形特徴判定基準にしたがって、
所定期間内の包絡線において振幅平均値を超える時間の合計が200ms以上ある場合に、上記包絡線が連続すると判定することを特徴とする請求項4または6に記載の情報解析装置。 - 上記波形特徴判定手段は、上記波形特徴判定基準にしたがって、
上記音波形が、インパルスノイズを1周期あたりに10個以上含む場合に、上記音波形が一定数以上のインパルスノイズを含むと判定することを特徴とする請求項5に記載の情報解析装置。 - 上記波形特徴判定手段は、
周波数成分分布に係る波形特徴判定基準にしたがって、上記音波形の周波数成分分布が、正常の傾向を示すか否か、または、異常の傾向を示すか否かを判定し、
上記音種別判定手段は、
上記音波形の周波数成分分布が正常の傾向を示すと判定された場合に、上記生体音情報が、「正常呼吸音」および「呼吸音減弱」の少なくともいずれか一方に属する可能性があると判定し、
上記音波形の周波数成分分布が異常の傾向を示すと判定された場合に、上記生体音情報が、「連続性雑音」および「断続性雑音」の少なくともいずれか一方に属する可能性があると判定することを特徴とする請求項3から8までのいずれか1項に記載の情報解析装置。 - 上記波形特徴判定手段は、上記波形特徴判定基準にしたがって、
上記音波形の周波数成分分布において、200Hz以下の周波数成分の和が全体の80%以上を占める場合に、当該周波数成分分布が、正常の傾向を示すと判定し、
上記音波形の周波数成分分布において、200Hz以上の周波数成分の和が全体の30%以上を占める場合に、当該周波数成分分布が、異常の傾向を示すと判定することを特徴とする請求項9に記載の情報解析装置。 - 上記波形特徴判定手段は、
音波形における周期性が強いか否かを区分するための波形特徴判定基準にしたがって、上記音波形の周期性が強いか否かを判定し、
上記音種別判定手段は、
上記音波形の周期性が強いと判定された場合に、上記生体音情報が、「正常呼吸音」および「呼吸音減弱」の少なくともいずれか一方に属する可能性があると判定し、
上記音波形の周期性が弱いと判定された場合に、上記生体音情報が、「連続性雑音」および「断続性雑音」の少なくともいずれか一方に属する可能性があると判定することを特徴とする請求項3から10までのいずれか1項に記載の情報解析装置。 - 上記波形特徴判定手段は、
時間周波数解析に基づく周波数成分分布に係る波形特徴判定基準にしたがって、上記音波形の、周波数帯域別の周期性の有無を判定し、
上記音種別判定手段は、
上記時間周波数解析に基づく周波数成分分布において、高周波数帯域において周期性が有ると判定された場合に、上記生体音情報が「正常呼吸音」に属する可能性があると判定し、
上記時間周波数解析に基づく周波数成分分布において、低周波数帯域において周期性が有って、高周波数帯域において周期性が無いと判定された場合に、上記生体音情報が「呼吸音減弱」に属する可能性があると判定することを特徴とする請求項3から11までのいずれか1項に記載の情報解析装置。 - 上記波形特徴判定手段は、上記波形特徴判定基準にしたがって、
上記音波形の自己相関関数が2~5秒間隔でピークを持つ場合、かつ、
上記自己相関関数の包絡線における、所定振幅値における、包絡線ピークの期間が呼吸周期の10%以下である場合に、上記音波形の周期性が強いと判定することを特徴とする請求項11に記載の情報解析装置。 - 上記波形特徴判定手段は、上記波形特徴判定基準にしたがって、
上記音波形の時間周波数解析に基づく周波数成分分布において、400Hz以上の帯域において周期性が認められるときに、高周波数帯域において周期性が有ると判定し、
上記音波形の時間周波数解析に基づく周波数成分分布において、周期性が認められるのが400Hz未満の帯域のときに、低周波数帯域において周期性が有って、高周波数帯域において周期性が無いと判定することを特徴とする請求項12に記載の情報解析装置。 - 上記音種別判定手段が、上記生体音情報が異常音に属する可能性があると判定した場合に、上記異常音の異常の程度を、上記波形特徴判定手段によって特定された上記音波形の特徴に基づいて判定する異常レベル判定手段を備えていることを特徴とする請求項1から14までのいずれか1項に記載の情報解析装置。
- 上記音種別判定手段は、
予め定義された音の種別ごとに、上記生体音情報が当該音の種別に該当するか否かを判定することを特徴とする請求項1から15までのいずれか1項に記載の情報解析装置。 - 上記音種別判定手段は、
上記生体音情報が、予め定義された複数の音の種別のいずれに該当するかを特定することを特徴とする請求項1から15までのいずれか1項に記載の情報解析装置。 - 上記音種別判定手段によって生成された、上記生体音情報が属する音の種別を示した音種別判定結果を、表示部に出力する結果出力手段を備えていることを特徴とする請求項1から17までのいずれか1項に記載の情報解析装置。
- 上記結果出力手段は、
上記音種別判定結果を上記生体音情報に関連付けて記憶部に記憶することを特徴とする請求項18に記載の情報解析装置。 - 請求項1から19までのいずれか1項に記載の情報解析装置を備えていることを特徴とする電子聴診器。
- 聴診器によって採取された生体音情報に含まれる音波形に対して、音波形の特徴を区分するための基準を示した波形特徴判定基準を適用して、上記音波形の特徴を特定する波形特徴判定ステップと、
上記波形特徴判定ステップにて特定された、上記音波形の特徴に基づいて、上記生体音情報が属する音の種別を判定する音種別判定ステップとを含むことを特徴とする情報解析方法。 - 被測定者に対して聴診を実施するための電子聴診器と、
上記電子聴診器によって採取された生体音情報を解析する請求項1から19までのいずれか1項に記載の情報解析装置と、
上記情報解析装置によって出力された、上記電子聴診器を用いて実施された聴診の結果を示す聴診結果に基づいて、上記被測定者に対して画像撮像処理を実施する画像撮像装置とを含み、
上記画像撮像装置は、
上記情報解析装置が上記生体音情報に基づいて判定した異常の有無と、上記生体音情報が採取された部位とを少なくとも含む聴診結果を取得する聴診結果取得手段と、
上記聴診結果取得手段によって取得された聴診結果に基づいて、異常が有ると判定された部位を特定する部位特定手段と、
上記部位特定手段によって特定された部位に対して、それ以外の部位に対する撮像とは異なる仕方で撮像を行い、上記被測定者の画像データを取得する撮像制御手段とを備えていることを特徴とする、測定システム。 - コンピュータを、請求項1から19までのいずれか1項に記載の情報解析装置の各手段として機能させるための制御プログラム。
- 請求項23に記載の制御プログラムを記録したコンピュータ読み取り可能な記録媒体。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US14/353,545 US20140276229A1 (en) | 2011-12-13 | 2012-12-10 | Information analyzing apparatus, digital stethoscope, information analyzing method, measurement system, control program, and recording medium |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2011-272785 | 2011-12-13 | ||
JP2011272785 | 2011-12-13 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2013089073A1 true WO2013089073A1 (ja) | 2013-06-20 |
Family
ID=48612520
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2012/081960 WO2013089073A1 (ja) | 2011-12-13 | 2012-12-10 | 情報解析装置、電子聴診器、情報解析方法、測定システム、制御プログラム、および、記録媒体 |
Country Status (3)
Country | Link |
---|---|
US (1) | US20140276229A1 (ja) |
JP (1) | JPWO2013089073A1 (ja) |
WO (1) | WO2013089073A1 (ja) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2015084853A (ja) * | 2013-10-29 | 2015-05-07 | パイオニア株式会社 | 信号処理装置及び方法、並びにコンピュータプログラム及び記録媒体 |
US20150196269A1 (en) * | 2014-01-15 | 2015-07-16 | Xerox Corporation | System and method for remote determination of acute respiratory infection |
EP3165164A4 (en) * | 2014-07-01 | 2018-02-21 | Pioneer Corporation | Respiratory sound analysis device, respiratory sound analysis method, computer program, and recording medium |
JP2019017844A (ja) * | 2017-07-20 | 2019-02-07 | パイオニア株式会社 | 呼吸音処理装置、呼吸音処理方法、及び、コンピュータプログラム |
US11529072B2 (en) * | 2012-06-18 | 2022-12-20 | AireHealth Inc. | Method and apparatus for performing dynamic respiratory classification and tracking of wheeze and crackle |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2017060828A1 (en) * | 2015-10-08 | 2017-04-13 | Cordio Medical Ltd. | Assessment of a pulmonary condition by speech analysis |
US9931099B1 (en) * | 2016-10-04 | 2018-04-03 | Faiz Fatteh | Electronic stethoscope lacking an earpiece assembly |
FR3061392B1 (fr) * | 2016-12-27 | 2019-08-30 | Somfy Sas | Procede de controle de presence et systeme de surveillance |
US11517256B2 (en) * | 2016-12-28 | 2022-12-06 | Koninklijke Philips N.V. | Method of characterizing sleep disordered breathing |
CN110352349B (zh) * | 2017-02-15 | 2023-01-31 | 日本电信电话株式会社 | 异常音检测装置、异常度计算装置、异常音生成装置、异常信号检测装置、及其方法、记录介质 |
EP3424432B1 (en) | 2017-07-04 | 2020-09-30 | Tata Consultancy Services Limited | Systems and methods for detecting pulmonary abnormalities using lung sounds |
US11547381B2 (en) * | 2017-10-10 | 2023-01-10 | University Of Southern California | Wearable respiratory monitoring system based on resonant microphone array |
US11284827B2 (en) | 2017-10-21 | 2022-03-29 | Ausculsciences, Inc. | Medical decision support system |
CN108784733B (zh) * | 2018-06-28 | 2022-04-12 | 上海掌门科技有限公司 | 听诊系统及方法 |
US20220160325A1 (en) * | 2020-11-24 | 2022-05-26 | RTM Vital Signs LLC | Method of determining respiratory states and patterns from tracheal sound analysis |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005066045A (ja) * | 2003-08-25 | 2005-03-17 | Konica Minolta Medical & Graphic Inc | 音データ処理装置及びプログラム |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5218969A (en) * | 1988-02-04 | 1993-06-15 | Blood Line Technology, Inc. | Intelligent stethoscope |
US6790183B2 (en) * | 1998-10-14 | 2004-09-14 | Raymond L. H. Murphy | Method and apparatus for displaying body sounds and performing diagnosis based on body sound analysis |
KR20030065228A (ko) * | 2002-01-31 | 2003-08-06 | 김종수 | 청진음향신호를 표시하는 출력시스템 |
US20080013747A1 (en) * | 2006-06-30 | 2008-01-17 | Bao Tran | Digital stethoscope and monitoring instrument |
JP2009106574A (ja) * | 2007-10-31 | 2009-05-21 | Konica Minolta Medical & Graphic Inc | 呼吸音解析装置及びプログラム |
JP5093537B2 (ja) * | 2008-10-16 | 2012-12-12 | 国立大学法人 長崎大学 | 音情報判定支援方法、音情報判定方法、音情報判定支援装置、音情報判定装置、音情報判定支援システム及びプログラム |
-
2012
- 2012-12-10 US US14/353,545 patent/US20140276229A1/en not_active Abandoned
- 2012-12-10 WO PCT/JP2012/081960 patent/WO2013089073A1/ja active Application Filing
- 2012-12-10 JP JP2013549260A patent/JPWO2013089073A1/ja active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005066045A (ja) * | 2003-08-25 | 2005-03-17 | Konica Minolta Medical & Graphic Inc | 音データ処理装置及びプログラム |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11529072B2 (en) * | 2012-06-18 | 2022-12-20 | AireHealth Inc. | Method and apparatus for performing dynamic respiratory classification and tracking of wheeze and crackle |
JP2015084853A (ja) * | 2013-10-29 | 2015-05-07 | パイオニア株式会社 | 信号処理装置及び方法、並びにコンピュータプログラム及び記録媒体 |
US20150196269A1 (en) * | 2014-01-15 | 2015-07-16 | Xerox Corporation | System and method for remote determination of acute respiratory infection |
EP3165164A4 (en) * | 2014-07-01 | 2018-02-21 | Pioneer Corporation | Respiratory sound analysis device, respiratory sound analysis method, computer program, and recording medium |
JP2019017844A (ja) * | 2017-07-20 | 2019-02-07 | パイオニア株式会社 | 呼吸音処理装置、呼吸音処理方法、及び、コンピュータプログラム |
JP7041911B2 (ja) | 2017-07-20 | 2022-03-25 | パイオニア株式会社 | 呼吸音処理装置、呼吸音処理方法、及び、コンピュータプログラム |
Also Published As
Publication number | Publication date |
---|---|
US20140276229A1 (en) | 2014-09-18 |
JPWO2013089073A1 (ja) | 2015-04-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2013089073A1 (ja) | 情報解析装置、電子聴診器、情報解析方法、測定システム、制御プログラム、および、記録媒体 | |
JP2013123494A (ja) | 情報解析装置、情報解析方法、制御プログラム、および、記録媒体 | |
EP2621336B1 (en) | Apparatus and method for diagnosing obstructive sleep apnea | |
US6261238B1 (en) | Phonopneumograph system | |
JP5977255B2 (ja) | 嚥下障害検出装置及びその作動方法 | |
WO2013089072A1 (ja) | 情報管理装置、情報管理方法、情報管理システム、聴診器、情報管理プログラム、測定システム、制御プログラムおよび記録媒体 | |
US20130060100A1 (en) | Contactless non-invasive analyzer of breathing sounds | |
JP2017536905A (ja) | 音響的監視システム、監視方法及び監視するコンピュータプログラム | |
Chamberlain et al. | Mobile stethoscope and signal processing algorithms for pulmonary screening and diagnostics | |
JP2013518607A (ja) | 携帯型モニタリングのための生理学的信号の品質を分類する方法およびシステム | |
US11484283B2 (en) | Apparatus and method for identification of wheezing in ausculated lung sounds | |
US20060116878A1 (en) | Asthma diagnostic apparatus, asthma diagnostic method, and storage medium storing asthma diagnostic program | |
JP2013123495A (ja) | 呼吸音解析装置、呼吸音解析方法、呼吸音解析プログラムおよび記録媒体 | |
US11813109B2 (en) | Deriving insights into health through analysis of audio data generated by digital stethoscopes | |
US20190388006A1 (en) | Non-invasive system and method for breath sound analysis | |
US20170258350A1 (en) | Heart Murmur Detection Device and Method Thereof | |
Mondal et al. | Diagnosing of the lungs status using morphological anomalies of the signals in transformed domain | |
Pal et al. | Automatic breathing phase identification based on the second derivative of the recorded lung sounds | |
Wang et al. | Using breath sound data to detect intraoperative respiratory depression in non-intubated anesthesia | |
Phettom et al. | Automatic Identification of Abnormal Lung Sounds Using Time-Frequency Analysis and Convolutional Neural Network | |
JP2012223509A (ja) | 聴診音解析装置および聴診音解析プログラム | |
Michiels et al. | Jamf-based representation for computational lung sound analysis |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 12858549 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2013549260 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14353545 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 12858549 Country of ref document: EP Kind code of ref document: A1 |