WO2001091691A1 - Artificial intelligence incubator system and control method thereof - Google Patents

Artificial intelligence incubator system and control method thereof Download PDF

Info

Publication number
WO2001091691A1
WO2001091691A1 PCT/KR2001/000767 KR0100767W WO0191691A1 WO 2001091691 A1 WO2001091691 A1 WO 2001091691A1 KR 0100767 W KR0100767 W KR 0100767W WO 0191691 A1 WO0191691 A1 WO 0191691A1
Authority
WO
WIPO (PCT)
Prior art keywords
nurture
treatment equipment
setpoint value
value
life body
Prior art date
Application number
PCT/KR2001/000767
Other languages
French (fr)
Inventor
Gun-Bok Lee
Sang-Hyun Byun
Original Assignee
P & M Co., Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by P & M Co., Ltd. filed Critical P & M Co., Ltd.
Publication of WO2001091691A1 publication Critical patent/WO2001091691A1/en

Links

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
    • A61G—TRANSPORT, PERSONAL CONVEYANCES, OR ACCOMMODATION SPECIALLY ADAPTED FOR PATIENTS OR DISABLED PERSONS; OPERATING TABLES OR CHAIRS; CHAIRS FOR DENTISTRY; FUNERAL DEVICES
    • A61G11/00—Baby-incubators; Couveuses
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
    • A61B5/02055—Simultaneously evaluating both cardiovascular condition and temperature
    • 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
    • A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
    • A61M16/00—Devices for influencing the respiratory system of patients by gas treatment, e.g. mouth-to-mouth respiration; Tracheal tubes
    • A61M16/021—Devices for influencing the respiratory system of patients by gas treatment, e.g. mouth-to-mouth respiration; Tracheal tubes operated by electrical means
    • A61M16/022—Control means therefor
    • A61M16/024—Control means therefor including calculation means, e.g. using a processor
    • A61M16/026—Control means therefor including calculation means, e.g. using a processor specially adapted for predicting, e.g. for determining an information representative of a flow limitation during a ventilation cycle by using a root square technique or a regression analysis
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems

Definitions

  • the present invention relates to an intelligent type incubator system and a control method thereof; and more particularly, to an intelligent type incubator system and a control method thereof, in which various kinds of information data concerning checked state of a patient are processed as control data of various kinds of treatment equipments for maintaining a normal state of a corresponding patient, and the outputs of the corresponding treatment equipments are automatically controlled, to thereby make a state monitoring of the patient and its diagnosis and treatment operation, closed-loop automatic controlled.
  • an incubator system is provided as an important medical equipment having a function of temporarily nursing a patient under an abnormal state like an immature infant, and of recovering to a normal state of the patient through an appropriate treatment.
  • Fig. 1 is a block diagram of a conventional incubator system, and this system is constructed by an incubator 12 for providing a proper environmental standard such as temperature, humidity and oxygen etc. within a closed space where there is a patient 11; an artificial respiratory machine 13 for controlling the breathing number and breathing pressure of the patient 11 etc.
  • a therapeutic medicine injector 14 for injecting various therapeutic medicines to the patient 11; other equipments (not shown) for a treatment of the patient; and several kinds of measuring equipments 15 for measuring an electrocardiogram (ECG), blood pressure and the breathing number of the patient 11, etc.
  • ECG electrocardiogram
  • the present invention is directed to an intelligent type incubator system and a control method thereof that substantially obviates one or more of the limitations and disadvantages of the related art.
  • a primary object of the present invention is to provide an intelligent type incubator system and a control method thereof, which is capable of making a state monitoring of a patient and its diagnosis and treatment operation, closed-loop automatic controlled.
  • the intelligent type incubator system is constructed by a nurture/treatment equipment which includes an environment controlling part for controlling an air environment containing temperature, humidity and an oxygen amount within a given space according to a corresponding setpoint value, an artificially breathing part for artificially breathing of a lungs-breath life body according to the corresponding setpoint value, and an injection part for injecting various kinds of injection agents onto the life body according to the corresponding setpoint value; a measuring unit for measuring a state of the life body like a medical and biological metabolism etc., the life body being as the object body under a nurture and/or a treatment through the nurture/treatment equipment; and an artificial intelligence controlling unit for processing state data of the life body measured by the measuring unit, as a control variable value for various setpoint values of the nurture/treatment equipment, on the basis of a neural network driven-type fuzzy inference model based on statistical information for a correlation between a state of the life body and an output of the nurture/treatment equipment, and for controlling the output of the nurture/treatment equipment
  • the nurture/treatment equipment in a controlling method of the intelligent type incubator system having the nurture/treatment equipment and manually varying a setpoint value of the nurture/treatment equipment according to a state of a corresponding life body, the nurture/treatment equipment containing an incubator for controlling an air environment containing temperature, humidity and an oxygen amount within a given space according to the setpoint value, an artificial respiratory machine for artificially breathing of a lungs-breath life body according to the setpoint value, and an injecting apparatus for injecting various kinds of injection agents onto the life body according to the setpoint value; the method includes the steps of: constructing a plural number of fuzzy inference rules on the basis of a statistical distribution of the setpoint value of the nurture/treatment equipment on the state data of the life body, and forming the plural number of fuzzy inference rules as the structure of the neural network; dividing the state data of the life body and its based setpoint value of the nurture/treatment equipment into a plurality of groups corresponding to the plural number of fuzzy inference rules and establishing them,
  • a treatment experience of a corresponding expert doctor and a special knowledge for the life body are established as a knowledge -based system of an artificial intelligence so that system itself has a judgment function, thereby, a corresponding nurture/treatment equipment is appropriately controlled so as to lead the life body to be normally nurtured, treated and restored to health.
  • Fig. 1 is a block diagram of a conventional incubator system
  • Fig. 2 indicates a block diagram of an intelligent type incubator system in one embodiment of the present invention
  • Fig. 3 depicts a constructive diagram for an algorithm of a neural network driven-type fuzzy inference model of an artificial intelligence controller shown in Fig. 2;
  • Fig. 4 is a drawing showing, on a two dimensional plane, group data classified under an assumption that there are two inputs to embody a neural network driven-type fuzzy inference model of an artificial intelligence controller shown in Fig. 2;
  • Fig. 5 sets forth a drawing showing another constructive example for embodying a inference rule of an artificial intelligence controller shown in Fig. 2.
  • an intelligent type incubator system and a controlling method thereof are described in detail, as follows.
  • Fig. 2 is a block diagram of an intelligent type incubator system in one embodiment of the present invention
  • the intelligent type incubator system includes a nurture/treatment equipment 20 which has an incubator 21 for outputting temperature of air, humidity, oxygen, pressure, and air controlled by each setpoint value for an air flowing quantity, and controlling a quality of air supplied into the inside of a given space, an artificial respiratory machine 22 for outputting oxygen controlled according to each setpoint value for expiration/inspiration pressure, oxygen content, the breathing number per unit time, humidity and temperature, and for artificially breathing of the lungs breathing life body (hereinafter, referred to as "patient") 30, a therapeutic medicine injector 23 for controlling and outputting an actual injection amount according to a setpoint value concerning the injection amount and for injecting various kinds of a therapeutic medicine to the patient 30; a measuring equipment 40 for measuring and inputting various kinds of state data which contains an electrocardiogram (ECG) having irregular pulse and the pulse number representing a state of a medical and biological metabolism etc.
  • ECG electrocardiogram
  • an artificial intelligence controller 50 for processing various kinds of state data of the patient 30 measured by the measuring equipment 40, as a control variable value for various setpoint values of the nurture/treatment equipment 20, on the basis of a neural network driven-type fuzzy inference model based on statistical information for a correlation between the state data of the patient 30 and an output of the nurture/treatment equipment 20 on the state data, and for controlling various outputs of the nurture/treatment equipment 20 on the basis of the processed control variable value.
  • a closed-loop system constructed by the nurture/treatment equipment 20, the patient 30, the measuring equipment 40 and the artificial intelligence controller 50.
  • the artificial intelligence controller 50 receives in feedback a state of the patient 30 through the measuring equipment 40, and processes and outputs the inputted state data of the patient, as the control variable value for various setpoint values, on the basis of the neural network driven-type fuzzy inference model, and sends out it as an input signal of the nurture/treatment equipment 20 containing the incubator 21, the artificial respiratory machine 22 and/or the injector 23, to thereby control the control object like the patient 30 through various kinds of nurture/treatment equipments 20.
  • the artificial controller 50 performs a function of recovering the patient to a stabilized balance state, namely, a normal state, through an appropriate procedure, on the basis of knowledge for a current state and a peculiarity of the patient 30, namely, treatment experience of a doctor and medical knowledge for the human body etc.
  • An input of the incubator 21 become a setpoint values for humidity, oxygen partial pressure and temperature of air supplied to the incubator 21 inside, and an air flowing amount, and an output thereof becomes the air controlled by the setpointed value.
  • An input of the artificial respiratory machine 22 becomes a setpoint value for expiration/inspiration pressure, an oxygen content, the breathing number per unit time, humidity and temperature, and an output thereof becomes the oxygen controlled by the setpoint value.
  • An input of the therapeutic medicine injector 23 becomes a setpoint value for the amount of the therapeutic medicine injection, and an output thereof becomes an actual injection amount based on the setpoint value.
  • An input of the patient 30 becomes various kinds of outputs of the nurture/treatment equipment 20 which contains the incubator 21, the respiratory machine 22 and the injector 23, and an output thereof becomes, a state of the patient, the electrocardiogram (ECG) having the irregular pulse and the pulse number, blood pressure, the breathing number, the oxygen saturation degree, the percutaneous oxygen partial pressure, the percutaneous dioxide partial pressure, the body temperature, the weight and age, etc.
  • ECG electrocardiogram
  • the artificial intelligence controller 50 gains information for value of the above-mentioned input/output signal, and processes the current state data of the patient 30 measured by the measuring equipment 40, as the control variable value for various setpoint values of the nurture/treatment equipment 20, on the basis of the neural network driven-type fuzzy inference model established by the expert knowledge and statistical information for the correlation between various state data of the patient 30 and a selective output of the nurture/treatment equipment 20, and also, controls various outputs of the nurture/treatment equipment 20 on the basis of the processed control variable value.
  • the establishment of the neural network driven-type fuzzy inference model and its inventive control method are described in detail, as follows.
  • Fig. 3 is a constructive diagram for an algorithm of the neural network driven-type fuzzy inference model of the artificial intelligence controller 50 shown in Fig. 2.
  • the fuzzy inference model constructs a fuzzy inference rule by using the expert knowledge of a doctor and the statistic distribution of the setpoint value of the nurture/treatment equipment 20 for various kinds of state data of a patient, and the state data of the patient for a learning is divided into groups by the number of rules, and it is decided a coupling coefficient of units in each hierarchy of the neural network NNm through the learning.
  • the membership value OtFl, ⁇ 2, . . . , ⁇ fr) of actual state data (xil, xi2, xin) is decided.
  • Xj n is obtained from the learned neural network model (NN ⁇ NN r ), and is combined with the previously obtained membership value ( Fi, 2, • • ⁇ , flr), thus a setpoint value (y * ) of the nurture/treatment equipment 20 to control the state of the patient 30 is gained.
  • a driving algorithm of the neural network driven-type fuzzy inference model of such artificial intelligence controller 50 is described in detail step by step, as follows.
  • an input variable value concerning an output y is decided. For example, if the blood pressure, the breathing number and the body temperature of the patient becomes close to temperature within the incubator, an input variable Xi of the controller 50 becomes the blood pressure, the breathing number and the body temperature measured and fed back from the patient, and an output of the controller 50 becomes setpoint temperature of the incubator 21.
  • the coupling coefficient of a network is decided from a neural network structure NN m of a membership function of Fig. 3 and the data divided into the groups of the r number through repeated learning. For instances, in case that the input data Xi belongs to a group G 2 , an output of NN m becomes (0, 1, 0, 0, . . . , 0).
  • the coupling coefficient of the network is decided through the repeated learning from a neural network structure NNi. NN2, . . . , NN r of Fig. 3 and a divided data group, namely, (xr 1 , yr 1 ). For example, if the input data is the data belonging to the G2 group, the coupling coefficient is decided through an application to the NN2 structure.
  • a final fifth step data measured in real time and extracted, namely, the state data of the patient, will be generally distributed randomly within a given range.
  • an output value of the controller 50 is decided from the inference network structure decided in the fourth step on the following numerical expression 1.
  • Fig. 5 presents another constructive example for realizing a inference rule of the artificial intelligence controller shown in Fig. 2, and herewith, an output y * is obtained as follows.
  • a fuzzy inference rule is decided by the r number, from the statistical distribution of the input/output data.
  • r (the number of inference rules), and A ⁇ .i and A12.2 represent a membership function of the input variables xi, X 2 , and ii, 1 2 indicates a sort of a membership function for the respective input variables xi and X 2 , and ⁇ ; is an inference value corresponding to each inference rule decided by the learning.
  • an inference value y * of an output is decided on the basis of the following numerical expression 3.
  • the existing open loop incubator system adjusted manually and operated by a setpoint value is constructed as a closed-loop system by realizing an artificial intelligence controller based on a neural network driven-type fuzzy inference model, to whereby control a state of a patient automatically and recover to a normal state. Accordingly, according to an establishment of the inventive closed-loop automatic control system, there are advantages of remarkably lessening an economic waste caused by numerous manpower commitment to the existing incubator system and reducing an error in a diagnosis and treatment to be caused by a frequent monitoring on a state of the patient.

Abstract

In the intelligent type incubator system and a control method thereof, the system is constructed by various kinds of nurture/treatment equipments (20) which has an incubator (21), an artificial respiratory machine (22) and a therapeutic medicine injector (23), and which is provided for a nurture and/or a treatment of a patient; a measuring equipment for measuring and inputting various state data of the patient; and an artificial intelligence controller (50) for processing various kinds of state data of the patient (30) measured by the measuring equipment (40), as a control variable value for various setpoint values of the nurture/treatment equipment (20), on the basis of a neural network driven-type fuzzy inference model based on statistical information for a correlation between the state data of the patient (30) and an output of the nurture/treatment equipment (20) on the state data, and for controlling various outputs of the nurture/treatment equipment (20) on the basis of the processed control variable value.

Description

ARTIFICIAL INTELLIGENCE INCUBATOR SYSTEM AND CONTROL METHOD THEREOF
TECHNICAL FIELD The present invention relates to an intelligent type incubator system and a control method thereof; and more particularly, to an intelligent type incubator system and a control method thereof, in which various kinds of information data concerning checked state of a patient are processed as control data of various kinds of treatment equipments for maintaining a normal state of a corresponding patient, and the outputs of the corresponding treatment equipments are automatically controlled, to thereby make a state monitoring of the patient and its diagnosis and treatment operation, closed-loop automatic controlled.
BACKGROUND ART
In general, an incubator system is provided as an important medical equipment having a function of temporarily nursing a patient under an abnormal state like an immature infant, and of recovering to a normal state of the patient through an appropriate treatment. Fig. 1 is a block diagram of a conventional incubator system, and this system is constructed by an incubator 12 for providing a proper environmental standard such as temperature, humidity and oxygen etc. within a closed space where there is a patient 11; an artificial respiratory machine 13 for controlling the breathing number and breathing pressure of the patient 11 etc. and artificially making the patient breathing; a therapeutic medicine injector 14 for injecting various therapeutic medicines to the patient 11; other equipments (not shown) for a treatment of the patient; and several kinds of measuring equipments 15 for measuring an electrocardiogram (ECG), blood pressure and the breathing number of the patient 11, etc.
In managing the conventional incubator system like Fig. 1, at the present hospitals, numerous manpower such as doctors, nurses and/or managing workers required is needed for all times of 24 hours, to monitor a physical change of state of the patient through several kinds of measuring equipments 15, re-determine an environment of the incubator 12 through a manual input equipment equipped with several kinds of treatment equipments such as the incubator 12, the respiratory machine 13 and the injector 14 etc. according to the monitoring result, or re-adjust a breath of the artificial respiratory machine(13) or an injection amount of the injector 14, or perform a monitoring for an error operation of an entire system containing various kinds of measuring equipments 15.
However, in such conventional incubator system, a current state of the patient is recognized on the basis of measured values gotten by numerous measuring equipments, and a normal or abnormal state thereof is decided, and on the basis of doctor's own treatment experience and medical knowledge, reference inputs of various kinds of treatment equipments such as the incubator, the artificial respiratory machine and the injector etc. should be adjusted manually. Therefore, there are many problems for an economic waste caused by a commitment of much manpower and time, an inconvenience of a manual manipulation and a possibility for a diagnosis error which is caused by referring to a large number of state data for the sake of a decision of the doctor on a treatment method of the patient, etc.
DISCLOSURE OF THE INVENTION Accordingly, the present invention is directed to an intelligent type incubator system and a control method thereof that substantially obviates one or more of the limitations and disadvantages of the related art.
A primary object of the present invention is to provide an intelligent type incubator system and a control method thereof, which is capable of making a state monitoring of a patient and its diagnosis and treatment operation, closed-loop automatic controlled.
In accordance with the present invention for achieving the object, the intelligent type incubator system is constructed by a nurture/treatment equipment which includes an environment controlling part for controlling an air environment containing temperature, humidity and an oxygen amount within a given space according to a corresponding setpoint value, an artificially breathing part for artificially breathing of a lungs-breath life body according to the corresponding setpoint value, and an injection part for injecting various kinds of injection agents onto the life body according to the corresponding setpoint value; a measuring unit for measuring a state of the life body like a medical and biological metabolism etc., the life body being as the object body under a nurture and/or a treatment through the nurture/treatment equipment; and an artificial intelligence controlling unit for processing state data of the life body measured by the measuring unit, as a control variable value for various setpoint values of the nurture/treatment equipment, on the basis of a neural network driven-type fuzzy inference model based on statistical information for a correlation between a state of the life body and an output of the nurture/treatment equipment, and for controlling the output of the nurture/treatment equipment on the basis of the processed control variable value. In order to achieve the above objects in accordance with the present invention, in a controlling method of the intelligent type incubator system having the nurture/treatment equipment and manually varying a setpoint value of the nurture/treatment equipment according to a state of a corresponding life body, the nurture/treatment equipment containing an incubator for controlling an air environment containing temperature, humidity and an oxygen amount within a given space according to the setpoint value, an artificial respiratory machine for artificially breathing of a lungs-breath life body according to the setpoint value, and an injecting apparatus for injecting various kinds of injection agents onto the life body according to the setpoint value; the method includes the steps of: constructing a plural number of fuzzy inference rules on the basis of a statistical distribution of the setpoint value of the nurture/treatment equipment on the state data of the life body, and forming the plural number of fuzzy inference rules as the structure of the neural network; dividing the state data of the life body and its based setpoint value of the nurture/treatment equipment into a plurality of groups corresponding to the plural number of fuzzy inference rules and establishing them, on the basis of expert information on the nurture and/or treatment of the life body, and forming the respective established groups as the neural network structure; individually dividing various kinds of state data of the life body for the learning into groups based on the number of the inference rules; deciding a coupling coefficient of the neural network of the respective groups through a repeated learning based on the neural network driven type fuzzy inference rules and the divided data, and also deciding a coupling coefficient of the neural network of the respective group through repeated learning based on the neural network structure of the respective group and the divided data; deciding a membership value of the state data measured and inputted from the corresponding life body on the basis of the learned neural network driven-type fuzzy inference model; deciding an estimation value of a corresponding setpoint value of the nurture/treatment equipment for the state data measured and inputted, on the basis of the learned neural network model of the respective group; and calculating and getting the corresponding setpoint value of the nurture/treatment equipment on the basis of the decided membership value and estimation value, and controlling an output of a corresponding nurture/treatment equipment on the basis of the calculated setpoint value.
In this present invention, a treatment experience of a corresponding expert doctor and a special knowledge for the life body, on the basis of a recognition for a state of the life body, are established as a knowledge -based system of an artificial intelligence so that system itself has a judgment function, thereby, a corresponding nurture/treatment equipment is appropriately controlled so as to lead the life body to be normally nurtured, treated and restored to health.
BRIEF DESCRIPTION OF THE DRAWINGS
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention.
In the drawings:
Fig. 1 is a block diagram of a conventional incubator system; Fig. 2 indicates a block diagram of an intelligent type incubator system in one embodiment of the present invention;
Fig. 3 depicts a constructive diagram for an algorithm of a neural network driven-type fuzzy inference model of an artificial intelligence controller shown in Fig. 2;
Fig. 4 is a drawing showing, on a two dimensional plane, group data classified under an assumption that there are two inputs to embody a neural network driven-type fuzzy inference model of an artificial intelligence controller shown in Fig. 2; and
Fig. 5 sets forth a drawing showing another constructive example for embodying a inference rule of an artificial intelligence controller shown in Fig. 2.
BEST MODE FOR CARRYING OUT THE INVENTION
Reference will now be made in detail to the preferred embodiments of the present invention, examples of which are illustrated in the accompanying drawings.
In accordance with the present invention, an intelligent type incubator system and a controlling method thereof are described in detail, as follows.
Fig. 2 is a block diagram of an intelligent type incubator system in one embodiment of the present invention, and the intelligent type incubator system includes a nurture/treatment equipment 20 which has an incubator 21 for outputting temperature of air, humidity, oxygen, pressure, and air controlled by each setpoint value for an air flowing quantity, and controlling a quality of air supplied into the inside of a given space, an artificial respiratory machine 22 for outputting oxygen controlled according to each setpoint value for expiration/inspiration pressure, oxygen content, the breathing number per unit time, humidity and temperature, and for artificially breathing of the lungs breathing life body (hereinafter, referred to as "patient") 30, a therapeutic medicine injector 23 for controlling and outputting an actual injection amount according to a setpoint value concerning the injection amount and for injecting various kinds of a therapeutic medicine to the patient 30; a measuring equipment 40 for measuring and inputting various kinds of state data which contains an electrocardiogram (ECG) having irregular pulse and the pulse number representing a state of a medical and biological metabolism etc. of the patient 30, and also contains the breathing number, an oxygen saturation degree, percutaneous oxygen partial pressure, percutaneous dioxide partial pressure, body temperature, weight and ages; and an artificial intelligence controller 50 for processing various kinds of state data of the patient 30 measured by the measuring equipment 40, as a control variable value for various setpoint values of the nurture/treatment equipment 20, on the basis of a neural network driven-type fuzzy inference model based on statistical information for a correlation between the state data of the patient 30 and an output of the nurture/treatment equipment 20 on the state data, and for controlling various outputs of the nurture/treatment equipment 20 on the basis of the processed control variable value.
Referring to Fig. 2, in the present invention it is formed a closed-loop system constructed by the nurture/treatment equipment 20, the patient 30, the measuring equipment 40 and the artificial intelligence controller 50. The artificial intelligence controller 50 receives in feedback a state of the patient 30 through the measuring equipment 40, and processes and outputs the inputted state data of the patient, as the control variable value for various setpoint values, on the basis of the neural network driven-type fuzzy inference model, and sends out it as an input signal of the nurture/treatment equipment 20 containing the incubator 21, the artificial respiratory machine 22 and/or the injector 23, to thereby control the control object like the patient 30 through various kinds of nurture/treatment equipments 20.
In an aspect of a dynamic viewpoint, the artificial controller 50 performs a function of recovering the patient to a stabilized balance state, namely, a normal state, through an appropriate procedure, on the basis of knowledge for a current state and a peculiarity of the patient 30, namely, treatment experience of a doctor and medical knowledge for the human body etc.
It is described contents for the input/output signals constructing the inventive system, as follows. An input of the incubator 21 become a setpoint values for humidity, oxygen partial pressure and temperature of air supplied to the incubator 21 inside, and an air flowing amount, and an output thereof becomes the air controlled by the setpointed value. An input of the artificial respiratory machine 22 becomes a setpoint value for expiration/inspiration pressure, an oxygen content, the breathing number per unit time, humidity and temperature, and an output thereof becomes the oxygen controlled by the setpoint value. An input of the therapeutic medicine injector 23 becomes a setpoint value for the amount of the therapeutic medicine injection, and an output thereof becomes an actual injection amount based on the setpoint value. An input of the patient 30 becomes various kinds of outputs of the nurture/treatment equipment 20 which contains the incubator 21, the respiratory machine 22 and the injector 23, and an output thereof becomes, a state of the patient, the electrocardiogram (ECG) having the irregular pulse and the pulse number, blood pressure, the breathing number, the oxygen saturation degree, the percutaneous oxygen partial pressure, the percutaneous dioxide partial pressure, the body temperature, the weight and age, etc.
The artificial intelligence controller 50 gains information for value of the above-mentioned input/output signal, and processes the current state data of the patient 30 measured by the measuring equipment 40, as the control variable value for various setpoint values of the nurture/treatment equipment 20, on the basis of the neural network driven-type fuzzy inference model established by the expert knowledge and statistical information for the correlation between various state data of the patient 30 and a selective output of the nurture/treatment equipment 20, and also, controls various outputs of the nurture/treatment equipment 20 on the basis of the processed control variable value. The establishment of the neural network driven-type fuzzy inference model and its inventive control method are described in detail, as follows.
Fig. 3 is a constructive diagram for an algorithm of the neural network driven-type fuzzy inference model of the artificial intelligence controller 50 shown in Fig. 2. Describing first a decision method of the membership value, the fuzzy inference model constructs a fuzzy inference rule by using the expert knowledge of a doctor and the statistic distribution of the setpoint value of the nurture/treatment equipment 20 for various kinds of state data of a patient, and the state data of the patient for a learning is divided into groups by the number of rules, and it is decided a coupling coefficient of units in each hierarchy of the neural network NNm through the learning. From such learned neural network model, the membership value OtFl, μ¥2, . . . , μfr) of actual state data (xil, xi2, xin) is decided.
Next, describing a decision for an output value of the artificial intelligence controller 50, the state data of the patient for the learning, and its based output value, namely, the setpoint value of the nurture/treatment equipment 20, are classified according to the rule, and the coupling coefficient of the neural network is decided. An estimated output value
Figure imgf000014_0001
Vei2, . . • , yeir) for an actual state input (xu, Xi2, . . . . , Xjn) is obtained from the learned neural network model (NNι~NNr), and is combined with the previously obtained membership value ( Fi, 2, • • ■ , flr), thus a setpoint value (y *) of the nurture/treatment equipment 20 to control the state of the patient 30 is gained.
A driving algorithm of the neural network driven-type fuzzy inference model of such artificial intelligence controller 50 is described in detail step by step, as follows. In a first step, an input variable value concerning an output y; is decided. For example, if the blood pressure, the breathing number and the body temperature of the patient becomes close to temperature within the incubator, an input variable Xi of the controller 50 becomes the blood pressure, the breathing number and the body temperature measured and fed back from the patient, and an output of the controller 50 becomes setpoint temperature of the incubator 21. Representing in symbols, it becomes the input variable XiT = (xii, Xi2, X13), and xu represents the blood pressure, Xi2 indicates the breathing number, x;3 as the human body temperature, yi as the setpoint temperature, and i represents a time sequence. In a second step, the input/output data for the variable value decided like the first step is divided into groups of the r number based on the number of inferences as shown in Fig. 4, and each group is represented as Gj(j = 1, 2, . . . , r). For instances, data ensured from the treatment experience and medical knowledge of a doctor is classified into groups of the r number according to a statistical distribution of an output. Assuming that there are two inputs for convenience, the classified group data can be represented on the two-dimensional plane as shown in Fig. 4.
In a third step, the coupling coefficient of a network is decided from a neural network structure NNm of a membership function of Fig. 3 and the data divided into the groups of the r number through repeated learning. For instances, in case that the input data Xi belongs to a group G2, an output of NNm becomes (0, 1, 0, 0, . . . , 0).
In a fourth step, the coupling coefficient of the network is decided through the repeated learning from a neural network structure NNi. NN2, . . . , NNr of Fig. 3 and a divided data group, namely, (xr1, yr1). For example, if the input data is the data belonging to the G2 group, the coupling coefficient is decided through an application to the NN2 structure.
In a final fifth step, data measured in real time and extracted, namely, the state data of the patient, will be generally distributed randomly within a given range. At this time, an output value of the controller 50 is decided from the inference network structure decided in the fourth step on the following numerical expression 1.
Figure imgf000016_0001
For example, when an output of the membership function NNm is Fj(xi) = (0.5, 0.2, 0.2), j=l,2,3", and an output of NNj is "yj ei = (3, 1, 0.5)", it becomes "y * = 0.8/0.9 = 2", and a final output value is gained by applying a proper scaling factor determined in a data manipulation, and this value is provided as the output of the controller 50.
Fig. 5 presents another constructive example for realizing a inference rule of the artificial intelligence controller shown in Fig. 2, and herewith, an output y* is obtained as follows.
In a first step, a fuzzy inference rule is decided by the r number, from the statistical distribution of the input/output data. For instances, R1 : IF xi is Aii.i, and X2 is A12.2 THEN y = ωι( ι, X2). Herewith, it is assumed that the input variable is two of xi and X2, and i = 1, 2, . . . , r (the number of inference rules), and Aα.i and A12.2 represent a membership function of the input variables xi, X2, and ii, 12 indicates a sort of a membership function for the respective input variables xi and X2, and ω; is an inference value corresponding to each inference rule decided by the learning.
Ui = Aii.ι(xi) Ai2.2(X2)
Figure imgf000017_0001
In a final third step, an inference value y* of an output is decided on the basis of the following numerical expression 3.
Figure imgf000017_0002
INDUSTRIAL APPLICABILITY
As afore-mentioned, in the inventive intelligent type incubator system and a control method thereof, the existing open loop incubator system adjusted manually and operated by a setpoint value is constructed as a closed-loop system by realizing an artificial intelligence controller based on a neural network driven-type fuzzy inference model, to whereby control a state of a patient automatically and recover to a normal state. Accordingly, according to an establishment of the inventive closed-loop automatic control system, there are advantages of remarkably lessening an economic waste caused by numerous manpower commitment to the existing incubator system and reducing an error in a diagnosis and treatment to be caused by a frequent monitoring on a state of the patient.
It will be apparent to those skilled in the art that various controls and variations can be made in the present invention without deviating from the spirit or scope of the invention. Thus, it is intended that the present invention cover the controls and variations of this invention provided they come within the scope of the appended claims and their equivalents.

Claims

WHAT IS CLAIMED IS:
1. An intelligent type incubator system comprising: a nurture/treatment equipment which includes an environment controlling part for controlling an air environment containing temperature, humidity and an oxygen amount within a given space according to a corresponding setpoint value, an artificially breathing part for artificially breathing of a lungs-breath life body according to the corresponding setpoint value, and an injection part for injecting various kinds of injection agents onto the life body according to the corresponding setpoint value; a measuring unit for measuring a state of the life body as the object body under a nurture and/or a treatment through the nurture/treatment equipment; and an artificial intelligence controlling unit for processing state data of the life body measured by the measuring unit, as a control variable value for various setpoint values of the nurture/treatment equipment, on the basis of a neural network driven-type fuzzy inference model based on statistical information for a correlation between a state of the life body and an output of the nurture/treatment equipment, and for controlling an output of the nurture/treatment equipment on the basis of the processed control variable value.
2. A controlling method of an intelligent type incubator system, said intelligent type incubator system having a nurture/treatment equipment and being for manually varying a setpoint value of the nurture/treatment equipment according to a state of a corresponding life body, to nurture and/or treat the life body, wherein said nurture/treatment equipment contains an incubator for controlling an air environment containing temperature, humidity and an oxygen amount within a given space according to the setpoint value, an artificial respiratory machine for artificially breathing of a lungs-breath life body according to the setpoint value, and an injecting apparatus for injecting various kinds of injection agents onto the life body according to the setpoint value, said method comprising the steps of: constructing a plural number of fuzzy inference rules on the basis of a statistical distribution of the setpoint value of the nurture/treatment equipment on state data of the life body, and forming the plural number of fuzzy inference rules as a neural network structure; dividing the state data of the life body and its based setpoint value of the nurture/treatment equipment into a plurality of groups corresponding to the plural number of fuzzy inference rules, and establishing them, on the basis of expert information on the nurture and/or treatment of the life body, and forming the respective established groups as the neural network structure; separating individually various kinds of state data of the life body for a learning into groups based on the number of the inference rules; deciding a coupling coefficient of the neural network through a repeated learning based on the neural network driven type fuzzy inference rules and the divided data, and also, deciding the coupling coefficient of the neural network of the respective groups through the repeated learning based on the neural network structure of the respective groups and the divided data; deciding a membership value of the state data measured and inputted from a corresponding life body on the basis of the learned neural network driven-type fuzzy inference model; deciding an estimation value of a corresponding setpoint value of the nurture/treatment equipment for the state data measured and inputted, on the basis of the learned neural network model of the respective group,' and calculating the corresponding setpoint value of the nurture/treatment equipment on the basis of the decided membership value and estimation value, and controlling an output of a corresponding nurture/treatment equipment on the basis of the calculated setpoint value.
3. The method of claim 2, wherein said setpomt value of said incubator contains a setpoint value for temperature, humidity, oxygen partial pressure and a flowing amount of air supplied to the incubator inside, said setpoint value of said artificial respiratory machine contains a setpoint value for expiration/inspiration pressure, an oxygen content, the breathing number per unit time, the humidity and the temperature, and said setpoint value of said injecting apparatus contains a setpoint value of an injection amount.
4. The method of claim 2 or 3, wherein said state data of the life body contains an electrocardiogram (ECG) having irregular pulse and the pulse number, the breathing number, an oxygen saturation degree, percutaneous oxygen partial pressure, percutaneous dioxide partial pressure, body temperature, weight and ages.
PCT/KR2001/000767 2000-06-01 2001-05-12 Artificial intelligence incubator system and control method thereof WO2001091691A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR10-2000-0030129A KR100367908B1 (en) 2000-06-01 2000-06-01 Artificial Intelligence Incubator System and control method thereof
KR2000/30129 2000-06-01

Publications (1)

Publication Number Publication Date
WO2001091691A1 true WO2001091691A1 (en) 2001-12-06

Family

ID=19671027

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2001/000767 WO2001091691A1 (en) 2000-06-01 2001-05-12 Artificial intelligence incubator system and control method thereof

Country Status (2)

Country Link
KR (1) KR100367908B1 (en)
WO (1) WO2001091691A1 (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6884211B2 (en) 2002-07-12 2005-04-26 Pontifica Universidad Catolica Del Peru Neonatal artificial bubble
CN103309370A (en) * 2013-06-01 2013-09-18 中南林业科技大学 Incubating room humidity control method based on BP (back-propagation) neural network
CN103631285A (en) * 2013-11-28 2014-03-12 马从国 CAN bus-based barton environment temperature control system
CN104536474A (en) * 2014-11-13 2015-04-22 无锡悟莘科技有限公司 Fuzzy control method of aquaculture oxygenation
WO2021195138A1 (en) * 2020-03-24 2021-09-30 Vyaire Medical, Inc. System and method for assessing conditions of ventilated patients
CN115050454A (en) * 2022-05-26 2022-09-13 深圳先进技术研究院 Method, device, equipment and storage medium for predicting mechanical ventilation offline
CN115551579B (en) * 2020-03-24 2024-04-12 维亚埃尔医疗股份有限公司 System and method for assessing ventilated patient condition

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20140115129A (en) * 2013-03-20 2014-09-30 제이더블유중외메디칼 주식회사 User interface of incubator
KR101592123B1 (en) 2014-11-26 2016-02-04 세명대학교 산학협력단 Oxygen Treatment System having a Oxygen Chamber
KR102347888B1 (en) 2020-05-08 2022-01-05 인제대학교 산학협력단 Shading device for newborn baby incubator

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4509505A (en) * 1982-03-12 1985-04-09 La Calhene Societe Anonyme Isolator for confining and transporting human beings in a sterile atmosphere
US5316542A (en) * 1992-02-14 1994-05-31 Dragerwerk Aktiengesellschaft Coupled control of operating parameters of an incubator
US5446934A (en) * 1993-11-30 1995-09-05 Frazier; Richard K. Baby monitoring apparatus

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4509505A (en) * 1982-03-12 1985-04-09 La Calhene Societe Anonyme Isolator for confining and transporting human beings in a sterile atmosphere
US5316542A (en) * 1992-02-14 1994-05-31 Dragerwerk Aktiengesellschaft Coupled control of operating parameters of an incubator
US5446934A (en) * 1993-11-30 1995-09-05 Frazier; Richard K. Baby monitoring apparatus

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6884211B2 (en) 2002-07-12 2005-04-26 Pontifica Universidad Catolica Del Peru Neonatal artificial bubble
CN103309370A (en) * 2013-06-01 2013-09-18 中南林业科技大学 Incubating room humidity control method based on BP (back-propagation) neural network
CN103631285A (en) * 2013-11-28 2014-03-12 马从国 CAN bus-based barton environment temperature control system
CN104536474A (en) * 2014-11-13 2015-04-22 无锡悟莘科技有限公司 Fuzzy control method of aquaculture oxygenation
WO2021195138A1 (en) * 2020-03-24 2021-09-30 Vyaire Medical, Inc. System and method for assessing conditions of ventilated patients
CN115551579A (en) * 2020-03-24 2022-12-30 维亚埃尔医疗股份有限公司 System and method for assessing ventilated patient condition
CN115551579B (en) * 2020-03-24 2024-04-12 维亚埃尔医疗股份有限公司 System and method for assessing ventilated patient condition
CN115050454A (en) * 2022-05-26 2022-09-13 深圳先进技术研究院 Method, device, equipment and storage medium for predicting mechanical ventilation offline
CN115050454B (en) * 2022-05-26 2023-04-07 深圳先进技术研究院 Method, device, equipment and storage medium for predicting mechanical ventilation offline

Also Published As

Publication number Publication date
KR20010109443A (en) 2001-12-10
KR100367908B1 (en) 2003-01-14

Similar Documents

Publication Publication Date Title
Nemoto et al. Automatic control of pressure support mechanical ventilation using fuzzy logic
CN104958074A (en) Patient monitor with integrated closed loop controller
US6148814A (en) Method and system for patient monitoring and respiratory assistance control through mechanical ventilation by the use of deterministic protocols
DE60020842T2 (en) MONITORING SYSTEM FOR VENTILATOR
WO2001091691A1 (en) Artificial intelligence incubator system and control method thereof
US20230256187A1 (en) Artificial ventilation system and related control method
Kwok et al. Rule-base derivation for intensive care ventilator control using ANFIS
Sun et al. Fuzzy logic assisted control of inspired oxygen in ventilated newborn infants.
Dojat et al. Knowledge-based systems for automatic ventilatory management.
CN114887169A (en) Intelligent control decision method and system for breathing machine
Tzavaras et al. A neuro-fuzzy controller for the estimation of tidal volume and respiration frequency ventilator settings for COPD patients ventilated in control mode
Giabbanelli A novel framework for complex networks and chronic diseases
Jensen et al. Identification of causal relations between haemodynamic variables, auditory evoked potentials and isoflurane by means of fuzzy logic.
CN111643328B (en) Massage force control system and method
Bailey et al. Adaptive and neural network adaptive control of depth of anesthesia during surgery
Zhu et al. Ventilator control based on a fuzzy-neural network approach
WO2019020136A1 (en) Apparatus and method for linking and evaluating information in clinical data in a medical device
JPH11206884A (en) Automatic weaning system for respirator applying fuzzy theory control and record medium recorded with automatic weaning program
Tzavaras et al. Classical approaches and intelligent systems in ventilation management: a survey
US20240071620A1 (en) Monitorization and forecast of pain and illness in patients with quantum computing
Neat et al. A hybrid adaptive control approach for drug delivery systems
Goodman Regulation and control in physiological systms: 1960–1980
Mohamad-Samuri et al. Absolute EIT coupled to a blood gas physiological model for the assessment of lung ventilation in critical care patients
Imamura et al. A Computer Control System of Applying Anesthesia Using Fuzzy Logic for Medical Operation: Categorization of anaesthesiologists thinking represented by artificial intelligence and comparison of each category
CN114141352A (en) Intelligent management system and method for blood purification center

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): CN JP US

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR

121 Ep: the epo has been informed by wipo that ep was designated in this application
122 Ep: pct application non-entry in european phase
NENP Non-entry into the national phase

Ref country code: JP