ELECTROMYOGRAPHIC FORCE MEASURING SYSTEM AND METHOD
CROSS REFERENCE TO RELATED APPLICATIONS
This application claims priority to U.S. Provisional Patent Application No. 60/467,375, filed May 2, 2003, which is herein incorporated by reference in its entirety.
BACKGROUND OF THE INVENTION
1. Field of the Invention The subject invention relates to the general field of electromyographic systems and methods, and more particularly to a wireless, electromyographic system and method for unobtrusively inferring the forces applied to an object by a subject.
2. Background of the Related Art
An electromyograph is a device that records the electrical waves (i.e., biopotential signal) associated with the activity of skeletal and smooth muscles beneath the skin. Electromyographs have been used for years to aid in the diagnosis of neuromuscular disorders, such as Parkinson's disease. The electromyograph, converts the detected electrical activity of the muscle into a perceptible, usually visual record for analysis. Electromyographic (EMG) equipment has heretofore typically taken the form of one or more electrodes that are attached to an individual's skin adjacent the skeletal or smooth muscle of interest, an EMG display monitor, and a set of wires connecting the electrode or electrodes to the monitor.
U.S. Patent No. 5,755,675 to Sihvoven, which is herein incorporated by reference, discloses a method for measuring the function of joints and associated muscles. The method includes the step of measuring, on the one hand, the mobility of
a person in a desired area and, on the other hand, simultaneously by means of electromyography (EMG) measuring the electrical activity of muscles in the same area. Then, the clinician evaluates the abnormality in the mobility and in the function of the muscles of said area, caused especially by pain, by comparing the measured values with reference values compiled in advance.
Scientific research directed to the measuring of the forces imparted by an individual to an object has received considerable attention over the last quarter century. Applications have varied from basic biomechanics research, to medical procedures, such as surgery and newborn delivery, to rehabilitation. Prior attempts at measuring these subject-to-object forces have been cumbersome and obtrusive. All prior attempts limit where the measurements can be taken because of instrumentation and may interfere with the grasping process and therefore, limit their utility in many applications, such as the monitoring of the forces imparted on a newborn by the clinician during delivery or monitoring uterine contractions electromyographically during labor.
In view of the foregoing, a need exists for an improved electromyographic system and method for unobtrusively inferring the forces applied to an object by a subject.
SUMMARY OF THE INVENTION
The subject application is directed to a system and method for unobtrusively measuring/inferring a force imparted by a subject to an object. The system includes, inter alia, an electromyographic (EMG) unit and a data acquisition (DAQ) unit. The EMG unit detects the electrical waves associated with a subject's muscle activity and creates a biopotential signal based thereon. The EMG unit has a sensing portion and a
biopotential signal-processing portion. The sensing portion has at least one surface electrode that is attached to the subject's skin and detects the electrical waves therefrom. In a single electrode embodiment, it is envisioned that the electrode includes the positive the negative and the ground leads. In alternative embodiments, electrodes/sensors can be used which include the positive, the negative and the ground electrodes in a single unit, similar to that disclosed in U.S. Patent No. 6,643,541 to Mok et al. The biopotential signal-processing portion of the EMG unit preferably includes a mechanism for filtering unwanted noise from the biopotential signal. Those skilled in the art would readily appreciate that in alternative embodiments, several surface electrodes can be employed, each detecting electrical waves from different muscles. Such an arrangement would be used in a multiple- channel system similar to that described in U.S. Patent No. 6,643,541 to Mok et al, which is herein incorporated by reference. Still further, it is presently envisioned that each EMG measurement includes a pair of leads and a ground. The DAQ unit receives the filtered or unfiltered biopotential signal from the
EMG unit. The DAQ unit includes a memory component, an analysis component and a display device. The memory component is adapted for receiving and storing data that correlates the subject's biopotential signal to a force imparted by the subject to an object. The analysis component of the DAQ unit includes a mechanism for comparing the biopotential signal to the stored data so as to determine the force imparted by the subject to the object. The inferred force imparted by the subject to the object is then displayed in real-time by the display device. Additionally, it is envisioned that the DAQ unit stores the biopotential signal in memory for subsequent retrieval and analysis.
In a preferred embodiment, the sensing portion of the EMG unit is adapted to be worn by the subject and includes a transmitter with an internal antenna for transmitting the biopotential signal to the biopotential signal-processing portion of the EMG unit. In this embodiment, the biopotential signal-processing portion of the EMG unit includes an antenna, preferably internal, for receiving the biopotential signal transmitted by the sensing portion of the EMG unit.
The disclosed system preferably includes a calibration mechanism that generates the data, which correlates the subject's biopotential signal to a force imparted by the subject to an object. It is envisioned that the calibration mechanism includes a dynamometer or extensometer. In one embodiment, the dynamometer measures a tension or compression force imparted thereon by the subject. In an alternative embodiment, the dynamometer measures a grip or pinch force imparted thereon by the subject.
Preferably, the DAQ unit is adapted for providing instructions to an operator for using and calibrating the system. In alternative embodiments, the calibration process is completely automated.
In applications where it is desired to measure the hand-applied forces (traction and/or gripping) a subject imparts to an object, it is envisioned that the surface of the electrode or electrodes are applied to the subject's forearm and other secondary muscles. The disclosed system may be used in a variety of applications, including, but not limited to, clinical applications. The disclosed system is also suitable for orthopedic procedures, agricultural applications, entertainment purposes, and professional athletic training and analysis. Moreover, the disclosed system can also be adapted for use in entertainment applications, such as virtual reality games. For example, the system can be adapted to provide muscular feedback to the user of the
gamming system so as to enhance perceived realism. In agricultural applications the system can be combined with other robotic devices to provide a more efficient control over farming equipment.
For all of the above-described applications, the placement of the surface electrode(s) is dependent on the relevant muscle groups associated with the physical activity to be monitored in the application. When a particular task requires a muscle contraction, there will be increased motor unit recruitment (i.e., nerve impulses) to that muscle. Once all of the motor units of a single muscle are activated, if more force is required, motor units in secondary muscle groups will be recruited to increase the force needed. In applications that require the recruitment of secondary muscle groups, it would be advantageous to use a plurality of surface electrodes to supply multiple biopotential signals to a multi-channel DAQ unit.
Also disclosed herein is a system for unobtrusively measuring/inferring the forces imparted by a clinician to a newborn during the delivery process. The force inferring system includes an electromyographic (EMG) unit and a data acquisition
(DAQ) unit. The EMG unit detects the electrical waves associated with the activation of at least one of a clinician's forearm muscles and creates a biopotential signal based thereon. The EMG unit has a sensing portion and a biopotential signal-processing portion. The sensing portion has at least one surface electrode which is attached to the clinician's skin and detects the electrical waves from the forearm muscle(s) and, if warranted, other secondary muscles. The biopotential signal processing portion preferably filters unwanted noise from the biopotential signal.
The DAQ unit receives the filtered or unfiltered biopotential signal from the EMG unit and includes a memory component, an analysis component and a display device. The memory component of the DAQ is adapted for receiving and storing data
that correlates the clinician's biopotential signal to a force imparted by the clinician to a newborn. The analysis component of the DAQ unit includes means for comparing the biopotential signal to the stored data so as to determine the force imparted by the clinician to the newborn. Lastly, the display device provides to the clinician a real- time visual indication of the magnitude of the imparted force.
In a preferred embodiment, the sensing portion of the EMG unit is adapted to be worn by the clinician and includes a transmitter with internal antenna for transmitting the biopotential signal to the biopotential signal-processing portion of the EMG unit. In this embodiment, the biopotential signal-processing portion of the EMG unit is remotely located and includes an internal antenna for receiving the biopotential signal transmitted by the sensing portion of the EMG unit.
It is presently preferred that the disclosed force inferring system further includes a mechanism for calibrating the system and generating the data that correlates the clinician's biopotential signal to a force imparted by the clinician to an newborn.
Also disclosed is a method for unobtrusively measuring a force imparted by a subject to an object which includes the steps of selecting at least one muscle of a subject to be monitored; providing a force measuring/inferring system as discussed above which includes EMG and DAQ units; attaching to the subject's skin at least one surface electrode which is operatively associated with the sensing portion of the EMG unit; storing in the memory component of the DAQ unit, data that correlates the subject's biopotential signal for the selected muscle(s) to a force imparted by the subject to an object; comparing with the analysis component of the DAQ unit the filtered or unfiltered biopotential signal to the stored data so as to determine the force
imparted by the subject to the object; and providing to the subject, a real-time visual indication of the magnitude of the imparted force .
The disclosed method preferably also includes the step of attaching the entire sensing portion of the EMG unit to the subject, wherein the EMG unit includes a transmitter with preferably an internal antenna for transmitting the biopotential signal to the biopotential signal-processing portion of the EMG unit.
In an alternative embodiment, the force inferring method includes the step of calibrating the DAQ unit so as to generate the data that correlates the subject's biopotential signal to a force imparted by the subject to an object.
BRIEF DESCRIPTION OF THE DRAWINGS
So that those having ordinary skill in the art to which the present application appertains will more readily understand how to make and use the same, reference may be had to the drawings wherein: Fig. 1 is a perspective view of the wireless electromyographic force measuring system of the present invention;
Fig. 2 is a perspective view of the sensing portion of the EMG unit used in the wireless electromyographic force measuring system illustrated in Fig. 1 ;
Fig. 3 is a representative schematic diagram of the force measuring system of the present invention;
Fig. 4 is a perspective view, which illustrates the calibration unit of the present invention that includes a pull dynamometer;
Fig. 5a is a graphical representation of the data obtained during typical calibration traction tests of four different forearm muscles;
Fig. 5a is a graphical representation of the data obtained during typical calibration gripping tests for two subjects;
Fig. 6a is a graphical representation showing the EMG inferred traction force applied to a newborn during a routine delivery; Fig. 6b is a calibration curve for a clinician;
Fig. 6c is a graphical representation which illustrates the inferred traction forces applied to the newborn during the delivery; and
Fig. 7 is a graphical representation of the EMG data obtained during a vacuum-assisted vaginal delivery of a 3570-gram newborn.
These and other features of the electromyographic force measuring system of the present application will become more readily apparent to those having ordinary skill in the art form the following detailed description of the preferred embodiments.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS Referring now to the drawings, there is illustrated in Fig. 1, a wireless electromyographic (EMG) force measuring system designated generally by reference numeral 100. Electromyographic system 100 unobtrusively measures/infers a force imparted by an individual 19 to an object. It is noted that the embodiment described hereinbelow is a representative system which is used to measure hand-applied forces imparted on an object by an individual. Those skilled in the art would readily appreciate that the inventive features disclosed herein are not limited to applications that require the measurement of hand-applied forces. For example, electromyographic system 100 can easily be adapted for applications that require the measurement of foot-applied forces, or muscle activity, or the measurement of uterine smooth muscle activity or forces during contraction.
Electromyographic system 100 includes a wireless electromyographic (EMG) unit 20 and a data acquisition (DAQ) unit 30. The EMG unit 20 detects the electrical waves associated with the individual's skeletal and smooth muscle activity and creates a biopotential signal based thereon. The EMG unit 20 has a sensing portion 22 (Fig. 2) and a biopotential signal-processing portion 32. The sensing portion 22 includes surface electrodes 24 (Fig. 2) that is attached to the individual's skin and detects electrical waves from the specific muscle of interest. Leads 28 attach the electrodes 24 to the housing 26 of the EMG unit 20.
Referring now to Fig. 3, which provides a representative schematic diagram for the electromyographic system 100. The biopotential signal detected by the electrodes 24 passes through an instrumentation amplifier 40, which takes the weak electric signal (0.1 - 5mV) from the muscle and increases its amplitude to above 0.5V, so that it can be further processed and recorded. The instrumentation amplifier 40, a Burr-Brown INA128P for example, is connected with a 505-ohm variable resistor so that amplification gain can be adjusted for each user/individual. An ideal gain for most clinical applications relating to hand-applied force ranges from 100X to 10000X; however, a higher gain is needed for smooth muscle contraction detection. There are two light-emitting diodes (LED's) 46 connected to comparators 44, which are designed to signal gain saturation. Once amplified, the biopotential output signal passes through a bandpass filter that isolates frequencies between 0.1 kHz and 1 kHz; this improves the signal-to-noise ratio (SNR). The filtered biopotential signal then passes through the wireless transmission unit 50.
The sensing portion 22 of the EMG unit 20 includes a transitter module 50, a Linx TXM-916-ES for example, and a transmitter antenna (e.g., Linx ANT-916SP),
both of which are located in housing 26. The biosignal processing portion 32 of the EMG unit, is remotely located relative to the sensing portion (shown attached to indivdual 19). The biosignal processing portion 32 includes a receiver module 51, such as a Linx RXM-916-ES, and an internal receiver antenna (Linx ANT-916-SP). These components are located in the housing 33 for the biosignal processing unit 32 which is connected to the DAQ system 30.
The wireless EMG unit 20, preferably operates at a frequency of about 916.5 MHz, which is compatible with known EMG frequency ranges and is capable of transmitting analog signals of 20 Hz to 28 kHz up to 20 meters. To avoid interference in clinical applications, the frequency at which the wireless EMG unit 20 operates is higher than the two medical frequency bands prescribed by the Federal Communications Commission (FCC) (i.e., 174-216 MHz and 450-470 MHz). The frequency at which the wireless EMG unit 20 operates is also higher than the television frequency band, 470-862 MHz. Thus, electromyographic system 100 does not interfere with wireless transmissions from other equipment present in a clinical environment, but still operates within a safe range for medical applications.
Both the sensing portion 22 and the biosignal processing portion 32 of the EMG unit 20 are powered by batteries. The approximate total current usage for the sensing portion 22 without LEDs is 17.8 mA. System 100 works reliably within a range of about 7.2V to 9V of battery voltage. Based on this voltage range and assuming that LEDs are not on, the total alkaline battery life for the biosignal processing portion 32 is about 29 hours and sensing portion 22 is about 33 hours.
To ensure preservation of the desired biopotential signal within the appropriate frequency range and to remove noise acquired during wireless transmission, the transmitted biopotential signal passes through a second bandpass
filter 54 with cutoff frequencies between 0.1 kHz and 1 kHz. An operational amplifier boosts the gain of the signal after wireless reception. A half- wave rectifier 58 discards the negative portion of the signal. Finally, the positive biopotential signal is integrated by integrator 60, smoothing the spiked signal into a more easily comprehensible temporal envelope. This signal then arrives at the DAQ unit 30 for recording and data analysis. Those skilled in the art would readily appreciate that if the user of system 100 desires a raw data output from the DAQ unit 30, the signal conditioning elements of this system, such as the filters, rectifiers and integrators, can be eliminated. The DAQ unit 30 receives the filtered biopotential signal from the EMG unit
20. The DAQ unit 30 includes a memory component, an analysis component and a display device. The memory component is adapted for receiving and storing data that correlates the subject's biopotential signal to a force imparted by the subject to an object. As will be discussed hereinbelow, the correlation data is typically provided by a calibration unit.
The analysis component of the DAQ unit includes a mechanism for comparing the filtered or unfiltered biopotential signal to the stored data so as to determine the force imparted by the subject to the object. The inferred force imparted by the subject to the object is then displayed in real-time by the display device. Additionally, the DAQ unit stores the filtered or unfiltered biopotential signal in memory for subsequent retrieval and analysis.
The DAQ system 30 is specifically designed to digitize and process incoming analog signals. The representative DAQ system disclosed herein is implemented in Lab VIEW™, which provides for live-data display and data storage for post- processing. The program also prompts the user for application-related information,
which is linked with output files generated by that specific execution. To increase the portability of system 100, the code can be implemented on a laptop computer 35 (Fig. 1) using a 16-bit PCMCIA card for data acquisition and a terminal block; the hardware is programmed to acquire data at rates up to 1 kHz. Referring now to Fig. 4, system 100 is preferably used in conjunction with a calibration unit 70. Calibration unit 70 allows system 100 to be adapted for a particular user and application while maintaining accuracy in force measurement.
Callibration unit 70 allows system 100 to be calibrated prior to each use so as to account for different muscular responses of different individuals. Although, it has been discovered by the present inventors that, for a prescribed force range, a linear relationship exists between the amount of force imparted to an object by an individual's hand and the biopotential signal, the present inventors have also discovered that the relationship is unique to each individual and is impacted by factors such as fatique and placement of the electrodes.. Additionally, the angle at which the pull force is applied to the object, for example, impacts the accurancy of the data obtained during use of the system.
Calibration unit 70 can be manually operated or automated. An embodiment of the manual design is disclosed in Fig. 4 and consists of, inter alia, a dynamometer 72. A Chatillon DFM 100 is shown, but those skilled in the art would readily appreciate that other model or an extensometer can be used as well. A vice 74 holds the dynamometer 72 at an appropriate angle by clamping force. Dynamometer 72 has a digital level attachment which will display that angle at which the dynamoter is placed. A model skull 76 is attached to the hook 78 on the dynamometer 72 and is pulled by the individual. In this embodiment, a model skull is used to assist in the simulatation of tractions forces applied by a clinitian to a newborn during the
delivery process. Those skilled in the art would appreciate that any object could be used in place of the skull.
An automated embodiment for the calibration unit 70 utilizes two load cells; one compression and one S-tension load cell, (e.g., Omega LC302-50 and Omega LC105-50, respectively). If the system 100 is being calibrated for use in assesing the forces imparted by a clinician to a newborn during the delivery processs, then the model skull 76 is attached directly to the tension load cell which is attached to a 20" aluminum beam, similar to the beam shown in Fig. 4. The tension load cell measures the axial pull force exerted by the individual to the skull 76. The skull 76 is also attached to the 20" beam. The beam is fixed to a vertical bar at one end and is free to pivot around that point. The compression load cell is placed under the rotating beam to measure the vertical component of the pull force. The automated system directly coordinates with the DAQ unit to generate calibration curves by integrating force measurments provided by the load cells with the biosignal provided by the EMG unit.
The DAQ unit 30 guides the user through a manual calibration process, which in this embodiment involves a series of pulls at different pre-set force ranges. Depending on the application, calibration force values ranged from 20 N to HO N. After the calibration process, the program displays the real-time biopotential signal measured by the electomyograph system 100 during actual use and correlates that signal, using the calibration curve, to an inferred force imparted by an individual to an object. The biosignal and force data obtained from the wireless EMG and DAQ units are then stored for post-processing.
A representative method for using system 100 is as follows: 1. Turn on all of the system components.
2. Select at least one muscle of an individual to be monitored.
3. Place a pair of electrodes at the origin and insertion of the selected muscle(s) and also place a ground electrode nearby.
4. Strap the wireless sensing portion of the EMG unit to the individual and connect the leads of the electrodes thereto.
5. Start the DAQ program.
6. Calibrate the system as follows: a. Manual i. Orient the calibration unit to a suitable angle. ii. Pull to 51bs and hold for 5 seconds. Release and rest. iii. Pull to 10 lbs and hold for 5 seconds. Release and rest. b. Automated i. Pull at a desired angle for 20 seconds.
7. Began using the EMG unit and acquiring data. Results of Experimentation
The efficasy of system 100 was tested in a series of laboratory tests and then used in vivo clinical testing after obtaining Institutional Review Board approval. Most hand-gripping and pulling efforts require contraction of the forearm and upper arm muscles. To determine placement of electrodes in the single-channel embodiment of the system, two calibration systems: a JAMAR™ hand-dynamometer for grip force and a modified Chatillon™ dynamometer for traction forces. Five forearm and four upper arm muscles were tested, including the flexor carpi radialis, flexor carpi ulnaris, brachioradiahs, flexor digitorum profundus, palmaris longus, and extensor indicis proprius in the forearm and the triceps brachii, biceps brachii, lateral deltoids and lower trapezius muscles in the upper arm. Fig. 5a shows a typical
calibration for a traction test, where the R2 values ranged between 0.90 and 0.99 for traction tests up to 90 N for the four muscles in the forearm. Fig 5b shows a typical a typical calibration for a grip test up to 300N in two subjects.
Similar results were obtained for other subjects and in tests with the upper arm muscles. For force applications in the range of up to 110 N, the largest gradient and voltage output were obtained from muscle contractions recorded from the palmaris longus. As this force range was the primary range of interest, all further testing was conducted using this muscle.
Seven subjects performed traction tests with the Chatillon™ dynamometer. To mimic the downward traction of an obstetrician delivering a baby, each subject pulled axially when the dynamometer was oriented at 45° beneath the horizontal generating forces up to 110 N in four equal increments. Using this calibration, traction forces generated during clinical procedures (such as delivering a baby) can be inferred from the EMG signal measured by the system. The average R2 values for each of the seven subjects ranged from 0.92 to 0.96. In addition, subjects were asked to repeat the calibration process for four to six cycles. The consistency of the slopes is represented by the percent deviation from the average of the seven subjects, and this value was found to be less than 20%. Temporal variability (up to a factor of 9) was found in a series of tests performed at the same time each day for the same subjects, which was attributable to variations in electrode placement, in the resting or post-exercised state of the muscle being tested and in the gain of the system. Despite the larger variation in slope, linearity was maintained (R2 >0.9). The results dictate that calibration for each laboratory or clinical experiment is best done near the time of actual testing. Another series of calibration studies tested the effect of force application rate
on system reliability. Subjects were also asked to apply a specified force from zero to maximum in a preset time. Different rates of applied traction produced variations in the calibration curves. Specifically, slower pull speeds (0.2 Hz) had a steeper slope and higher R2 value (0.95), as compared to the faster pull speed (1 Hz), which had an R2 value (0.84).
In a pilot clinical study, the system was used in a series of vaginal deliveries during which obstetricians providing written informed consent wore the electrodes on the forearm of their dominant upper extremity. Oral consent for the provider's participation in the study was obtained from the laboring patient. For each delivery, the clinician performed a calibration a short time prior to or after each delivery. Fig. 6a shows a sample time history of EMG data during a routine delivery of a 2650- gram newborn. Fig. 6b provides the corresponding calibration curve for that clinician and Fig. 6c illustrates the inferred traction applied to the newborn during the delivery. In addition to validating the efficiency and reliability of the system in the clinical setting, two additional clinical studies were performed. In one difficult, vacuum-assisted vaginal delivery of a 3570-gram newborn, shown in Figure 7, the upward traction on the vacuum by the clinician's right hand was estimated to be about 60 N, based on her calibration data for that delivery. A force-time history during an obstetric emergency, known as shoulder dystocia, (i.e., when the fetal shoulders become impacted in the maternal pelvis after the fetal head is delivered) was also captured. The force time history shown in Figure 7 is for a severe shoulder dystocia delivery of a 3477-gram newborn. As expected, the intern equipped with the wireless EMG system disclosed herein applied increasing levels of traction (beyond 1 ION, the linear calibration limit) in her attempt to dislodge
the infant's shoulders that had become impacted behind the maternal pubic bone. An attending physician was able to step in after about 90 seconds and perform maneuvers that minimized traction on the fetal head, which then resulted in atraumatic delivery. While the invention has been described with respect to preferred embodiments, those skilled in the art will readily appreciate that various changes and/or modifications can be made to the invention without departing from the spirit or scope of the invention as defined by the appended claims.