GB2145228A - Electromyogram - Google Patents

Abstract

An electromyogram has a pair of electrodes 12, 14 which are secured to the skin surface over a muscle to be investigated, to measure the small voltages produced at the skin surface by muscle activity. Because there is no invasion of the body, the instrument is quick and easy to use. The patient's body is earthed by a further electrode 10 to reduce mains induced voltages in the body and circuitry is provided to take account of residual mains induced voltages. Electrodes 12 and 14 are connected to transistors T1 and T2 arranged as a differential pair feeding a differential amplifier A2. After further processing by lum filters based on transistors T5 and T6, a variable gain amplifier IC2, and a bandgrass filter based on amplifier A3, the measurement signal may be displayed on an oscilloscope or recorded. The portion of the signal above a threshold level is applied to a meter M1 and to an oscillator IC4 which feeds pulses to a loudspeaker LS1 at a rate proportioned to the measured signal. <IMAGE>

Description

SPECIFICATION Electromyogram An electromyogram (EMG) is a sensitive instrument that measures the current waveforms generated by contractions in muscles. An EMG is used to measure the extent of damage or weakness to muscle which may have been caused by injury, disease or damage to the spinal cord.

Known EMG's measure these waveforms using needle-type electrodes inserted into the muscle tissue.

This type of measurement is invasive and therefore distressing to the patient. This measurement can only be performed by specially trained personnel, to whom the patient is referred to by a specialist. Conventional EMG's also tend to be expensive to purchase, large in size and costly to run.

It is for these reasons that this type of measurement is not carried out as often as is desirable.

According to the present invention, there is provided an electromyogram which has electrodes adapted to be secured on the surface of the skin in the vicinity of a muscle in order to sense voltages produced by muscle activity.

Three electrodes are preferably provided of which two serve to sense voltages produced by muscle activity, whilst the third serves to connect the patient to ground.

The two voltage sensing electrodes are preferably capacitatively coupled to the base electrodes of respective transistors driven by a common constant current source; the source and coupling circuits being such as to maximise common mode rejection.

The electromyogram may have means for producing an audible output signal having a frequency dependent on the sensed muscle voltage.

The electrical circuit of the electromyogram may include an amplifier with a hum filter on either side of the amplifier.

The invention will now be further described, by way of example, with reference to the accompanying drawing which is a circuit diagram of an electromyogram according to the invention.

The function of the electromyogram is to measure minute electrical waveforms produced by contractions of the muscles; this signal is present at the skin surface at levels around luV. Also present are 50 Hz Mains induced signals at levels exceeding 7V peak to peak.

The EMG has three input electrodes 10, 12 and 14. Electrode 10 is an earth which is connected between any suitable part of the patient's body and the ground of the EMG. This has the effect of dropping the Mains induced signals to approximately 1V peak to peak.

Input signals are obtained by attaching the remaining two electrodes 12 and 14 over the muscle of interest.

Signals from these electrode drive transistors T1 and T2, which are arranged as a differential pair. Emitter and collector currents for T1 and T2 are derived from a precision constant current source made by T3, T4 and Al. Transistors T1 and T2 share the current supplied by the constant current source; e.g. if T1 is driven harder than T2, then while the collector current of T2 increases there will be a corresponding decrease in the collector current of T2. The collectors of T1 and T2 are connected to the input of A2.

Operational amplifiers Al to A4 amplify current differences applied to their inputs and are part of a single DIL package.

To ensure a high common mode rejection ratio (CMRR), the quiescent collector currents of T1 and T2 must be held close to a fixed reference. Hence the precision constant current source. To derive this constant current source for T1 and T2 the bases of T3 and T4 are driven by the output of Al. The non-inverting input of Al is driven by a fixed voltage derived from a voltage divider (R12 and R13) from the positive supply rail. C11 is a bypass capacitor to prevent supply rail variations modulating this reference voltage. The inverting output of Al is coupled to the emitter of T4 placing this transistor in the feedback loop of Al.The operational amplifier Al will attempt to maintain the current flowing through its inputs at a constant level thus maintaining the base-emitter current through T4 and therefore the collector current at nominally 100 mA.

Assuming T3 has a similar gain to that of T4, its collector current will be the same. A 1K preset P4 allows adjustment of the two collector currents to offset any slight differences in gain.

The input transistors T1 and T2 and the constant current drivers T3 and T4 have the same characteristics within each pair. Due to inherent differences between two transistors even of the same type it was found physically necessary to thermally couple T1 to T2 and T3 to T4 to bring their characteristics as close as possible to each other.

The input stage gain is determined by the value of the resistance between the emitters of T1 and T2. The lower this resistance the higher the gain. A switch S1 simply connects a 100R resistor in parallel with R3 thus increasing the gain. Capacitors C7 and C8 ensure high frequency stability through bypassing the bases of T1 and 12 at frequencies above the range of interest (100-500Hz). To ensure good CMRR the bases of T1 and T2 each receive the same level of input signal. As the input is AC coupled, the characteristics of the input capacitors must closely match each other. Normally if stranded polyester 10% types are used the slightly different impedances of each would limit the CMRR.The solution adopted was to use several capacitors in parallel so that the slight capacitance variations and corresponding impedance variations average out. These six capacitors C1 to C6 are therefore of the same type. Supply rail decoupling for the input stages is provided by R25, R26, C22 and C23.

Two 50 Hz hum filters are employed. One immediately following the differential input stage, the other between the variable gain stage and the bandpass filter. Both filters consist of a twin-T-network. In the first hum filter a transistor T5 is connected as an emitter follower. The twin-T components are connected to provide feedback at 50Hz in order to obtain a high circuit Q and thus good rejection at 50 Hz. The value of the resistance formed by R16 and R17 (paralleled) must be as close as possible to half the value of R14 and 15. As the latter pair are 47K resistors, the best way to obtain a value of half is to connect two 47K resistors in parallel. Similarly for the second hum filter, T6 is the active component and the filter consists of C24, C25, C26, C27 and R27, R28, R29 and R30.Resistors R28 and R29 form a resistance half that of R27 and R30 to provide good rejection at the notch frequency. These stages provide a total of 20 dB rejection at 50Hz.

Following the first hum filter is a variable gain stage employing an operational amplifier (IC2). Gain variation is provided by P1, a 1 M potentiometer connected in the feedback path of the IC2. P1 is a front panel control. Gain is variable between 10 and 1000. To avoid problems arising from large offset voltages and unstable gain settings, the feedback for 1C2 has been arranged via a voltage divider consisting of R23 and R24, the gain potentiometer P1 connected between the output of 1C2 and these two resistors.

The gain of this circuit is given by the equation:

Signal levels at the output of the variable gain stage are around iv. Any hum exceeding this level could easily cause clipping in succeeding stages and the purpose of the second hum filter is to prevent this.

The bandpass filter employs one operational amplifier A3. A filter network consisting of R34, R35, R36 and R37 and C29 and C30 is connected around a feedback path between the operational amplifier output and its inverting input. This provides a bandpass extending from 100 Hz to 500 Hz which encompasses the range of interest for the muscle fibre signals. At midband ( 250 he ) the gain of this stage is approximately 4.

A monitor output is provided from the output of A3 so that the filtered muscle activity waveforms may be viewed on an oscilosciope or recorded on a penchart if desired. This consists of a precision rectifier that passes only than a preset DC voltage determined by the threshold potentiometer P2 on the front panel. The output of the bandpass filter is mixed with a DC voltage derived from the positive supply rail via P2. The resultant signal, the AC muscle activity signal superimposed on a DC voltage is then applied to the input of the precision rectifier. This involves A4, D1 D2 and R39, R40, R41 and R42. The latter two resistors convert the current differencing input of the LM3900 into a cnventional operational amplifier.Positive going signals of less than 0.6 V above the voltage present on the junction of R39 and R40 will be amplified by the full open gain of A4. The output of this stage increase rapidly until D2 conducts. The stage then has only unity gain (x 1) determined by the ratio of R42 and R39. Output from the precision rectifier is taken from the cathode of D2 and consists of the amplified positive going part of the muscle fibre signals that are above the voltage set by the threshold potentiometer P1. A diode D1 ensures that the gain of this stage remains at unity for the negative going portions of the muscle fibre signals from the output of A3.

An operational amplifier (IC3) with an emitter follower stage (T7) connected in the negative feedback path drives a meter Mi. The threshold stage output is coupled to the input of 1C3 (a 741) via a 100nF capacitor C34.

Resistor R47 limits the base current of T7 to a safe value as the 741 will provide much more current than the transistor will stand. A signal from the output of the threshold circuit is amplified by 1C3 causing T7 to turn on thus charging C38 and increasing the meter reading. The circuit will respond quickly to increasing input signals showing a corresponding increase in the meter reading. As the signal decreases with decreasing muscle activity, so the meter reading decays at a rate depending on the capacitance between the emitter of T7 and ground. This provides for some integration of the signal level variations. The integrate switch S2 connects a 470 u capacitor C37 in parallel with C38 (47u). With this in circuit (integrate switch ON) the meter takes approximately 4 seconds to drop from full scale to zero.

An audio output consisting of a series of pulses whose repetition rate increases with increasing muscle activity is also provided. The emitter of T7 is coupled to IC4 (a 555 timer) via R44. Current through this resistor charges C33 until the voltage on pin 6 of IC4 reaches two-thirds of the voltage on pins 4 and 8. At this point pin 7 of the 555 previously appearing as an open circuit will conduct, discharging C33 via R45. Once the voltage on pin 2 drops to one third of that on pins 4 and 8, pin 7 returns to an open circuit condition allowing C33 to charge again. In this manner the 555 oscillates, providing pulses on pin 3 to a loudspeaker LS1 via P3 which serves as a volume control. As the voltage at the emitter of T7 varies according to the variation in muscle activity signals, the rate at which C33 charges will vary. This varies the pulse repetition rate of the 555 oscillator in sympathy with the variations in muscle activity.

The instrument is driven by batteries B1 and B2.

The unit also includes a power ON/OFF switch S4, and a battery monitoring facility via S3 and R49, utilising the meter M1.

The electrodes can be a standard type. For example, the Red Dot special procedure type sold by the 3M's company which are self-adhesive silver/silver chloride skin surface types are suitable.

Claims (6)

1. An electromyogram which has electrodes adapted to be secured on the surface of the skin in the vicinity of a muscle in order to sense voltages produced by muscle activity.

2. An electromyogram as claimed in Claim 1, wherein three electrodes are provided of which two serve to sense voltages produced by muscle activity, whilst the third serves to connect the patient to ground.

3. An electromyogram as claimed in Claim 1 or Claim 2, wherein the two voltage sensing electrodes are capacitatively coupled to the base electrodes of respective transistors driven by a common constant current source; the source and coupling circuits being such as to maximise common mode rejection.

4. An electromyogram as claimed in any preceding claim and having means for producing an audible output signal which has a frequency dependent on the sensed muscle voltage.

5. An electromyogram as claimed in any preceding claim, wherein the electrical circuit includes an amplifier with a hum filter on either side of the amplifier.

6. An electromyogram substantially as herein described with reference to the accompanying drawing.