SMUAP Decomposition Method Considering Estimated Distance

from Surface Electrodes to Motor Unit during Voluntary Isovelocity

Elbow Flexion

Jun Akazawa

and Ryuhei Okuno

School of Health Sciences and Medical Care, Meiji University of Integrative Medicine, Kyoto, Japan

Department of Electrical and Electronics Engineering, Setsunan University, Osaka, Japan

Keywords: Electromyogram, Motor Unit, Decomposition, Elbow Flexion.

Abstract: The purpose of this study was to decompose Surface Motor Unit Acton Potential (SMUAP) clearly even in

case that the estimated distance from surface electrodes to motor units changing during voluntary

isovelocity elbow flexion. We developed decomposition algorithm focusing on SMUAP Profile, and

investigated motor unit (MU) recruitment and firing rate in biceps short head muscle during isovelocity

elbow flexion using our developed method. As a result, we concerned that calculated MUs firing rates were

almost same as the results in the previous studies, and the estimated MU’s territory was changed with elbow

flexion. It was shown that the developed algorithm was useful for decomposing SMUAP when the

estimated distance from surface electrodes to MU changing during voluntary isovelocity elbow flexion.

1 INTRODUCTION

In the field of spots science and rehabilitation, they

had been interested in the methode to analyze the

motor units (MUs) recruitment and firing rates

during voluntary isovelocity elbow flexion

(hereinafter referred to as elbow flexion).

During elbow flexion, the distance from surface

electrodes to muscle fibres of single MU changes

slightly. It was sometimes difficult to decompose

Surface Motor Unit Action Potential (SMUAP)

clearly. So it was important to estimate the distance

between surface electrodes and the muscle fibres. In

the previous study, at first we had devised the

system to decompose SMUAP for short period with

template maching during the elbow flexion, and

investigated firing rate of MU in biceps short head

muscle (Akazawa, 2013), though, we didn’t

confirmed that the distance form surface electoredes

to motor units changing during the elbow flexion.

In this study, in order to decompose SMAUP

clearly even in case that the estimated distance from

surface electrodes to motor units changing during

the elbow flexion, we developed the decomposition

algorithm, in which we used not only template

matching method for short period but also we added

the algorithm conditions of SMUAP profile

(Akazawa, 2005).

2 MEATHOS

Figure 1: Schematic overview of the experimental setup.

Schematic overview of the experimental setup is

shown at Fig. 1. The subject was instructed to flex

the elbow joint smoothly to approximately 75 degree

at constant angular velocity (5 degree/s) against the

weight (2.0 kg). The experiments were performed

with one healthy subject who gave informed consent

and the investigation was approved by the local

Ethics Committee (Meiji University of Integrative

Medicine).

The eight-channel surface electrode was used.

The SEMG signal was amplified with the gain 80 dB

255

Akazawa J. and Okuno R..

SMUAP Decomposition Method Considering Estimated Distance from Surface Electrodes to Motor Unit during Voluntary Isovelocity Elbow Flexion.

DOI: 10.5220/0005248102550257

In Proceedings of the International Conference on Bio-inspired Systems and Signal Processing (BIOSIGNALS-2015), pages 255-257

ISBN: 978-989-758-069-7

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

and the band pass filter was set at 43 Hz - 2.8k Hz.

In the algorithm, firstly the measured SEMG was

extracted for 3 seconds by every 1.5 seconds

(Akazawa, 2013).

In this study, the developed algorithm condition

of SMUAP profile was as follows: The SMUAP

channel which outputs the max value won’t change

for less than two channels during the short period of

3.0 sec. We estimated location and shape of MU

(Akazawa 2005). The central coordinate was set at

the center between CH4 and CH5 of surface

electrodes.

3 RESULTS

Fig. 2 shows a typical decomposed result of MUAP

Train, and firings of nine motor units, MU1 to MU9

were identified. The results of calculated MU’s

firing rates are in agreement with the generally

accepted behavior of MUs firing rates.

MU1 which was firing for the longest time, so

template of MU1 which was made with calculating

the arithmetic mean (average) of the MU1 signals.

Figure 2: Solid lines: elbow joint angle (average velocity

was approximately 5 degree/s). Identified firings of nine

MUs are shown. Each bar represents one firing and firings

of the same MUs are aligned horizontally. The average

frequency of Motor Unit 1(MU1) was 22.52 Hz, and the

standard deviation was 11.35 Hz. The average frequency

of MU2 was 20.86 Hz, and the standard deviation was

10.90 Hz.

We made four templates of MU1 during time

period of 3.0 sec so that we could confirm the shape

change of SMUAP with changing the distance from

surface electrodes to the motor units. The results are

shown at Fig. 3. We can know that the highest

voltage value of SMUAP was changing from CH8 to

CH5. As increasing the subject’s elbow angle, the

amplitude of SMUAP was increasing gradually.

As shown at Fig. 3 (d), the channel which output

the highest value is CH5. So, the condition that the

SMUAP which was made by MU1 is that the

channel output the highest value is CH4 to 6.

Figure 3: Template of MU1 The short period of Fig, (a)

was 3.5 to 6.5 sec, (b) was 8.0 to 11.0 sec, (c) was 14.0 to

17.0 sec, (d) was 21.5 to 24.5 sec. At fig, (a) the estimated

territory’s radius (R) was 5.0 mm, depth (D) was 2.0 mm,

and horizontal shift (Y) from the middle of the electrodes

was 10.16 mm, fig, (b) R was 5.0 mm, D was 1.9 mm, and

Y was 7.62 mm, fig, (c) R was 5.0 mm, D was 2.20 mm,

and Y was 5.08 mm, fig, (d) R was 0.25 mm, D was 2. 0

mm, and Y was 2.54 mm,

4 CONCLUSIONS

In this study, in order to decompose SMUAP clearly

even in case that the estimated distance from surface

electrodes to motor units changing during elbow

flexion, we developed the decomposition algorithm,

in which we used not only template matching

method for short period but also we added the

algorithm conditions of SMUAP profile. The results

of calculated MU’s firing rates are in agreement

with the generally accepted behavior of MUs firing

rates.

It was shown that the proposed method was

useful for decomposing MUAPs during elbow

flexion.

ACKNOWLEDGEMENTS

This work was supported by JSPS KAKENHI Grant

Numbers 26350678, 24500667.

REFERENCES

Akazawa, J., Okuno, R., 2013. A Method for Quantitative

SEMG Decomposition and MUAP Classification

during Voluntary Isovelocity Elbow Flexion. 35th

Annual International Conference of the IEEE EMBS,

pp.6776-6779, Osaka, Japan.

Akazawa, J., Sato, T., Minato, K., Yoshida, M., 2005.

“Method of estimating location and territory of motor

BIOSIGNALS2015-InternationalConferenceonBio-inspiredSystemsandSignalProcessing

256

units in human first dorsal interosseous muscle with

multi-channel surface electromyograms,” JSMBE, vol.

43(4), pp. 595-604 (in Japanese).

SMUAPDecompositionMethodConsideringEstimatedDistancefromSurfaceElectrodestoMotorUnitduringVoluntary

IsovelocityElbowFlexion

257