Keywords

1 Introduction

The action potential of the muscle fibers that constitute a skeletal muscle is generated via chemical action at the neuromuscular junction, and it propagates along the muscle fiber from the neuromuscular junction to the tendons at both ends. The conduction velocity of the action potential is called the muscle fiber conduction velocity. It is derived via surface electromyography (EMG) by using methods such as cross-correlation method. Notably, the waveform obtained from the surface EMG is not the action potential of a single motor unit but the interference potential of multiple motor units. Therefore, if we pay attention to the waveform shape propagating over multiple channels, a new index different from the propagation velocity will be derived.

Kosuge et al. [1] studied a method of quantitatively determining the conduction wave obtained from a multi-channel surface electromyogram (hereinafter, the method is referred to as the m-channel method) and a method of calculating the conditions and conduction velocity. All the propagating waves were extracted from the surface EMG waveforms using the array electrodes by the m-channel method, and the characteristics of each propagating wave, such as propagation velocity, amplitude, wavelength, etc., were examined. From the results, it became possible to consider a more detailed muscle contraction mechanism.

In a previously conducted study by Kawagoe et al. [2], experiments were performed using an array electrode with large ground area and a cross electrode with small area, and the relative frequency distributions of the amplitude and velocity of the conducting waves differed along the same direction. From these results, the aforementioned difference might be attributed to the following three factors: the difference in the electrode area, difference in the measurement time, and difference in the position of the paste. Therefore, a new electrode, as depicted in Fig. 1, was proposed. The feature of the proposed electrode is that pure silver wire with a diameter of 1 mm and length of 10 mm and circular pure silver wire with a diameter of 1 mm are alternately arranged in a line at 8 mm intervals at 5 mm intervals. By using this electrode, three causes can be investigated.

Fig. 1.
figure 1

Proposed electrode

Full size image

In this study, we aim to perform measurements using both the m-channel method and proposed electrode. We also consider the differences in the characteristics of the conducting waves obtained because of the differences in the electrode shape and muscle contraction state.

2 Methods

2.1 Experimental Method

The test subjects comprised six healthy adult males and one female, and the test muscles were the biceps of the dominant arm, which was self-reported by the test subjects. Each subjects held the elbow joint angle at 90° in the sitting position, following which we measured the maximum exertion muscle strength (100% MVC). Subsequently, the same posture was maintained for 30 s with a load of 10% MVC and 40% MVC, respectively, and the surface EMG data were acquired. In addition, considering the muscle fatigue between trials, we took sufficient breaks and performed the measurements multiple times.

The proposed electrode was used to derive the surface EMG data. The sampling frequency in the experiment was 5 kHz. In addition, the amplifier settings were as follows: High Cut of 1 kHz, Low Cut of 5 Hz, and amplification factor of 80 dB (Fig. 2).

Fig. 2.
figure 2

Experimental protocol

Full size image

2.2 Analysis Method

The m-channel method was used to perform the analysis. In the method, one of the two adjacent electrodes, both of which are of the same shape, is defined as the conduction source and the other one as the conduction destination. The section where zero crossing occurs twice from the source is extracted as one waveform. We then determined whether the signal had propagated over multiple channels and then calculate the conduction speed. While performing the conduction judgment, one waveform obtained from the conduction source is used as the conduction-wave candidate, and the conduction wave candidate of the conduction destination existing 10 ms before and after the start point of the conduction wave candidate of the conduction source is extracted. Subsequently, to calculate even the waveforms with different wavelengths, the conduction-wave candidates were resampled using the sampling theorem, and the similarity ratio, amplitude ratio, and wavelength ratio were calculated.

While identifying a conduction wave over multiple channels, thresholds are decided for the similarity ratio, amplitude ratio, and wavelength ratio based on the idea that if the waveform shapes between adjacent channels are similar, then the action potential must have propagated between both the channels. Subsequently, when the conduction-wave candidate is equal to or greater than the threshold, it is determined as a conduction wave. The conduction speed is defined as the time difference Δt between channels, and it is obtained by dividing the distance between channels (5 mm) by Δt. In addition, the conduction-velocity-variation coefficient (hereinafter, referred to as CV) is used as a conduction-determination condition to consider the velocity variable of the waveform for which conduction is determined. In this study, the conduction judgment conditions were as follows: the similarity ratio and the wavelength ratio of 0.9 or higher, the amplitude ratio of 0.7 or higher, and CV was 30% or less. Only the conduction waves over three channels were extracted and used for analysis.

3 Result

3.1 Relative Frequency Distribution and the Number of Conducting Waves

To compare the conduction wave obtained using the analysis method with the relative frequency distribution of the amplitude and conduction velocity for each electrode, the total number of conduction waves was set to 100%, and the ratio of the amplitude to conduction velocity was calculated. The results for subject A are depicted in Fig. 3.

Fig. 3.
figure 3

Relative frequency distribution for Subject A (left column: results for a rectangular electrode; right: results for a circular electrode)

Full size image

The figure shows that the number of conduction waves obtained using the rectangular electrode is greater than those obtained using the circular electrode, for both the loads. A similar distribution was observed when the load was 10% MVC. In addition, when the load was 40% MVC, more conduction waves with the conduction velocity of 5–7.5 m/s were extracted with a circular electrode, compared with a rectangular electrode, and the amplitude of the conduction wave with the conduction velocity of 2.5–7.5 m/s was observed.

3.2 EMG Data

The myoelectric data obtained for Subject A are depicted in Fig. 4. A comparison of the propagation waves obtained simultaneously using both the types of electrodes showed that the values of amplitude, muscle fiber conduction velocity, and wavelength were almost the same. However, comparing the propagation waveforms, a smooth waveform was observed in each channel with the rectangular electrode, whereas a non-smooth waveform was observed with the circular electrode.

Fig. 4.
figure 4

Raw data for Subject A (10% MVC)

Full size image

The EMG data for Subject D obtained with the circular electrode are depicted in Fig. 5. The muscle fiber conduction velocity of the conducting waves obtained from the EMG data was 1 m/s, and this phenomenon was also observed in each of the cases of Subjects E, F, and G.

Fig. 5.
figure 5

Raw data obtained with the circular electrode (10% MVC, Subject D)

Full size image

4 Discussion

4.1 Relative Frequency Distribution and the Number of Conducting Waves

It is considered that the extraction of the propagating wave was easier in the case of the rectangular electrode because more muscle fibers interfered and averaged when compared to the circular electrode. In addition, the muscle activity can be considered in more detail because the circular electrode extracted propagation waves with various propagation speeds compared with the rectangular one.

4.2 EMG Data

In Fig. 4, the interference potential might be attributed a large number of muscle fibers in the case of the rectangular electrode. However, in the case of the circular electrode, the interference potential is attributed to the small number of muscle fibers.

In Fig. 5, the conduction wave obtained from the EMG data cannot consider the muscle activity, and the reason for this is the AC noise (hum).

5 Conclusion

In this study, we examined the differences in the characteristics of the conducting waves obtained based on the difference in the electrode shape and muscle contraction state. Consequently, it became clear that the extraction of the conducting waves was easy because the number of conducting waves obtained with the rectangular electrode was large. However, it also became clear that a more detailed muscle activity could be considered with the circular electrode. In the future, measurements will be continued, and the shape of electrodes will be studied.