CN110720911B - Muscle movement unit extraction method - Google Patents

Abstract

The invention provides a muscle movement unit extraction method, which comprises the steps of firstly collecting multichannel surface electromyogram signals, then respectively extracting muscle movement units from two parts of surface electrode signals of odd-numbered channels and even-numbered channels, finally sorting and issuing results, confirming the movement units found by the two parts at the same time, and confirming the muscle movement units found by the odd-numbered channels or the even-numbered channels through differential signals. The muscle movement unit extraction method provided by the invention can effectively improve the correctness of the result by mutual verification through the respective extraction results of the odd and even part electrodes. The invention is simple to realize and meets the requirement of practical application.

Description

Muscle movement unit extraction method

Technical Field

The invention relates to a muscle movement unit extraction method.

Background

The distribution of the Motor Unit (MU) is the basis of the surface EMG (surface EMG, sEMG), and the detailed information of the Motor Unit obtained by the sEMG is helpful for understanding the working mechanism of the neuromuscular. The method for extracting MU researched by scholars at home and abroad comprises the following steps: a K-means clustering algorithm, a template matching method, an Artificial Neural Network (ANN) algorithm, a real-time linear aliasing blind signal separation algorithm, an independent component analysis and reduction (ICA), a convolution kernel compensation algorithm and the like. The effects of the methods can only be verified through simulation, the effect of extracting MU from the real sEMG is difficult to verify, and in general, how to verify the extraction accuracy is one of the difficulties in the electromyography research field because the extraction of the myomotor unit by using the multichannel surface electromyography is still in the exploration stage.

Disclosure of Invention

In view of the above problems, an object of the present invention is to provide a muscle motion unit extraction method, including the steps of:

collecting multichannel surface electromyographic signals under muscle contraction, filtering the signals, and weakening interference;

step two, extracting surface electromyographic signals of odd number channels, and calculating an inverse matrix of a signal cross-correlation matrix and a distribution sequence xi _d (n)；

Extracting surface electromyographic signals of even number channels, and calculating an inverse matrix of a signal cross-correlation matrix and a distribution sequence xi _e (n)；

Step four, the issuing sequence xi _d (n)，ξ _e (n) comparing to confirm that the same dispensing sequence is a muscle movement unit;

step five, determining the positions of the muscle movement units of the rest issuing sequences by adopting differential signals, determining the muscle movement units with the positions outside the surface electrodes, and if not, not identifying the muscle movement units;

and step six, classifying and sorting all the movement units, removing unreasonable muscle movement units and optimizing the result.

Because the muscle movement unit extracted from the surface electromyogram signal has no gold standard, and whether the extraction result is correct or not is not checked, the muscle movement unit extraction method provided by the invention can effectively improve the correctness of the result, and mutual verification is further carried out through the respective extraction results of the odd-even two parts of electrodes. For muscle movement units that are not found in both parts at the same time, as muscle movement units that may be present outside the area covered by the surface electrodes, only some of them may be detected while the other part is not. Therefore, for this case, the differential signal is further employed to extract the muscle motion unit. The invention is simple to realize and greatly improves the accuracy of muscle movement unit extraction.

Drawings

FIG. 1 is a flow chart of the present invention.

Fig. 2 is a schematic surface electrode numbering diagram according to an embodiment of the invention.

Detailed Description

The present invention is further described in detail below with reference to the accompanying drawings, and those skilled in the art can easily realize the present invention from the disclosure of the present specification.

Fig. 1 shows a flow chart of the present invention. The invention provides a muscle movement unit extraction method, which is characterized by comprising the following steps:

collecting multichannel surface electromyographic signals under muscle contraction, filtering the signals and weakening interference. Because the sEMG signal contains various interference signals, the preprocessing firstly needs to adopt a band-pass filter, reserves a frequency band signal of 10 Hz-500 Hz, and then adopts a notch filter to filter 50Hz power frequency interference.

Step two, extracting surface electromyographic signals of odd number channels, and calculating an inverse matrix of a signal cross-correlation matrix and a distribution sequence xi _d (n) of (a). The parity order lane distribution is shown in fig. 2. Firstly, a cross-correlation matrix and a cross-correlation matrix inverse matrix of odd-number channel signals are calculated, wherein the cross-correlation matrix is expressed as:

C＝E(S(n)S ^T (n))

where n is the sampling instant, S (n) is the multichannel signal at the nth sampling instant, S ^T (n) is the signal transposition of the nth sampling instant, E (-) is the number order expectation, the inverse C of the cross correlation matrix is calculated ^-1 I.e. by

C ^-1 ＝[E(S(n)S ^T (n))] ^-1

The sampling time n is the median of sEMG signal energy, and the energy is calculated according to the following formula:

Δ＝S ^T (n)C ^-1 S(n)

taking the time n corresponding to the energy median ₀ . Obtaining a distribution sequence according to the signal correlation, and calculating a signal distribution sequence xi of an odd number channel _d (n)：

ξ _d (n)＝S ^T (n ₀ )C ^-1 S(n)

Extracting surface electromyogram signals of even number channels, calculating the inverse matrix of the signal cross-correlation matrix and the issuing sequence xi _e And (n) the method is the same as the second step.

Step four, the issuing sequence xi _d (n)，ξ _e (n) comparing to confirm that the same dispensing sequence is a muscle motor unit. And confirming the muscle movement units which can be found in the surface electromyographic signals of the channels with the even serial numbers and the channels with the odd serial numbers, wherein in the comparison process, the error within 3 is allowed to exist at the moment of issuing the sequence.

And step five, determining the positions of the muscle movement units by adopting differential signals for the rest issuing sequences, and determining the muscle movement units with the positions outside the surface electrodes, otherwise, not identifying the muscle movement units. When the muscle movement units are distributed outside the electrodes, the muscle movement units can be found only on even-numbered channels or odd-numbered channels, so that the positions of the muscle movement units are determined through differential signals, and then the muscle movement units are determined.

And step six, classifying and sorting all the movement units, removing unreasonable muscle movement units and optimizing the result. For dispensing rejections with dispensing times less than 15ms, it is possible to supplement for times that are significantly missing.

In summary, the muscle movement unit extraction method provided by the invention can effectively improve the accuracy of the result, and further verify each other through the respective extraction results of the odd and even electrodes. For muscle movement units that may be present outside the area covered by the surface electrodes, only some of them may be detected while the other part is not. Therefore, for this case, the differential signal is further employed to extract the muscle motion unit. The invention is simple to realize and greatly improves the accuracy of muscle movement unit extraction.

The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Those skilled in the art can modify or change the above-described embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which may be made by those skilled in the art without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims (1)

1. A muscle movement unit extraction method is characterized by comprising the following steps:

collecting multichannel surface electromyographic signals under muscle contraction, filtering the signals, and weakening interference;

step two, extracting surface electromyographic signals of odd number channels, and calculating a cross-correlation matrix and a cross-correlation matrix inverse matrix of the odd number channels, wherein the cross-correlation matrix is expressed as:

C＝E(S(n)S ^T (n)) ；

where n is the sampling instant, S (n) is the multichannel signal at the nth sampling instant, S ^T (n) is the signal transposition for the nth sampling instant, E (-) is the order expectation, the inverse C of the cross correlation matrix is calculated ^-1 I.e. by

C- ¹ ＝[E(S(n)S ^T (n))]- ¹ ；

The sampling time n is the median of sEMG signal energy, and the energy is calculated according to the following formula:

Δ＝S ^T (n)C- ¹ S(n)

the time n corresponding to the energy median is taken ₀ Obtaining a distribution sequence according to the signal correlation, and calculating a signal distribution sequence xi of the odd number channel _d (n)：

ξ _d (n)＝S ^T (n ₀ )C ^-1 S(n)；

Step three, extracting the surface electromyographic signals of the channels with even serial numbers, and calculating the inverse matrix and the distribution sequence xi of the cross-correlation matrix of the signals of the channels with even serial numbers according to the algorithm of the step two _e (n)；

Step four, the issuing sequence xi _d (n)，ξ _e (n) comparing, and confirming that the same dispensing sequence is a muscle movement unit;

step five, determining the positions of the muscle movement units of the rest issuing sequences by adopting differential signals, determining the muscle movement units with the positions outside the surface electrodes, and if not, not identifying the muscle movement units;

and step six, classifying and sorting all the movement units, removing unreasonable muscle movement units and optimizing the result.

Images