Evaluation of a signal-adapted filter for processing of periodic electromyography signals in horses walking on a treadmill.

The research article is about how the use of a signal-adapted filter can improve the analysis of periodic electromyography (EMG) signals in horses walking on a treadmill.

Objective and Methodology

The study wanted to test the effectiveness of an adaptive-filter method, which is a filter with transfer functions matched to the characteristics of the EMG signals, in processing and analysing periodic electromyography signals:

• The research team conducted the experiment on 15 healthy adult horses without any signs of back pain.
• They installed surface electrodes bilaterally to record the EMG signals from the left and right longissimus dorsi muscles, middle gluteal muscles, and the triceps brachii muscle of the horses as they walked on a treadmill.
• Along with this, they placed reflective markers on the hoof of the left hind limb of the horses for simultaneous kinematic measurement of motion cycles.

Procedure

• The recorded EMG signal’s absolute value was convoluted using a filter signal of equivalent length to three motion cycles.
• The team calculated the signal-to-noise ratio (SNR) from the autocorrelation function.
• This SNR was then compared with the SNR of the EMG signal that was unfiltered and one that was low-pass filtered.

Results

The results of the study were promising:

• The implementation of the signal-adapted filter showed a significant increase in the SNR by 7.3 dB if compared to the low-pass filter and an increase of 11.1 dB when compared to the unfiltered EMG signal.

Conclusion

Based on the finding of this study, it can be concluded:

• The signal-adapted filter method was beneficial in eliminating signal parts that were not correlated to periodic motion, leading to a much cleaner analysis of the EMG signals.
• This makes the adapted filter method very useful in increasing the applicability of periodic EMG signals as a clinical tool, especially, in this study, for monitoring and analyzing the gait of horses.