We introduced an algorithm for automatic identification of true positive (TP) and false positive (FP) spikes in the motor unit spike train, identified by blind source separation (BSS) of high-density surface electromyograms (HDsEMG). The algorithm selects predefined number of spikes, so called witnesses, from identified spike train. The other spikes in the spike train are called test spikes and are classified into TP or FP spikes by our algorithm. For this purpose, the algorithm constructs as many motor unit filters as there are test spikes, using the information from all the witnesses and each individual test spike. Afterwards, it applies each motor unit filter to HDsEMG to get new estimate of MU spike train for each selected test spike and calculates previously introduced Pulse-to-Noise Ratio (PNR) on preselected witnesses in this new spike train. When accumulated over all the test spikes, these PNR values exhibit bimodal distribution with the peak at lower PNR values representing FPs and the peak at higher PNR values representing TPs. Therefore, FPs and TPs can be discriminated by applying computationally efficient segmentation algorithm to corresponding PNR values. We also proposed and mutually compared different witness selection strategies and showed that selection of about 40 spikes with maximal amplitude in the identified spike train minimizes the selection of FPs as witnesses and maximizes the TP vs. FP discrimination power. In our tests on 20 s long experimental HDsEMG signals from biceps brachii muscle the number of FPs decreased from 23.9 ± 4.7 to 4.1 ± 4.4 when the proposed algorithm was used.
COBISS.SI-ID: 22931734
We analyzed the efficiency of motor unit (MU) filter prelearning from high-density surface electromyographic (HDEMG) recordings of voluntary muscle contractions in the identification of the motor unit firing patterns during elicited muscle contractions. Motor unit filters were assessed from 10 s long low level isometric voluntary contractions by gradient-based optimization of three different cost functions and then applied to synthetic HDEMG recordings of elicited muscle contractions with dispersion of motor unit firings ranging from 13 ms to 1 ms. We demonstrate that the number of identified MUs and the precision of MU identification depend significantly on the selected cost function. Regardless the selected cost function and MU synchronization level, the median precision of motor unit identification in elicited contraction is ? 95 % and is comparable to the precision in voluntary contractions. On the other hand, median miss rate increases significantly from ( 1 % to ~ 3 % with the tested level of MU synchronization. The manuscript has been accepted for presentation at EMBC2020 international conference.
COBISS.SI-ID: 00000001
In the last decade, accurate identification of motor unit (MU) firings received a lot of research interest. Different decomposition methods have been developed, each with its advantages and disadvantages. In this study, we evaluated the capability of three different types of neural networks (NNs), namely dense NN, long short-term memory (LSTM) NN and convolutional NN, to identify MU firings from high-density surface electromyograms (HDsEMG). Each type of NN was evaluated on simulated HDsEMG signals with a known MU firing pattern and high variety of MU characteristics. Compared to dense NN, LSTM and convolutional NN yielded significantly higher precision and significantly lower miss rate of MU identification. LSTM NN demonstrated higher sensitivity to noise than convolutional NN.
COBISS.SI-ID: 00000002