Towards classification of patients based on surface EMG data of temporomandibular joint muscles using self-organising maps. (February 2022)

• Record Type: Journal Article
• Authors: Troka, Mateusz; Wojnicz, Wiktoria; Szepietowska, Katarzyna; Podlasiński, Marek; Walerzak, Sebastian; Walerzak, Konrad; Lubowiecka, Izabela
• Is Part Of: Biomedical signal processing and control. Volume 72(2022)Part B
• Publication Date: 2022-02
• Subjects: SOM self-organising maps -- sEMG surface electromyography -- CC cross-correlation coefficient -- TMJ temporomandibular joint -- TMD temporomandibular disorder -- MR masseter right -- ML masseter left -- TR temporalis right -- TL temporalis left -- RMS root mean square -- MVC maximum voluntary contraction; Surface electromyography (sEMG) -- Muscle activation -- Temporomandibular joint (TMJ) -- Jaw motion -- Cross-correlation -- Self-organising maps (SOM); Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2021.103322
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: The activity of muscles is registered during real human jaw motions using sEMG. The EMG signal is analysed in order to identify an instability disorder in TMJ. A methodology based on SOM combined with Cross-correlation is proposed. The approach allows one to classify subjects based on TMJ muscles activity. The methodology can be used to develop a tool to be applied in clinical practice. Abstract: The study considers the need for an effective method of classification of patients with a temporomandibular joint disorder (TMD). The self-organising map method (SOM) was applied to group patients and used together with the cross-correlation approach to interpret the processed (rectified and smoothed by using root mean square (RMS) algorithm) surface electromyography signal (sEMG) obtained from testing the muscles (two temporal muscles and two masseters) of the temporomandibular joint (TMJ) during selected jaw movements. SOM's Unified distance matrix (U-matrix) maps consist of formed clusters that correspond to similarities in input datasets. The results showed that SOM was able to encode muscular responses and create clusters. Information about the level of similarity between the activity of right, left, ipsilateral, and contralateral pairs of muscles was provided by intra cross-correlation coefficient (CC). A low intra CC value may indicate instability of the TMJ function. Information about the level of similarity between the sEMG signals of the same muscles tested in two different patients was provided by inter CC. SOM analysis can be used to interpret the activation of muscular systems, and by comparing the results of different individuals also to identify their TMJ health. Using the cross-correlation approach, one can find similarities in the sEMG data of different patients that can be used to provide clinically useful information. Such findings could be used to improve the clinical diagnosis of TMD and assess muscle activity during treatment.