A chaotic neural network model for biceps muscle based on Rossler stimulation equation and bifurcation diagram. (September 2022)

• Record Type: Journal Article
• Authors: khodadadi, Vahid; Nowshiravan Rahatabad, Fereidoun; Sheikhani, Ali; Jafarnia Dabanloo, Nader
• Is Part Of: Biomedical signal processing and control. Volume 78(2022)
• Publication Date: 2022-09
• Subjects: Chaotic neural network model -- Poincare section method -- Electromyogram -- Rossler dynamic -- Biceps muscle -- Bifurcation diagram; 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.2022.103852
• Languages: English
• ISSNs: 1746-8094

Abstract

Abstract: In this paper, a chaotic neural network was presented based on the muscle electromyogram (EMG) signal. Networks and biological processes are complex, and nonlinear systems composed of a large number of interconnected elements. A full understanding of the behavior of these systems leads to an understanding of the behavior of diseases and the use of their rehabilitation equipment. In this study, a chaotic neural network model based on the EMG signal of biceps muscle and bifurcation diagram in resting-state were presented. Thus, the biceps muscle was stimulated at a rest state by a signal emanating from Rossler nonlinear dynamics. Then, by synchronizing it, the EMG signal of the biceps muscle was recorded and entered the phase space. Then, using Poincare section method, the bifurcation diagram and the resulting map were obtained. Chaotic map was then trained to a feed-forward neural network with four hidden layers and presented as a chaotic model of EMG signal of biceps muscle. The results revealed that the proposed model has a high generalizability compared to training chaotic maps. Furthermore, biological signals entered the chaos through the period doubling passage.