Classification of Upper limb phantom movements in transhumeral amputees using electromyographic and kinematic features. (February 2018)

• Record Type: Journal Article
• Authors: Gaudet, Guillaume; Raison, Maxime; Achiche, Sofiane
• Is Part Of: Engineering applications of artificial intelligence. Volume 68(2017:Aug.)
• Page Start: 153
• Page End: 164
• Publication Date: 2018-02
• Subjects: Surface electromyography (sEMG) -- Pattern recognition -- Phantom limb movements -- Transhumeral amputee -- Neural networks -- Classification; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2017.10.017
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: Recent studies have shown the ability of transhumeral amputees to generate surface electromyography (sEMG) patterns associated to distinct phantom limb movements of the hand, wrist and elbow. This ability could improve the control of myoelectric prostheses with multiple degrees of freedom (DoF). However, the main issue of these studies is that these ones record sEMG from sites that cannot always be integrated in a prosthesis socket. This study aims to evaluate the classification accuracy of eight main upper limb phantom movements and a no movement class in transhumeral amputees based on sEMG data recorded exclusively on the residual limb. A sub-objective of this study is to evaluate the impact of kinematic data on the classification accuracy. Five transhumeral amputees participated in this study. Classification accuracy obtained with an artificial neural network ranged between 60.9% and 93.0%. Accuracy decreased if the number of DoF considered in the classification increased, and/or if the phantom movements became more distal. Adding a kinematic feature produced an average increase of 4.8% in accuracy. This study may lead to the development of a new myoelectric control method for multi-DoF prostheses based on phantom movements of the amputee and kinematic data of the prosthesis.