Biomedical Signal Processing and Control 61 (2020) 102045 Contents lists available at ScienceDirect Biomedical Signal Processing and Control jo ur nal homepage: www.elsevier.com/locate/bspc Revealing stroke survivor gait deficits during rehabilitation using ensemble empirical mode decomposition of surface electromyography signals Ming-Gui Tan a , Jee-Hou Ho a,∗ , Hui-Ting Goh b , Hoon Kiat Ng a , Lydia Abdul Latif c , Mazlina Mazlan c a Department of Mechanical, Materials and Manufacturing Engineering, Faculty of Science and Engineering, University of Nottingham Malaysia, Malaysia b School of Physical Therapy, Texas Woman’s University, United States c Department of Rehabilitation Medicine, Faculty of Medicine, University of Malaya, Malaysia a r t i c l e i n f o Article history: Received 21 September 2019 Received in revised form 14 April 2020 Accepted 5 June 2020 Available online 1 July 2020 Keywords: Stroke Ensemble empirical mode decomposition Surface electromyography Gait a b s t r a c t In this paper, an attempt is made to explain the stroke survivor gait recovery mechanism using Ensemble Empirical Mode Decomposition (EEMD) of surface electromyography signals (sEMG). The existing gait functionality indices could describe the gait recovery through the improvement of gait parameters such as stride length, heel clearance, stance and swing time etc. However these indices reveal little information related to the kinesiology status. To address this knowledge gap, we propose an approach to decompose the sEMG signals acquired during the rehabilitation treatment using EEMD so as to reveal gait deficits in the perspective of motor unit recruitment and its firing patterns. 15 stroke survivors were recruited and their sEMG signals acquired from Gastrocnemius Lateral (GL) and Tibialis Anterior (TA) muscles were further decomposed into different intrinsic mode functions (IMF) using EEMD. Each IMF contains superimposed motor unit action potential (MUAP) with its specific frequency range. The evolvement of IMFs over three recovery stages was observed. Results show that foot drop can be caused by lack of high frequency IMF components in TA during the swing phase. Besides that, spasticity was observed in all IMF components from GL muscle that leads to counteract with TA muscle. Lack of high frequency IMF components in GL during the stance phase would increase the stance time. Co-activation from TA during the stance phase could contribute to this effect as well. In conclusion, the proposed approach could reveal additional information at kinesiology level to explain how well a stroke survivor recovers. © 2020 Elsevier Ltd. All rights reserved. 1. Introduction According to American Heart Association (AHA) 2011 report [1], there were 795,000 people suffered from new or recurrent stroke. Gait rehabilitation is therefore an important strategy to regain stroke survivors walking ability. Kinesiology information such as electromyography (EMG) is a technique to measure muscle activity noninvasively. EMG reveals electrical activities of skeletal muscle during movement [2]. Human central nervous system (CNS) has two ways to control the force activated by muscles. There are spa- tial and temporal recruitment of motor units (MU). Temporal MU recruitment increases the activation frequency of muscle fiber con- traction whereas spatial MU recruitment increases the number of ∗ Corresponding author: E-mail address: JeeHou.Ho@nottingham.edu.my (J.-H. Ho). MU recruited. These stimulated MUs send pulses and they can be recorded by electrode and it is known as motor unit action poten- tial (MUAP) [2]. Studying the shapes, amplitudes and firing rate of MUAP could provide important information of stroke survivors [3]. Therefore, EMG decomposition has attracted research spot- light in the past decades [4–6]. Many reported works focused on decomposing needle EMG and high-density arrays EMG into MUAP. For surface EMG, it is very challenging to decompose it accurately due to its low spatial selectivity [3]. All MUAPs from surface EMG tend to look alike and they overlap with each other. Therefore, it is very challenging to extract useful MUAP information from a stroke patient using existing sEMG decomposition methods. In 1998, Huang et al. [7] proposed a new technique for ana- lyzing non-linear and non-stationary data. The key part of the method is the Empirical Mode Decomposition (EMD). EMD could decompose a signal into finite and small number of intrinsic mode functions (IMF). These IMFs yield instantaneous frequencies as https://doi.org/10.1016/j.bspc.2020.102045 1746-8094/© 2020 Elsevier Ltd. All rights reserved.