Analysis of muscle activity during gait cycle using fuzzy rule-based reasoning Huiying Yu a,1 , Murad Alaqtash a,1 , Eric Spier b , T. Sarkodie-Gyan a, * a Department of Electrical and Computer Engineering, University of Texas at El Paso, El Paso, TX 79968, USA b Mentis Neuro Rehabilitation, 7230 Gateway East Blvd. Suite E, El Paso, TX 79915, USA article info Article history: Received 14 October 2009 Received in revised form 26 April 2010 Accepted 27 April 2010 Available online 20 May 2010 Keywords: Neuromechanics Muscle activation Human walking pattern Motion capture system Movement stability Control Gait pathology detection abstract The purpose of this study was to determine the patterns of muscle activation as outcome measures of the ground reaction forces during normal walking tasks using optical motion analysis capture system, instrumented treadmill and electromyography (EMG), respec- tively. The recognition of the muscle patterns during gait dynamics offers insight into the control of skeletal position, joint stiffness, vibrations of the soft tissue packages, stabil- ity during ground contact, and propulsion for the movement task. Sixteen able-bodied par- ticipants were recruited to walk on a dual-belt instrumented treadmill with embedded force plates. A fuzzy rule-based reasoning algorithm for recognizing the activation patterns of the lower extremity muscles during normal walking maneuvers within the seven gait phases was developed. The resulting recognition will enable the determination of altera- tions in the locomotor control system, contribute to suggest symptoms of a neurological disease, disease severity, and also indications of improvements in response to therapeutic interventions. Published by Elsevier Ltd. 1. Introduction Hemiplegic stroke, paraparesis from spinal cord inju- ries, and other upper motor neuron syndromes frequently cause serious mobility-related impairments. The rehabili- tation process is labor intensive. For many disorders, the most effective types of therapeutic intervention vary and are difficult to determine. Patient evaluation is often sub- jective as is the assessment of treatment. Gait analysis is a clinically useful tool to quantify the mobility state of the neurological disorder. The monitoring of the joint kinematics, kinetics and the dynamic EMG data may provide insight into determining the neuromuscular and skeletal contributions for assessing gait abnormalities and treatment outcomes. EMG signals during gait provide insight into muscle recruitment patterns and neuromuscu- lar control of walking. According to [6], diagnostic assess- ment and treatment decisions may be based on the EMG behavior associated with the dynamics of gait. Previous work [1,10] also illustrated that EMG activity was repeated consistently over a locomotion gait cycle during preferred walking speed in normal adults. The muscles differ from the peak activity at each of the various phases of the gait cycle. This experiment takes advantage of the surface EMG (SEMG) to monitor patterns of muscle activation during the gait cycle. The SEMG is non-invasive and easy to per- form but exhibits some limitations based on its causative factors that affect the quality of the signals. De Luca [3] cat- egorized these causative factors into two groups: extrinsic and intrinsic [3]. The extrinsic factors are electrode config- uration, location of the electrodes and orientation of the 0263-2241/$ - see front matter Published by Elsevier Ltd. doi:10.1016/j.measurement.2010.04.010 * Corresponding author. Address: Electrical and Computer Engineering, The University of Texas at El Paso, Engineering Annex Building, Room A- 316, 500 West University Avenue, El Paso, TX 79968, USA. Tel.: +1 915 747 7011; fax: +1 915 747 7871. E-mail addresses: hyu@miners.utep.edu (H. Yu), msalaqtash@miner- s.utep.edu (M. Alaqtash), erictspier@hotmail.com (E. Spier), tsg@ece.ute- p.edu (T. Sarkodie-Gyan). 1 Address: Electrical and Computer Engineering, The University of Texas at El Paso, Engineering Building, Room E321, 500 West University Avenue, El Paso, TX 79968, USA. Measurement 43 (2010) 1106–1114 Contents lists available at ScienceDirect Measurement journal homepage: www.elsevier.com/locate/measurement