J. Biomedical Science and Engineering, 2012, 5, 720-728 JBiSE http://dx.doi.org/10.4236/jbise.2012.512090 Published Online December 2012 (http://www.SciRP.org/journal/jbise/ ) Characterization and quantification of gait deficits within gait phases using fuzzy-granular computing * Melaku A. Bogale 1 , Huiying Yu 2 , Thompson Sarkodie-Gyan 2 , Amr Abdelgawad 3 1 University of Texas at El Paso, Computational Science Program, El Paso, USA 2 Department of Electrical and Computer Engineering, University of Texas at El Paso, El Paso, USA 3 Texas Tech University Health Science Center, El Paso, USA Email: mabogale@miners.utep.edu Received 5 October 2012; revised 7 November 2012; accepted 23 November 2012 ABSTRACT People with neurological disorders like Cerebral Palsy (CP) and Multiple Sclerosis (MS) suffer associ- ated functional gait problems. The symptoms and sign of these gait deficits are different between sub- jects and even within a subject at different stage of the disease. Identifying these gait related abnormali- ties helps in the treatment planning and rehabilitation process. The current gait assessment process does not provide very specific information within the seven gait phases. The objective of this study is to investi- gate the possible application of granular computing to quantify gait parameters within the seven gait phases. In this process we applied fuzzy-granular computing on the vertical ground reaction force (VGRF) and surface electromyography (sEMG) data to obtain respective characteristic values for each gait phase. A fuzzy similarity (FS) measure is used to compare pa- tient values with age and sex matched control able- bodied group. We specifically applied and tested this approach on 10 patients (4 Cerebral Palsy and 6 Mul- tiple Sclerosis) to identify possible gait abnormalities. Different FS values for VGRF for right and left leg is observed. The VGRF analysis shows smaller FS values during the swing phase in CP and MS subjects that are evidence of associated stability problem. Similarly, FS values for muscle activates of the four-selected muscle display a broad range of values due to diffe- rence between subjects. Degraded FS values for dif- ferent muscles at different stage of the gait cycle are reported. Smaller FS values are sign of abnormal ac- tivity of the respective muscles. This approach pro- vides individual centered and very specific informa- tion within the gait phases that can be employed for diagnosis, treatment and rehabilitation process. Keywords: Fuzzy-Granular Algorithm; Gait Phases; Fuzzy Similarity; Cerebral Palsy; Multiple Sclerosis 1. INTRODUCTION For people with mobility disabilities gait analysis is used to provide diagnosis, evaluation and treatment planning information. The benefit of gait analysis is well estab- lished that it has now become a part of routine process in many rehabilitation centers [1]. People with Multiple Sc- lerosis (MS) may suffer from significant gait impairment even at early stage of the disease [2,3]. Gait analysis has been used to identify associated gait deficit with MS [2,4]. Gait variability study in people with MS revealed slower walking speed and more fatigue than control heal- thy group [5]. In the study [6], the effect of MS on the frequency content of vertical ground reaction force (VGRF) during walking was investigated. Compared with health controls significantly lower frequency content in VGRF and no difference in frequency content in ante- rior-posterior ground reaction forces [6]. Lee EH et al. [7] emphasized the importance of gait analysis in critical surgical decision-making in children with Cerebral Palsy (CP). In this study [7] surgical decisions on children with CP, based on clinical evalu- ation and gait analysis was shown to help improve gait quality after surgery compared to decisions solely made on clinical assessment. According to [8] gait analysis has been used to make surgical procedure decisions in patients with CP. There are a growing number of lite- ratures [9-15] related to gait analysis and Cerebral Palsy in diagnosis and treatment planning decisions making process. Wavelet analysis was applied to study surface electromyography (sEMG) signals acquired from lower extremity muscles in children with CP [14]. Probabilistic gait classification in children with CP reported in [15]. Cluster analysis was used for identification of sagittal gait patterns [16]. Principal component analysis was ap- plied to extract gait patterns in children with CP [12]. * Conflict of interest: The authors do not have any financial and per- sonal relationships with other people or organizations that could have inappropriately influence (bias) their work. OPEN ACCESS