Timing and Classification of Patellofemoral Osteoarthritis Patients Using Fast Large Margin Classifier

Abstract

Surface electromyogram (sEMG) processing and classi cation can assist neurophysiological standardization and evaluation and provide habitational detection. The timing of muscle activation is critical in determining various medical conditions when looking at sEMG signals. Understanding muscle activation timing allows identi cation of muscle locations and feature validation for precise modeling. This work aims to develop a predictive model to investigate and interpret Patellofemoral (PF) osteoarthritis based on features extracted from the sEMG signal using pattern classi cation. To this end, sEMG signals were acquired from ve core muscles over about 200 reads from healthy adult patients while they were going upstairs. Onset, offset, and time duration for the Transversus Abdominus (TrA), Vastus Medialis Obliquus (VMO), Gluteus Medius (GM), Vastus Lateralis (VL), and Multi dus Muscles (ML) were acquired to construct a classi cation model. The proposed classi cation model investigates function mapping from real-time space to a PF osteoarthritis discriminative feature space. The activation feature space of muscle timing is used to train several large margin classi ers to modulate muscle activations and account for such activation measurements. The fast large margin classi er achieved higher performance and faster convergence than support vector machines (SVMs) and other state-of-the-art classi ers. The proposed sEMG classi cation framework achieved an average accuracy of 98.8% after 7 s training time, improving other classi cation techniques in previous literature.