Automated Detection and Classification of Gastrointestinal Diseases using surface-EMG Signals

Gastrointestinal (GI) disorders affect human life widely and result in other health issues as well. The diagnostic measures in this area are mostly invasive, uncomfortable and possess health hazards. The objective here is to propose surface electromyography(sEMG) based framework for safe, easy and precise detection of GI disorders. In this research, abdominal sEMG signals were acquired from 15 subjects. 77 samples from afflicted individuals and 65 samples from healthy ones are collected and analyzed. The acquired signals are first preprocessed and segmented using empirical mode decomposition (EMD) and mel-frequency cepstral coefficients (MFCCs) are extracted to be the best feature for classification of signals into normal and diseased categories. Testing and training using support vector machine (SVM) accurately classified signals into normal and diseased classes. This framework is found to be 100% accurate in disease detection and classification. It is a new approach in GI ailments diagnosis by being non-invasive, accurate and safe.