Detection of Subacute Intestinal Obstruction from Surface Electromyography Signatures

Subacute intestinal obstruction (SAIO) has become an important health concern in the field of medical diagnosis as delayed diagnosis may be fatal. The currently practiced clinical diagnostic tools are either uncomfortable for the patient or may leave harmful side effects on the human body. The aim here is to design such a diagnostic methodology that is harmless, painless, and comfortable for the patient. For this, the surface electromyography (sEMG) technique is proposed. sEMG signals from the abdominal surface of diseased and normal human beings were acquired and analyzed. Three best features of mean, skewness, and kurtosis were extracted for accurate classification of sEMG signals into normal and SAIO classes. Support vector machine (SVM) was used for training and testing the dataset at 10-fold cross-validation, giving the mean accuracy of 96% with 95% specificity and 96.7% sensitivity in SAIO diagnosis using sEMG. This research work proposes an accurate, safe, and facile method for the detection of SAIO.