Automatic Evaluation of Gastrointestinal Motor Activity through the Analysis of Bowel Sounds

Abstract

Approximately 10%-20% of people have Irritable Bowel Syndrome (IBS) worldwide. It is well known that IBS is associated with altered motility of the intestine. Currently, the diagnosis of IBS is made through expensive invasive radiographic and endoscopic examinations. The bowel sounds (BSs) are produced by the propelled-air during peristalsis, and can be easily recorded using non-invasive and low-cost equipment. In this study we propose a novel autoregressive moving average (ARMA)-based method to automatically detect BSs, and we use the measure, BS duration (BSD), to capture gastrointestinal motor activity. The performance of the proposed method is investigated and evaluated via computer simulations. Working on healthy volunteers during mosapride administrations, we show that the proposed method for automatic detection could achieve a sensitivity of 83.4%, specificity of 93.3% and accuracy of 91.2%. Moreover, we show that the measure BSD can conveniently capture gastrointestinal motor activity. Our findings should have the potential to contribute towards developing bowel-sound based diagnosis of IBS.