Intestinal motility assessment based on Legendre fitting of logarithmic bowel sound spectrum

Abstract

A quantitative intestinal motility (IM) assessment method based on a benchmark model using Legendre fitting of bowel sound spectrum is presented. The logarithmic magnitude of bowel sound sets containing IM events is decomposed into Legendre polynomial series to create the benchmark. For a bowel sound data segment under assessment, the relative spectrum fluctuation and correlation with the benchmark model are calculated and quantitative assessment can then be obtained. Experimental results have proven the accuracy of the proposed method.