Colonic pressure data processing based on independent component analysis

Abstract

The Colonic manometry is an important technique to evaluate human colonic motor functions, which are critical for doctors to understand the pathology of intestinal diseases like slow transit constipation (STC) and colonic inertia (CI). However, in the obtained pressure signals, several patterns of colonic motor activities as well as noises mixed together, which made it difficult to observe the information people really needed. In this article, a new method was proposed to extract patterns of colonic motility from the mixed signals, so that researchers could study them thoroughly. Colonic pressure recordings from 26 volunteers were obtained by the water-perfused manometry catheters. Then independent component analysis (ICA) was introduced, which successfully separated colonic motility patterns and noises into four independent components. And according to the rhythm of contractions examined by ICA, subjects' colonic motility could be divided into three types: regular rhythm (12 subjects), slow rhythm (8 subjects) and disordered (6 subjects), which exactly accorded with their original diagnosis.

Keywords:

• ICA
• Colonic pressure recording
• Patterns of colonic motility

Acknowledgements

The authors thank the editors and reviewers for their efforts to help improve this article. This work was supported by China National Natural Science Foundation (No. 60875061).