A practical measuring system for intestinal pressure activity and its clinical application

Abstract

Aim To propose a practical, non-invasive method for monitoring intestinal pressure activity and introduce its clinical application on discrimination between the healthy subjects and the slow transit constipation (STC) patient subjects in the sense of medical statistics. Methods Nineteen healthy subjects and eight STC patient subjects participated in the experiments. The authors tried to find the optimal feature combination by using the probabilistic neural network (PNN) to separate these two groups and test their significant differences. Results The statistical analysis revealed that the combination of the frequency and the average pressure of the pressure activity induced by colonic peristalsis was the optimal feature combination among the possible feature subsets, which was composed of several typical extracted features, and the discrimination accuracy of test samples was about 88.89%. The one-way analysis of variance (ANOVA) analysis also showed that there were differences or significant differences between them. Conclusions Clinical experiments and analysis method related verified the practicability of the system, and the measurement technology developed for monitoring intestinal pressure activity was useful.