Detecting and measuring human gastric peristalsis using magnetically controlled capsule endoscope

TL;DR

This paper introduces a novel computer vision approach using deep learning to detect and measure gastric peristalsis from magnetically controlled capsule endoscopy videos, aiding gastric disease diagnosis.

Contribution

It presents the first computer vision-based algorithms for detecting and measuring gastric peristalsis using MCCE video sequences, including a camera motion detector to improve accuracy.

Findings

Effective detection of gastric contraction waves.

Accurate measurement of gastric peristalsis periods.

Potential to assist gastric disease diagnosis.

Abstract

Magnetically controlled capsule endoscope (MCCE) is an emerging tool for the diagnosis of gastric diseases with the advantages of comfort, safety, and no anesthesia. In this paper, we develop algorithms to detect and measure human gastric peristalsis (contraction wave) using video sequences acquired by MCCE. We develop a spatial-temporal deep learning algorithm to detect gastric contraction waves and measure human gastric peristalsis periods. The quality of MCCE video sequences is prone to camera motion. We design a camera motion detector (CMD) to process the MCCE video sequences, mitigating the camera movement during MCCE examination. To the best of our knowledge, we are the first to propose computer vision-based solutions to detect and measure human gastric peristalsis. Our methods have great potential in assisting the diagnosis of gastric diseases by evaluating gastric motility.