Detecting and measuring human gastric peristalsis using magnetically controlled capsule endoscope

Xueshen Li,Yu Gan,David Duan,Xiao Yang

DOI: https://doi.org/10.48550/arXiv.2301.10218

2023-01-25

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.

Image and Video Processing,Computer Vision and Pattern Recognition

What problem does this paper attempt to address?

This paper aims to solve the problem of detecting and measuring human gastric peristalsis through magnetically - controlled capsule endoscopy (MCCE). Specifically, the authors developed algorithms to detect and measure human gastric peristaltic waves in video sequences obtained using MCCE. The paper points out that traditional methods for evaluating gastric motility, such as manometry, gastric - emptying scintigraphy, and electrogastrography, all have their own limitations. As an emerging tool, MCCE has the advantages of being comfortable, safe, and not requiring anesthesia in diagnosing gastric diseases, and it can provide real - time, true - color visual images of the gastric environment that are easy to interpret.

To overcome the problem that the quality of MCCE video sequences is affected by camera motion, the authors designed a camera motion detector (CMD) to process MCCE video sequences and reduce camera movement during MCCE examinations. In addition, the authors also developed spatio - temporal deep - learning algorithms using convolutional neural networks (CNN) and long - short - term memory networks (LSTM) to detect gastric contraction waves and measure the cycle of human gastric peristalsis. To the best of the authors' knowledge, this is the first computer - vision - based solution proposed to detect and measure human gastric peristalsis.

Overall, the purpose of this study is to improve the ability to evaluate gastric motility by developing and applying these advanced computer - vision techniques, thereby assisting in the diagnosis of gastric diseases.