Non-invasive, high-precision gastrointestinal motility function assessment system and method

Abstract

Disclosed are a non-invasive gastrointestinal motility function assessment system and a multimodal physiological signal processing method. The system includes a multimodal signal acquisition module, a signal preprocessing and feature extraction module, a multi-source information fusion and intelligent analysis module, and a gastrointestinal motility assessment and report generation module. Gastrointestinal acoustic signals, bio-impedance signals, and abdominal wall micro-vibration signals are synchronously acquired using an attachable, flexible device, and then subjected to noise reduction, filtering, and feature extraction, thereby achieving multimodal features reflecting the frequency and intensity of peristalsis and the motion state of contents. A deep learning model based on an attention mechanism is employed for deep fusion of the multimodal features, and quantitative indicators, such as estimated gastric emptying rate, peristalsis frequency and average amplitude, motility index, and percentiles relative to a standard database, are output.

Images