A respiratory health auxiliary monitoring method and system for chronic lung disease

By collecting multimodal data and performing motion feature enhancement, noise suppression, and decoupling analysis, combined with an asynchronous adaptive fusion algorithm to generate a highly reliable comprehensive respiratory waveform, the problem of insufficient multimodal signal fusion in existing technologies is solved, and refined monitoring of chronic lung diseases is realized.

CN122140224APending Publication Date: 2026-06-05THE THIRD HOSPITAL OF CHANGSHA
View PDF 14 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
THE THIRD HOSPITAL OF CHANGSHA
Filing Date
2026-05-08
Publication Date
2026-06-05

AI Technical Summary

Technical Problem

Existing methods for monitoring respiratory health in chronic lung diseases rely on single sensor signals, fail to effectively integrate multimodal respiratory signals, and lack processing of respiratory physiological coupling relationships. This results in weak reliability and physiological correlation of signal fusion results, making it difficult to achieve refined monitoring of disease categories and severity.

Method used

Multimodal data was collected, including chest rise and fall motion image sequences, ambient sound streams, and chest and abdominal motion waveforms recorded by a wearable breathing belt. Through motion feature enhancement, noise suppression, and decoupling analysis, a highly reliable comprehensive respiratory waveform was generated. The signals were then integrated using an asynchronous adaptive fusion algorithm to extract respiratory rhythm, depth, and symmetry features, which were then input into a pre-trained classification model to determine disease category and severity.

Benefits of technology

It enables multi-dimensional characterization of respiratory motion stability and physiological correlation, outputs disease categories and severity levels that match pathological change characteristics, and improves the accuracy and reliability of monitoring.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122140224A_ABST
    Figure CN122140224A_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of respiratory health monitoring, in particular to a respiratory health auxiliary monitoring method and system for chronic lung diseases, comprising: collecting a chest fluctuation image sequence, an environmental sound stream and a wearable chest and abdominal movement waveform to form multi-modal respiratory data, and obtaining a respiratory effort signal, a respiratory sound feature and a chest and abdominal breathing component through targeted processing. Relying on the respiratory physiological coupling relationship, a high-credibility comprehensive respiratory waveform is generated through an asynchronous adaptive fusion algorithm, multi-dimensional features such as respiratory rhythm, depth and symmetry are extracted, a pre-trained chronic disease state classification model is input, and a corresponding disease category and severity grade are output. The method can weaken the time sequence difference of multi-source signals, strengthen the expression of respiratory physiological features, improve the recognition accuracy of abnormal respiratory patterns, and realize continuous and refined respiratory state monitoring and disease grading evaluation for chronic lung diseases.
Need to check novelty before this filing date? Find Prior Art