Multi-sensor respiration monitoring method based on long and short term memory (LSTM) network

• Main Authors: YAO LEIBO, ZHANG JUANMEI, REN YAFEI
• Format: Patent
• Language: Chinese; English
• Published: 23.08.2024
• Subjects: DIAGNOSIS; HUMAN NECESSITIES; HYGIENE; IDENTIFICATION; MEDICAL OR VETERINARY SCIENCE; SURGERY

A multi-sensor respiration monitoring method based on a long short term memory (LSTM) network comprises the following specific steps: S1, adopting a plurality of electrode plates to perform single-lead monitoring on ECG signals, and adopting a piezoresistive sensor to perform respiration signal pressure monitoring; s2, data preprocessing is conducted on the ECG signals and the respiratory signals collected in the step S1; and S3, extracting respiratory characteristics of the ECG signal and the respiratory signal preprocessed in the step S2 through a long-short term memory network, estimating respiratory rates from the two signals, obtaining a final respiratory rate result by adopting information entropy weighted fusion, and outputting and feeding back the final respiratory rate result. Single-lead monitoring of ECG signals is carried out on multiple parts at the same time through the multiple electrode slices, breathing signals of the multiple parts are collected at the same time through only one strip-shaped