Physiological signal detection system based on multi-channel flexible integration

Abstract

The invention provides a physiological signal detection system based on multi-channel flexible integration. The physiological signal detection system is characterized by comprising n physiological signal detection channels, each physiological signal detection channel is provided with a single-channel feature extraction module, the output end of each physiological signal detection channel is connected with a soft-decision evidence generating module, the soft-decision evidence generating modules of the n channels are all connected in a multi-channel information integration module, and the output end of the multi-channel information integration module is connected with a diagnosis decision-making module. The physiological signal detection system has the advantages that the conflicting information integration problem existing during multi-channel binary detection is solved, robust detection of weak electrophysiological feature signals is achieved, the physiological significance of the channels are retained completely in the feature extraction process, the system conforms to the interpretation habit of a doctor, experience threshold values summarized for a long term can be clinically and normally used in the soft-decision process, and the system has very good universality for multi-channel detection of weak electrophysiological signals.

Description

technical field [0001] The invention relates to biomedical electrical technology, in particular to a physiological signal detection system based on multi-channel flexible fusion. Background technique [0002] In the fields of telemedicine and vital sign monitoring, the detection and analysis of electrophysiological characteristic signals such as EEG, ECG, and EMG are the basic technologies for health monitoring, early warning, and diagnosis and treatment. Because the electrophysiological characteristic signals of the human body are usually very weak, susceptible to various noise interference, and the signal itself is non-stationary, it is a common practice in the field of electrophysiological signal processing to use data collected by multiple channels for spatial redundancy detection. . For example, clinical 12-lead ECG examination produces 8 independent acquisition channels, scalp EEG signal detection uses 8*8 electrode array (64 channels), atrial epicardial mapping uses ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problems that the above-mentioned multi-channel phase-or-type fusion method cannot handle the fusion of conflicting detection results, the data utilization rate is low, and the matrix decomposition fusion method loses the physiological meaning of the channel, the present invention proposes a physiological signal detection system based on multi-channel flexible fusion, The system first extracts the characteristic signal on a single channel according to the detection method that is clinically in line with the doctor's diagnostic habits, obtains the channel detection evidence through soft judgment, and flexibly fuses multi-channel evidence under the framework of uncertain information fusion theory to solve the problem of multi-channel binary detection. In the process of feature extraction, the system fully retains the physiological meaning of the channel, which is in line with the interpretation habits of doctors. The long-term clinical experience threshold can be used in the soft judgment process. In normal use, and various existing single-channel determination methods can be used alone or at the same time, and it has strong versatility for multi-channel detection of weak electrophysiological signals

Images