Ballistocardiogram ventricular fibrillation auxiliary diagnosis system based on three-channel image and transfer learning

Abstract

The invention belongs to the technical field of medical instruments, and particularly relates to a ballistocardiogram ventricular fibrillation auxiliary diagnosis system based on three-channel images and transfer learning. The system comprises a signal preprocessing module, an image conversion module, a feature extraction module, a normalization processing module and a classifier. According to the method, a one-dimensional BCG signal is converted into three single-channel two-dimensional diagrams in a user-defined mode, then the three-channel two-dimensional diagrams are spliced into a three-channel image, the three-channel image serves as an input tensor, a plurality of CNN models obtained through pre-training of a large image library are migrated to a BCG domain for learning, feature values are extracted through middle layers of all the models and spliced into one-dimensional feature vectors, and the one-dimensional feature vectors are subjected to one-dimensional feature extraction; and then automatic discrimination of ventricular fibrillation, sinus rhythm and motion artifacts is realized through a full connection layer or a machine learning classifier and the like. The method can effectively solve the problem that ventricular fibrillation recognition is difficult due to BCG waveform diversity and small sample size characteristics, and a feasible scheme is provided for long-term health monitoring of family cardiovascular diseases.

Description

technical field [0001] The invention belongs to the technical field of medical instruments, and in particular relates to a shock-cardiogram ventricular fibrillation auxiliary diagnosis system based on three-channel images and migration learning. Background technique [0002] Sudden cardiac death (SCD) is a major public health burden worldwide, and due to medical constraints, most SCDs occur out of hospital. Ventricular fibrillation is the primary cause of SCD, which is uncertain and highly critical. Once it occurs, the ventricular output drops sharply and the patient's circulation is interrupted. If not intervened in time, sudden death will occur within minutes. Electrical defibrillation is the only effective method for the treatment of ventricular fibrillation, and its probability of success is inversely proportional to the duration of ventricular fibrillation, and the success rate of resuscitation decreases by nearly 10% for every minute of delay. Therefore, a non-contact ...

AI Technical Summary

Problems solved by technology

[0004] The present invention aims at the above-mentioned ventricular fibrillation detection algorithm requirements and the technical deficiencies of the existing BCG research, and provides a three-channel image and transfer learning-based ballistocardiogram (BCG) ventricular fibrillation auxiliary diagnosis system to solve the problems caused by the diversity and small size of BCG waveforms. Difficulty in identifying ventricular fibrillation due to sample size characteristics

Images