Portable ECG and PCG wireless acquisition system and multiscale CNN feature fusion Bi-LSTM network for coronary artery disease diagnosis.
Journal:
Computers in biology and medicine
PMID:
40239232
Abstract
Coronary artery disease (CAD) is a major cause of mortality, especially among aging populations, making timely and accurate diagnosis essential. In this work, a portable wireless device powered by artificial intelligence for CAD detection is proposed, which synchronously captures electrocardiograms (ECG) and phonocardiograms (PCG) signals and transmits them for real-time analysis and visualization. To ensure the reliability of the acquired signals, a Hidden Semi Markov model is applied to validate data quality. Then, a multiscale convolutional neural network (CNN) feature fusion model extracts critical features from the PCG and ECG signals. All these features and signal information are later processed by a bidirectional long short-term memory (Bi-LSTM) network. Our network achieves impressive metrics and maintains reliable performance in practical tests. This straightforward diagnostic system offers a practical and technically feasible solution for the effective diagnosis of CAD, leveraging advanced neural network architectures for robust clinical application.