An ECG Signal Acquisition and Analysis System Based on Machine Learning with Model Fusion

Area of Science:

• Biomedical Engineering
• Cardiology
• Machine Learning

Background:

• Cardiovascular disease is a leading global cause of death.
• Abnormal heart rate signals are key indicators of cardiovascular disease.
• Current ECG devices lack portability and rely on manual analysis, hindering efficient data processing.

Purpose of the Study:

• To develop a portable ECG acquisition and analysis system using machine learning.
• To improve the accuracy and efficiency of ECG signal classification for cardiovascular disease detection.

Main Methods:

• Developed an ECG analysis system comprising data preprocessing and machine learning models.
• Employed traditional models (logistic regression, SVM, XGBoost) with morphological and wavelet features.
• Integrated deep learning models (CNNs, LSTMs) for enhanced classification through model fusion.

Main Results:

• Achieved a classification accuracy of 99.13% for ECG signals.
• Demonstrated the effectiveness of model fusion combining traditional and deep learning approaches.

Conclusions:

• The proposed machine learning-based ECG system offers a highly accurate solution for cardiovascular disease detection.
• Future work will focus on model optimization and the development of field-deployable portable instruments.