Deep learning hybrid model ECG classification using AlexNet and parallel dual branch fusion network model.

Area of Science:

• Biomedical Engineering
• Artificial Intelligence in Healthcare
• Cardiology

Background:

• Cardiovascular diseases are a leading cause of mortality, necessitating precise diagnostic methods.
• Electrocardiography (ECG) is vital for detecting various cardiac conditions, but manual analysis is time-consuming and error-prone.
• Automated ECG analysis is essential for improving diagnostic efficiency and accuracy.

Purpose of the Study:

• To evaluate the performance of AI models, specifically AlexNet and a dual branch model, for automated ECG signal classification.
• To compare the proposed models against existing state-of-the-art methods in ECG analysis.
• To demonstrate the potential of AI in providing accurate and adaptable solutions for cardiovascular diagnostics.

Main Methods:

• Utilized the PTB Diagnostic ECG Database for training and testing AI models.
• Employed data preprocessing techniques including standardization, balancing, and reshaping of ECG signals.
• Implemented and evaluated AlexNet and a dual branch fusion network for ECG classification.

Main Results:

• AlexNet achieved a validation accuracy of 98.64% and a test set accuracy of 99%.
• The dual branch fusion network model attained a test set accuracy of 99%.
• Both models demonstrated high precision, sensitivity, and specificity, outperforming other advanced models.

Conclusions:

• The proposed AI models, AlexNet and the dual branch network, show exceptional performance in ECG classification with 99% accuracy.
• These findings highlight the significant advantages of integrating machine learning into automated ECG analysis systems.
• The developed models offer scalable and effective solutions for diverse healthcare settings, including remote and rural areas.