An Approach for Cardiac Coronary Detection of Heart Signal Based on Harris Hawks Optimization and Multichannel Deep...

Area of Science:

• Cardiology
• Artificial Intelligence
• Signal Processing

Background:

• Early detection of cardiovascular disease is crucial for patient outcomes.
• Electrocardiogram (ECG) analysis is a primary tool for diagnosing cardiac conditions like arrhythmia.
• Current diagnostic models face challenges in real-time processing efficiency and computational load.

Purpose of the Study:

• To develop an efficient deep neural network model for automatic arrhythmia diagnosis using ECG signals.
• To enhance the temporal and spatial fusion of information from ECG data.
• To improve the computational efficiency and accuracy of arrhythmia detection.

Main Methods:

• A deep neural network (DNN) model was developed, incorporating Harris Hawks optimization.
• The model was designed for flexible input length and achieved a reduction in parameters and computations.
• Temporal and spatial information fusion from ECG signals was implemented.

Main Results:

• The proposed DNN model demonstrated high diagnostic performance with 96.04% sensitivity, 93.94% specificity, and 95.00% accuracy.
• Significant reductions in model parameters (halved) and real-time processing computations (over 50%) were achieved.
• The approach showed practical advantages over existing methods in performance and efficiency.

Conclusions:

• The Harris Hawks-optimized DNN model offers a highly accurate and computationally efficient solution for automatic arrhythmia detection from ECG.
• This method presents a significant advancement for early cardiovascular disease diagnosis and prevention.
• The model's practical advantages make it suitable for real-time clinical applications.