ECG based Myocardial Infarction detection using Hybrid Firefly Algorithm.

Area of Science:

• Cardiology
• Biomedical Engineering
• Artificial Intelligence in Medicine

Background:

• Myocardial Infarction (MI) poses a significant global health burden, causing mortality, disability, and economic loss.
• Current diagnostic methods for MI are often invasive and lack optimal detection accuracy.
• Extensive feature extraction from ECG signals can lead to computational inefficiency and reduced performance.

Purpose of the Study:

• To develop an optimized ECG signal analysis method for accurate Myocardial Infarction detection.
• To bypass traditional feature extraction limitations by directly optimizing raw ECG signals.
• To enhance the diagnostic accuracy and efficiency of cardiovascular disorder assessment.

Main Methods:

• Utilized a Hybrid Firefly and Particle Swarm Optimization (FFPSO) algorithm for direct ECG signal optimization.
• Applied the FFPSO algorithm to raw ECG signals, bypassing conventional feature extraction techniques.
• Integrated an Artificial Neural Network (ANN) with the optimized ECG signals for classification.

Main Results:

• The FFPSO-ANN model achieved 99.3% accuracy in detecting Myocardial Infarction.
• Demonstrated high sensitivity (99.97%) and specificity (98.7%) for MI detection using the MIT-BIH and NSR databases.
• The optimization approach significantly improved diagnostic performance compared to traditional methods.

Conclusions:

• ECG signal feature optimization using FFPSO is a highly effective method for diagnosing heart conditions.
• The proposed approach offers a more accurate and efficient alternative for cardiovascular disorder diagnosis.
• Direct optimization of ECG signals presents a promising direction for improving cardiac patient care.