Wavelet-Based ECG Signal Processing for Early Detection of Myocardial Infarction

Heart attacks remain a leading global cause of mortality, and early detection is essential for enhancing patient survival and reducing long-term complications. This study addresses the challenge of timely myocardial infarction detection using real-time electrocardiogram (ECG) signal processing. The proposed method integrates wavelet denoising for noise suppression, the Pan-Tompkins algorithm for accurate QRS complex detection, and a rule-based approach for ST segment elevation identification, a key marker of acute heart attacks. validation was conducted in two stages: Stage 1 used PhysioNet ECG datasets to test the algorithm’s performance, while Stage 2 implemented real-time acquisition using an AD8232 sensor interfaced with Arduino for live monitoring. Results demonstrate that wavelet-based preprocessing combined with Pan-Tompkins QRS detection significantly enhances signal clarity, and the integrated system reliably identifies ST elevation. A convolutional neural network (CNN) was employed for classification, achieving high accuracy in differentiating normal and abnormal signals. The findings confirm that combining wavelet signal processing with real-time embedded monitoring provides a cost-effective and efficient solution for early heart attack detection. Clinically, the system demonstrates significant potential for deployment in wearable devices or remote health monitoring setups, enabling timely alerts and intervention in resource-limited environments. This approach contributes to preventive cardiology, enhances patient safety, and may reduce the overall burden of cardiovascular disease.