Machine Learning and Deep Learning-Based Myocardial Infarction Detection and Heart Activity Analysis Using ECG Signals

Abstract: Myocardial Infarction (MI) remains one of the leading causes of mortality worldwide, necessitating accurate and timely diagnosis to reduce fatal outcomes. Electrocardiogram (ECG) signals serve as a primary non-invasive diagnostic tool for detecting cardiac abnormalities; however, manual interpretation is time-consuming and prone to human error. This paper presents an intelligent framework for automated myocardial infarction detection and heart activity analysis using machine learning (ML) and deep learning (DL) techniques. The proposed system integrates ECG signal acquisition, preprocessing, feature extraction, and classification into a unified pipeline. Noise and artifacts such as baseline wander and power-line interference are removed using digital filtering techniques. Key features, including R–R intervals, QRS complex duration, and morphological characteristics, are extracted to represent cardiac activity effectively. Multiple ML algorithms, including Support Vector Machine (SVM), Random Forest (RF), and K-Nearest Neighbors (KNN), are employed and compared with deep learning models such as Convolutional Neural Networks (CNN) and hybrid CNN-LSTM architectures. Experimental results demonstrate that the proposed approach achieves high accuracy, sensitivity, and specificity in distinguishing normal and MI-affected ECG signals. The integration of deep learning enhances the system’s ability to capture complex temporal and morphological patterns, improving diagnostic performance. The developed system is computationally efficient and suitable for real-time applications in telemedicine, wearable health monitoring systems, and remote cardiac diagnosis. This work highlights the potential of combining signal processing with intelligent algorithms to provide reliable, scalable, and automated cardiac health assessment.