Advancing Artificial Intelligence Framework for Accurate and Early Heart Attack Detection and Evaluation

Abstract

"Abstract Cardiovascular diseases (CVDs) represent prevalent chronic conditions that constitute significant threats to global public health. The electrocardiogram (ECG) is a widely utilized noninvasive diagnostic technique that records the dynamic bioelectrical activity of the heart. By employing electrodes affixed to the skin39;s surface, the ECG enables continuous monitoring of the heart’s cyclical phases of contraction and relaxation. Atrial fibrillation (AF), characterized by irregular cardiac rhythms and the frequent absence of a discernible P-wave, markedly increases the risk of adverse clinical outcomes if left untreated. These complications include heart failure, ischemic stroke, cardiovascular mortality, thromboembolic events, blood coagulation disorders, and cognitive decline. The asymptomatic and often clinically silent nature of AF presents substantial challenges for early detection, potentially leading to fatal consequences. The manual diagnosis of atrial fibrillation (AF) using electrocardiograms (ECGs) presents notable limitations, as it requires substantial clinical expertise and entails a laborintensive, time- consuming process. Hence, recent advancements in Artificial Intelligence have had a transformative impact on the domain of cardiovascular diagnostics. This study proposes an automated framework capable of capturing both temporal and frequency features of cardiac signals to facilitate accurate AF classification. The initiative aligns with the broader vision of integrating the Internet of Things (IoT) and AI technologies, thereby establishing a forwardlooking model for proactive cardiovascular health monitoring. The classification efficacy of the system was rigorously evaluated using a Support Vector Machine (SVM) algorithm, achieving a classification accuracy of 94% across two categories: Atrial Fibrillation (AF) and Normal Sinus Rhythm (NSR). This novel approach holds substantial promise for deployment in clinical environments, enabling real-time detection of AF during routine ECG assessments, ultimately enhancing patient care while alleviating the diagnostic burden on medical professionals.