AI Edgebeat-Towards the Development of Digital Stethoscope for the Detection of Heart Valve Disorders

Heart ailments encompass a wide range of conditions that affect heart structure and function. While heart sound auscultation is a primary diagnostic method, it requires significant expertise and is time-consuming. Research indicates that heart sound changes often precede ECG abnormalities, underscoring its diagnostic importance. Although deep learning models have shown promise in identifying valvular heart diseases (VHDs) using phonocardiograms (PCGs), their real-time applications remain underexplored. This highlights the need for hardware-based systems to improve the efficiency and accuracy of heart disorder detection. This study lays the foundation for the hardware integration towards the development of a digital stethoscope, enabling real-time classification of heart sound signals for VHDs. The PCG signals were analyzed and classified using an artificial intelligence (AI)-enabled edge-computing platform. This approach achieved a diagnostic accuracy of up to 98.07% in detecting heart disorders. Furthermore, it serves as a valuable tool for clinicians in decision-making processes, potentially preventing the development of cardiac abnormalities in numerous cases.