Basic Research for High-speed Heart Sound Determination using AI

In recent years, many researchers have focused on enhancing the current situation of home healthcare systems.In this paper, we developed a system that uses AI to quickly determine whether auscultation sounds are normal or not.In our analysis, heart sounds were imaged, and their abnormalities were identified using machine learning tools.The proposed approach uses a Yolo_v7 model-based anomaly detection and the Wavelet Transform was employed to analyze the acquired heart sound data.Proposed system resulted with an 90% accuracy rate or over on the basis of 5 seconds of heartbeat data, in various recording environments.