Automated System for the Detection of Heart Anomalies Using Phonocardiograms: A Systematic Review

Phonocardiogram (PCG) signals generated by the heart contain information about heart conditions. This review examines how PCG analysis identifies and diagnoses heart issues. We studied traditional signal processing and artificial intelligence techniques and provided a complete picture of the current state of this field. Adhering to the systematic review guidelines, our comprehensive review covers 103 studies from reputed journals. It includes Machine Learning (ML) and Deep Learning (DL) techniques used to develop the computer-aided diagnostic tools using PCG signals. This review evaluates the strengths and weaknesses of various ML and DL methods, emphasizing their effectiveness in diagnosing several abnormalities. Additionally, we examine the obstacles and challenges limiting the widespread adoption of PCG-based diagnostic systems in clinical settings. We outline a plan for future research to develop improved versions of PCG analysis models. These models will be more robust, precise, and user-friendly. They will improve cardiovascular care by enabling machines to screen for problems automatically and intelligently.