Design and Implementation of a Low-cost Digital Stethoscope-Based Phonocardiogram (PCG) Acquisition System for Telemedicine Application

This paper presents the development of an affordable, digital stethoscope-based Phonocardiogram (PCG) acquisition system designed for telemedicine applications. The proposed system utilizes modern DSP and ML techniques with low-cost hardware components to capture and analyze heart sounds, providing real-time data transmission to healthcare professionals. The core design consists of a traditional stethoscope paired with a microphone, which records heart sounds via a smartphone or PC. By incorporating signal processing techniques, including denoising and machine learning algorithms, the device can classify heart conditions with promising accuracy. In trials involving 20 participants, PCG data was collected in both resting and post-exercise conditions, and heart rate measurements from the system were compared with ECG and oximeter readings. The results showed that the proposed PCG system maintained an overall error rate of less than 5% when compared to standard medical devices. Further classification of the PCG signals, using features like variance and skewness, demonstrated the system’s ability to distinguish between normal and abnormal heart sounds. With an average classification accuracy of 78.39%, sensitivity of 77.85%, and specificity of 78.94%, the device proved to be a practical, low-cost solution for telemedicine. By improving access to cardiac care, this PCG acquisition system could significantly enhance early detection and monitoring of heart diseases, particularly in remote areas where traditional medical services are scarce.