Blood Vessel Blockage Detection using Phonocardiogram

The cardiovascular diseases (CVDs) are still one of the top killers globally, and early detection is needed for proper management. The current research offers an IoT-based system incorporating a NodeMCU microcontroller, an electronic stethoscope (PCG signal acquisition), and cloud processing using ThingSpeak. The heart sounds signals are preprocessed and analyzed by wavelet transform in MATLAB, using machine learning algorithms—Decision Tree (DT), Random Forest (RF), and Naïve Bayes (NB)—for cardiac condition classification. The architecture allows real-time observation, remote diagnosis, and predictive algorithms for increased accessibility and precision in CVD detection. Performance assessment of the algorithms proves the efficiency of the proposed method in detecting abnormal heart sounds. The low-cost and portable device provides a viable alternative for early CVD screening, particularly in low-resource environments. Future development can involve integrating deep learning and mobile-based diagnostic apps for increased clinical applicability.