Classification of Abnormal and Normal Heart Sounds Using the MEMS based High Performance PhonoCardioGraphy System

The paper focuses on the classification of normal and abnormal heart sounds through a low-cost, high-performance PhonoCardioGraphy system using MEMS microphones with good sensitivity for the heart sounds. The proposed algorithm classifies the heart sound components using wavelet decomposition and Shannon energy. The PhonoCardioGraphy system, with the proposed algorithm tested by taking 1756 results, includes both normal and abnormal heart sounds. Experimental results revealed a good sensitivity of 99.17% and detection error rate of 1.5%.