Effective S1 S2 detection system with beat track method

Abstract

Heart sound diagnosis or auscultation is one of the most simple, non-invasive and costless method used to evaluate heart health and one of the basic and foremost routine of a doctor while reviewing a patient. Detecting cardiac abnormality by auscultation demands a physician's experience, and even then has a high scope of committing error. In this paper, a light, fast and low computation speed method is presented for detection of ‘lub’ ‘dub’ heart sounds. It is a light version of beat track method that is extensively used in the music industry to detect the periodic beats in a particular music by adapting to the timing intervals and can be used satisfactorily on both normal and pathological heart sounds. Moreover heartbeats can be identified in both de-noised and noised environment as it independent of external disturbances. The results showed over 94.73% accuracy in recognizing heart beats in noised environment and almost 100% in de-noised conditions. Moreover, the exact position detection error is less than 1%.