Diagnosis and segmentation of S1 and S2 heart sounds with digital analysis methods

Listening with stethoscope is a preferential method that the doctors use in order to differentiate normal cardiac systems from the abnormal ones that come out. On the other hand, with the ECG method which is used to diagnosis heart abnormalities, is experienced the difficulties at diagnosis heart murmur and abnormal heart sounds. At the health center or village clinic, due to use of low cost equipment and deficiency of medical expert, the heart auscultation is preferred because of diagnosis. But, for the diagnosis of heart abnormalities, doctor must need experience and skill with traditional auscultation equipment. To determinate of heart abnormalities and segmentation of heart sounds, development auxiliary systems should be important for doctors who have not experience and skills. In this thesis, a system has been developed for doctors that it can be assist to doctor and it analyze the heart sounds using the Digital Signal Processing (DSP) technique. In this system, sounds that are taken with electronic stethoscope will be untied and sounds will be subjected to Digital Signal Processing technique. These techniques are Fast Fourier Transform (FFT), Discrete Fourier Transform (DFT), Shannon Energy, and Power Spectrum Density (PSD). For more powerful application, sounds are more convenient using Hamming, Hanning and Blackman windowing functions for DSP technique. As a result of, these DSP techniques will be comparing and optimum DSP technique and windowing function will be chose for heart sound analysis.