Analisis Dan Ekstraksi Ciri Sinyal Suara Jantung Menggunakan Dekomposisi Wavelet
Heart disease is ranked top cause of death in the world. For this reason, auscultation is the main test performed by a doctor to evaluate the condition of the heart by listening to the heart sounds through a stethoscope, auscultation is a basic component in cardiac diagnosis.
Based on the reasons above, it is required a way to ease diagnosis by recording sounds of the heart then analyzing the heart sound signal and proccess it to be able to detect and recognize the patterns of the heart sound signal. In this research, we analyzed and extracted features of real data signal of heart sounds that were acquired by auscultation (using digital stethoscope), utilizing the decomposition of discrete wavelet transformation. Several types of mother wavelet were employed for different kinds of order, various level on each order and three sorts of frequency sampling that are 8KHz, 44,1KHz, and 48KHz. Parameters that are used for analysis are energy and the standards deviation. The our finding shows that pattern or characterization of normal heart sounds for frequency sampling 8 KHz produces the largest energy decomposition in D6 ( 62.5Hz-125Hz ). On the other hand at frequency sampling 44,1 KHz it produces the largest energy decomposition in D9 ( 43.066Hz-86.133Hz). For frequency sampling 44,1 KHz, it produces the largest energy decomposition in D9 (46.88Hz-93.75Hz).
- There are currently no refbacks.