利用短時傅立葉轉換及支持向量機對心音訊號做自動分析

Abstract

心臟疾病已成為國人第二大死因，大多數心臟疾病是由心瓣膜的不正常所造成。人的耳朵可經由電子聽診器聽取心音探查心臟疾病的類型，但解釋心音是一個非常特殊的技巧，必須要接受嚴格的訓練才能做正確的心音聽診。由於這個原因，自動心音分析的電腦系統將對醫務人員會有很大的幫助。本文提出了一種完整的心音分析系統涵蓋從分割心臟週期到最後確定心臟疾病的類型。心臟週期的分割與識別是根據短時傅立葉變換(STFT)和支持向量機(SVM)。實驗過程中，心音資料來源是來至德州心臟學會公開的心音資料，並非常有希望達到不錯的辨識率。

The heart disease has become the second cause of death, and most of heart diseases result from heart valve disorders. skilled cardiologists probe heart sounds by electronic stethoscope through human ears, but interpretation of heart sounds is a very special skill which is quite difficult to teach in a structured way. Because of this reason, automatic heart sound analysis in computer systems would be very helpful for medical staff. This paper presents a complete heart sound analysis system covering from the segmentation of beat cycles to the final determination of heart conditions. The kernels of heart beat cycle segmentation and recognition are based on autocorrelation, short-time Fourier transform, and support vector machines. The experiments are done with a public heart sound database released by Texas Heart Institute, with very promising recognition rate achieved.