Automatic diagnosis of septal defects based on tunable-Q wavelet transform of cardiac sound signals
作者：
Highlights：
• We propose a new method for diagnosis of septal defects using TQWT.
• New feature set based on SAMDF derived from TQWT has been proposed.
• The effects of Q and decomposition levels on classification performance have been evaluated.
• Results have been evaluated with classification performance evaluation parameters and ROC graphs.
• Performance has been compared with existing TQWT based method with same datasets.
摘要
•We propose a new method for diagnosis of septal defects using TQWT.•New feature set based on SAMDF derived from TQWT has been proposed.•The effects of Q and decomposition levels on classification performance have been evaluated.•Results have been evaluated with classification performance evaluation parameters and ROC graphs.•Performance has been compared with existing TQWT based method with same datasets.
论文关键词：Septal defects,Cardiac sound signals,Segmentation,Classification,Heart beat cycles,Tunable-Q wavelet transform (TQWT),Sum of average magnitude difference function (SAMDF),Least squares support vector machine (LS-SVM)
论文评审过程：Available online 4 December 2014.
论文官网地址：https://doi.org/10.1016/j.eswa.2014.11.046
