Hierarchical Diagnosis of Vocal Fold Disorders

被引:0
作者
Nikkhah-Bahrami, Mansour [1 ]
Ahmadi-Noubari, Hossein [2 ]
Aghazadeh, Babak Seyed [1 ]
Heris, Hossein Khadivi [1 ]
机构
[1] Univ Tehran, Dept Mech Engn, Tehran, Iran
[2] Univ Tehran, Elect & Comp Engn Dept, Tehran, Iran
来源
ADVANCES IN COMPUTER SCIENCE AND ENGINEERING | 2008年 / 6卷
关键词
Hierarchical classification; wavelet packets; nonlinear signal analysis; vocal fold disorders;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper explores the use of hierarchical structure for diagnosis of vocal fold disorders. The hierarchical structure is initially used to train different second-level classifiers. At the first level normal and pathological signals have been distinguished. Next, pathological signals have been classified into neurogenic and organic vocal fold disorders. At the final level, vocal told nodules have been distinguished from polyps in organic disorders category. For feature selection at each level of hierarchy. the reconstructed signal at each wavelet packet decomposition sub-band in 5 levels of decomposition with mother wavelet of (db10) is used to extract the nonlinear features of self-similarity and approximate entropy. Also. wavelet packet coefficients are used to measure energy and Shannon entropy features at different spectral sub-bands. Davies-Bouldin criterion has been employed to find the most discriminant features. Finally. support vector machines have been adopted as classifiers at each level of hierarchy resulting in the diagnosis accuracy of 92%.
引用
收藏
页码:897 / +
页数:2
相关论文
共 7 条
[1]  
AKAY M, 2000, NONLINEAR BIOMEDICAL, V2, P72
[2]  
[Anonymous], 1994, DIS VOIC DAT VERS 1
[3]   Trying different wavelets on the search for voice disorders sorting [J].
Guido, RC ;
Pereira, JC ;
Fonseca, E ;
Sanchez, FL ;
Vieira, LS .
PROCEEDINGS OF THE THIRTY-SEVENTH SOUTHEASTERN SYMPOSIUM ON SYSTEM THEORY, 2005, :495-499
[4]   Analysis of vocal disorders in a feature space [J].
Matassini, L ;
Hegger, R ;
Kantz, H ;
Manfredi, C .
MEDICAL ENGINEERING & PHYSICS, 2000, 22 (06) :413-418
[5]   Feature analysis of pathological speech signals using local discriminant bases technique [J].
Umapathy, K ;
Krishnan, S .
MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING, 2005, 43 (04) :457-464
[6]   Meaningful MRA initialization for discrete time series [J].
Veitch, D ;
Taqqu, MS ;
Abry, P .
SIGNAL PROCESSING, 2000, 80 (09) :1971-1983
[7]   Classification of surface EMG signals using optimal wavelet packet method based on Davies-Bouldin criterion [J].
Wang, Gang ;
Wang, Zhizhong ;
Chen, Weiting ;
Zhuang, Jun .
MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING, 2006, 44 (10) :865-872