Discrete wavelet transform and support vector machine applied to pathological voice signals identification
ABSTRACT An algorithm able to classify pathological and normal voice signals based on Daubechies discrete wavelet transform (DWT-db) and support vector machines (SVM) classifier is presented. DWT-db is used for time-frequency analysis giving quantitative evaluation of signal characteristics to identify pathologies in voice signals, particularly nodules in vocal folds, of subjects with different ages for both male and female. After using a linear prediction coefficients (LPC) filter, the signals mean square values of a particular scale from wavelet analysis are entries to a nonlinear least square support vector machine (LS-SVM) classifier, which leads to an adequate larynx pathology classifier which over 95% of classification accuracy.
- SourceAvailable from: Thomas Elssner[Show abstract] [Hide abstract]
ABSTRACT: Automatic classification of high-resolution mass spectrometry data has increasing potential to support physicians in diagnosis of diseases like cancer. The proteomic data exhibit variations among different disease states. A precise and reliable classification of mass spectra is essential for a successful diagnosis and treatment. The underlying process to obtain such reliable classification results is a crucial point. In this paper such a method is explained and a corresponding semi automatic parameterization procedure is derived. Thereby a simple straightforward classification procedure to assign mass spectra to a particular disease state is derived. The method is based on an initial preprocessing stage of the whole set of spectra followed by the bi-orthogonal discrete wavelet transform (DWT) for feature extraction. The approximation coefficients calculated from the scaling function exhibit a high peak pattern matching property and feature a denoising of the spectrum. The discriminating coefficients, selected by the Kolmogorov–Smirnov test are finally used as features for training and testing a support vector machine with both a linear and a radial basis kernel. For comparison the peak areas obtained with the it ClinProt-System 1  were analyzed using the same support vector machines. The introduced approach was evaluated on clinical MALDI-MS data sets with two classes each originating from cancer studies. The cross validated error rates using the wavelet coefficients where better than those obtained from the peak areas2.Computing and Visualization in Science 03/2009; 12(4):189-199.
Conference Paper: Different mother wavelets and pathological voice[Show abstract] [Hide abstract]
ABSTRACT: Early diagnosis of different maladies and pathologies of human vocal system using noninvasive methods and diverse signal processing technics is a problem that is particularly considered by biomedical engineering and signal processing researchers, recently. Automatic detection of voice pathology from speech signal is a new topic and has not been progressed enough. An algorithm able to classify two pathological voice signals based on wavelet packets (WP) and Fisher's linear discriminant (FLD) is presented in this research. We use WP and different mother wavelets (Daubechies, Coiflet, and Symmlet) for time-frequency analysis giving quantitative evaluation of signal characteristics to identify pathologies in voice signals of subjects with different ages for both male and female. Choosing Coiflet mother wavelet, we use FLD to find the best tree among Coiflet Wavelet Packet trees. After selecting best features from terminal nodes of the best tree with contribution to genetic algorithm, we apply support vector machines to separate voice pathologies. Applying our algorithm to seperate polyp from some other pathologies we come to much higher conclusions in contrast to previous works that use Daubechies mother wavelet instead of Coiflet mother wavelet (e.g. 92.5% in comparison to 82.5% for separating polyp from adductor spasmodic dysphonia).Computer, Control and Communication, 2009. IC4 2009. 2nd International Conference on; 03/2009