MyMedR

Displaying all 4 publications

Abstract:

Sort:

Fulltext Dual-tree complex wavelet packet transform for voice pathology analysis

Farah Nazlia Che Kassim, Muthusamy, Hariharan, Vijean, Vikneswaran, Zulkapli Abdullah, Rokiah Abdullah

Pertanika Journal of Science & Technology, 2020;28(3):839-858.
MyJurnal

Voice pathology analysis has been one of the useful tools in the diagnosis of the pathological voice, as the method is non-invasive, inexpensive, and can reduce the time required for the analysis. This paper investigates feature extraction based on the Dual-Tree Complex Wavelet Packet Transform (DT-CWPT) using energy and entropy measures tested with two classifiers, k-Nearest Neighbors (k-NN) and Support Vector Machine (SVM). Massachusetts Eye and Ear Infirmary (MEEI) voice disorders database and Saarbruecken Voice Database (SVD) were used. Five datasets of voice samples were used from these databases, including normal and abnormal samples, Cysts, Vocal Nodules, Polyp, and Paralysis vocal fold. To the best of the authors’ knowledge, very few studies were done on multiclass classifications using specific pathology database. File-based and frame-based investigation for two-class and multiclass were considered. In the two-class analysis using the DT-CWPT with entropies, the classification accuracy of 100% and 99.94% was achieved for MEEI and SVD database respectively. Meanwhile, the classification accuracy for multiclass analysis comprised of 99.48% for the MEEI database and 99.65% for SVD database. The experimental results using the proposed features provided promising accuracy to detect the presence of diseases in vocal fold.
Fulltext Real and complex wavelet transform approaches for Malaysian speaker and accent recognition

Rokiah Abdullah, Muthusamy, Hariharan, Vijean, Vikneswaran, Zulkapli Abdullah, Farah Nazlia Che Kassim

Pertanika Journal of Science & Technology, 2019;27(2):737-752.
MyJurnal

A new approach for speaker and accent recognition based on wavelets, namely Discrete Wavelet Packet (DWPT), Dual Tree Complex Wavelet Packet Transform (DT- CWPT) and Wavelet Packet Transform (WPT) based non-linear features are investigated. The results are compared with conventional MFCC and LPC features. k-Nearest Neighbors (k-NN), Support Vector Machine (SVM) and Extreme Learning Machine (ELM) classifier are used to quantify the speaker and accent recognition rate. The database for the research was developed using English digits (0~9) and Malay words. The highest accuracy for speaker recognition obtained is 93.54% while for accent recognition; it is 95.86% using Malay words. Combination of features for speaker recognition is obtained from ELM classifier is 98.68 % and for accent recognition is 98.75 % using Malay words.
Empirical Wavelet Transform Based Features for Classification of Parkinson's Disease Severity

Oung QW, Muthusamy H, Basah SN, Lee H, Vijean V

J Med Syst, 2017 Dec 29;42(2):29.
PMID: 29288342 DOI: 10.1007/s10916-017-0877-2

Parkinson's disease (PD) is a type of progressive neurodegenerative disorder that has affected a large part of the population till now. Several symptoms of PD include tremor, rigidity, slowness of movements and vocal impairments. In order to develop an effective diagnostic system, a number of algorithms were proposed mainly to distinguish healthy individuals from the ones with PD. However, most of the previous works were conducted based on a binary classification, with the early PD stage and the advanced ones being treated equally. Therefore, in this work, we propose a multiclass classification with three classes of PD severity level (mild, moderate, severe) and healthy control. The focus is to detect and classify PD using signals from wearable motion and audio sensors based on both empirical wavelet transform (EWT) and empirical wavelet packet transform (EWPT) respectively. The EWT/EWPT was applied to decompose both speech and motion data signals up to five levels. Next, several features are extracted after obtaining the instantaneous amplitudes and frequencies from the coefficients of the decomposed signals by applying the Hilbert transform. The performance of the algorithm was analysed using three classifiers - K-nearest neighbour (KNN), probabilistic neural network (PNN) and extreme learning machine (ELM). Experimental results demonstrated that our proposed approach had the ability to differentiate PD from non-PD subjects, including their severity level - with classification accuracies of more than 90% using EWT/EWPT-ELM based on signals from motion and audio sensors respectively. Additionally, classification accuracy of more than 95% was achieved when EWT/EWPT-ELM is applied to signals from integration of both signal's information.
Improved binary dragonfly optimization algorithm and wavelet packet based non-linear features for infant cry classification

Hariharan M, Sindhu R, Vijean V, Yazid H, Nadarajaw T, Yaacob S, et al.

Comput Methods Programs Biomed, 2018 Mar;155:39-51.
PMID: 29512503 DOI: 10.1016/j.cmpb.2017.11.021

BACKGROUND AND OBJECTIVE: Infant cry signal carries several levels of information about the reason for crying (hunger, pain, sleepiness and discomfort) or the pathological status (asphyxia, deaf, jaundice, premature condition and autism, etc.) of an infant and therefore suited for early diagnosis. In this work, combination of wavelet packet based features and Improved Binary Dragonfly Optimization based feature selection method was proposed to classify the different types of infant cry signals.
METHODS: Cry signals from 2 different databases were utilized. First database contains 507 cry samples of normal (N), 340 cry samples of asphyxia (A), 879 cry samples of deaf (D), 350 cry samples of hungry (H) and 192 cry samples of pain (P). Second database contains 513 cry samples of jaundice (J), 531 samples of premature (Prem) and 45 samples of normal (N). Wavelet packet transform based energy and non-linear entropies (496 features), Linear Predictive Coding (LPC) based cepstral features (56 features), Mel-frequency Cepstral Coefficients (MFCCs) were extracted (16 features). The combined feature set consists of 568 features. To overcome the curse of dimensionality issue, improved binary dragonfly optimization algorithm (IBDFO) was proposed to select the most salient attributes or features. Finally, Extreme Learning Machine (ELM) kernel classifier was used to classify the different types of infant cry signals using all the features and highly informative features as well.
RESULTS: Several experiments of two-class and multi-class classification of cry signals were conducted. In binary or two-class experiments, maximum accuracy of 90.18% for H Vs P, 100% for A Vs N, 100% for D Vs N and 97.61% J Vs Prem was achieved using the features selected (only 204 features out of 568) by IBDFO. For the classification of multiple cry signals (multi-class problem), the selected features could differentiate between three classes (N, A & D) with the accuracy of 100% and seven classes with the accuracy of 97.62%.
CONCLUSION: The experimental results indicated that the proposed combination of feature extraction and selection method offers suitable classification accuracy and may be employed to detect the subtle changes in the cry signals.