Title: Feature selection based on binary particle swarm optimisation and neural networks for pathological voice detection

Authors: Taciana Araújo De Souza; Vinícius J.D. Vieira; Micael Andrade De Souza; Suzete E.N. Correia; Silvana C. Costa; Washington C. De Almeida Costa

Addresses: Postgraduate Program in Electrical Engineering, Federal University of Campina Grande, Campina Grande, Brazil ' Postgraduate Program in Electrical Engineering, Federal University of Campina Grande, Campina Grande, Brazil ' Academic Unity of Industry, Federal Institute of Education, Science and Technology of Paraíba, João Pessoa, Brazil ' Academic Unity of Industry, Federal Institute of Education, Science and Technology of Paraíba, João Pessoa, Brazil ' Academic Unity of Industry, Federal Institute of Education, Science and Technology of Paraíba, João Pessoa, Brazil ' Academic Unity of Industry, Federal Institute of Education, Science and Technology of Paraíba, João Pessoa, Brazil

Abstract: In this work, 52 Haralick texture features, extracted from two-dimensional wavelet coefficients of speech signals from recurrence plots (RPs) pathologies are used for pathological voice discrimination. Here, three pathologies are considered for analysis: vocal fold paralysis, edema and nodules. For feature selection, a binary particle swarm optimisation (PSO) algorithm using multilayer perceptron (MLP) neural network with cross validation is employed. The adopted fitness function is based on the maxima average accuracy rate. Statistical tests for individual measures were made and their results show statistical significance for several employed measures. The measures were combined and the more relevant ones based on the highest accuracy were selected by the PSO. The comparison with and without PSO by applying the statistical test of mean difference showed that the PSO use increased the accuracy rates. Furthermore, the PSO use reduced the amount of features for almost half of all initially used.

Keywords: detection of laryngeal pathologies; acoustic analysis; recurrence plots; Haralick texture features; particle swarm optimisation; PSO; wavelet transform.

DOI: 10.1504/IJBIC.2018.091234

International Journal of Bio-Inspired Computation, 2018 Vol.11 No.2, pp.91 - 101

Received: 03 Feb 2016
Accepted: 31 May 2016
Published online: 17 Apr 2018 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article