Table of Contents Author Guidelines Submit a Manuscript
Computational Intelligence and Neuroscience
Volume 2017 (2017), Article ID 4205141, 11 pages
Research Article

Pathological Brain Detection Using Weiner Filtering, 2D-Discrete Wavelet Transform, Probabilistic PCA, and Random Subspace Ensemble Classifier

1Department of Information and Communication Engineering, Chosun University, 309 Pilmun-Daero, Dong-Gu, Gwangju 61452, Republic of Korea
2School of Electrical Engineering, Korea University, 145 Anam-ro, Sungbuk-gu, Seoul 02841, Republic of Korea

Correspondence should be addressed to Goo-Rak Kwon

Received 10 May 2017; Revised 7 August 2017; Accepted 23 August 2017; Published 3 October 2017

Academic Editor: George A. Papakostas

Copyright © 2017 Debesh Jha et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


Accurate diagnosis of pathological brain images is important for patient care, particularly in the early phase of the disease. Although numerous studies have used machine-learning techniques for the computer-aided diagnosis (CAD) of pathological brain, previous methods encountered challenges in terms of the diagnostic efficiency owing to deficiencies in the choice of proper filtering techniques, neuroimaging biomarkers, and limited learning models. Magnetic resonance imaging (MRI) is capable of providing enhanced information regarding the soft tissues, and therefore MR images are included in the proposed approach. In this study, we propose a new model that includes Wiener filtering for noise reduction, 2D-discrete wavelet transform (2D-DWT) for feature extraction, probabilistic principal component analysis (PPCA) for dimensionality reduction, and a random subspace ensemble (RSE) classifier along with the -nearest neighbors (KNN) algorithm as a base classifier to classify brain images as pathological or normal ones. The proposed methods provide a significant improvement in classification results when compared to other studies. Based on cross-validation (CV), the proposed method outperforms 21 state-of-the-art algorithms in terms of classification accuracy, sensitivity, and specificity for all four datasets used in the study.