Stroke is one of the principal causes of death, since it has limited treatments. Early detection of stroke at acute stage is essential for the emergency treatment to the stroke affected patients. This work proposes a novel feature selection using evolutionary algorithm and Hybrid Multi-Layer Perception (HMLP) classifier to label the input brain Diffusion Weighted Imaging (DWI) image into Stroke and Non-Stroke. Features are obtained from the Region of Interest (ROI) using Independent Component Analysis (ICA). Itis proposed to use Genetic Algorithm (GA) to choose the best set of features for classification. HMLP classifier is trained using the selected features to label the input images. Theweights of MLP are optimized by the combination of GA and Local Search (LS). The performance of the proposed HMLP classifier is evaluated usingclassification accuracy, precision and recall. Results show improvements in the classification accuracy using the proposed method.
Cerebral Infraction, Stroke Classification, Independent Component Analysis, MLP Classifier, Genetic Algorithm, Hybrid Optimization
Share This Article
© The Author(s) 2015. Open Access. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License which permits unrestricted use, sharing, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.