Delineation of Ischemic Lesion from Brain MRI using Symmetric Bit Plane Pattern and Curvelet Co-occurrence Matrix
Karthik. R1, R. Menaka2
1Karthik. R, Department of Electronics Engineering, VIT Chennai Campus, Chennai (Tamil Nadu), India.
2R.Menaka, Department of Electronics Engineering, VIT Chennai Campus, Chennai (Tamil Nadu), India.
Manuscript received on 07 March 2019 | Revised Manuscript received on 20 March 2019 | Manuscript published on 30 March 2019 | PP: 201-206 | Volume-8 Issue-5, March 2019 | Retrieval Number: E2959038519/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Developing a precise segmentation algorithm to delineate ischemic lesion from brain MRI is a challenging research issue in the field of medical image analysis and neuroradiology. These lesions are generally complex in nature and exhibit heterogeneity in their intensity profile and morphological properties. To address these challenges, a novel segmentation algorithm using Symmetric Bit Plane Pattern analysis is presented in this work. Unlike the classical segmentation algorithms which fail to extract the region of interest in the presence of scattered structures with intensity in-homogeneities, the proposed segmentation algorithm considers the left-right symmetricity property of the brain for better estimation of segmentation parameters. An adaptive filter function is designed based on the gray level profile of the brain tissues to segment the intended region of interest. Once the region of interest is delineated, multi scale co-occurrence matrix based features in Curvelet space are extracted and its significance in detection of ischemic lesion is highlighted. Finally, Support Vector Machine is used to train the learning model for classification. Experimental results of the proposed work have obtained better classification accuracy of 98.8%.
Keyword: Curvelet Transform, Ischemic Lesion, MRI.
Scope of the Article: Pattern Recognition