MRI Brain Lump Based on Consistency Feature and Classification using Neural Network
Mohanraj.R1, Ramya2, Hema3
1Mohanraj R, Assistant Professor, Department of Electronics and Communication Engineering, Bharath Institute of Higher Education and Research, Chennai (Tamil Nadu), India.
2Ramya, Assistant Professor, Department of Electronics and Communication Engineering, Bharath Institute of Higher Education and Research, Chennai (Tamil Nadu), India.
3Hema, Assistant Professor, Department of Electronics and Communication Engineering, Bharath Institute of Higher Education and Research, Chennai (Tamil Nadu), India.
Manuscript received on 14 September 2019 | Revised Manuscript received on 23 September 2019 | Manuscript Published on 10 October 2019 | PP: 393-396 | Volume-8 Issue-6S2, August 2019 | Retrieval Number: F11080886S219/19©BEIESP | DOI: 10.35940/ijeat.F1108.0886S219
Open Access | Editorial and Publishing 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: MRI brain tumor image are segmented using texture feature and Artificial Neural Network are used for taxonomy. The proposed system uses ROI, seed selection and cellular automata based grow cut method for segmentation. The selection based on energy and entropy quality of Grey echelon Co-occurrence matrix, then Long run emphasis and Run length non homogeny is compared with Co-occurrence feature to get a feasible seed point from abnormal region. With this seed point cellular automata based grow cut method is proposed for segmenting the tumor region from MRI image. Morphological process is the smoothing process applied on obtained tumor part for highlighting it by removing distortion, noise and coarse region. By means of the Radial basis occupation of Artificial Neural Network which was accuracy, the tumor part is classified into normal, benign and malignant.
Keywords: Feature Extraction ,Grey Level Co-Occurrence Matrix, Run Length Features, Seed Point Selection, Grow Cut Method, Cellular Automata.
Scope of the Article: Network Modelling and Simulation