Experiment to Classify Autism through Brain MRI Analysis
B.J. Bipin Nair1, N. Shobha Rani2, S. Saikrishna3, C. Adith4
1B.J. Bipin Nair, Department of Computer Science, Amrita School of Arts and Sciences, Amrita Vishwa Vidyapeetham, Coimbatore (Tamil Nadu), India.
2N. Shobha Rani, Department of Computer Science, Amrita School of Arts and Sciences, Amrita Vishwa Vidyapeetham, Coimbatore (Tamil Nadu), India.
3S. Saikrishna, Department of Computer Science, Amrita School of Arts and Sciences, Amrita Vishwa Vidyapeetham, Coimbatore (Tamil Nadu), India.
4C. Adith, Department of Computer Science, Amrita School of Arts and Sciences, Amrita Vishwa Vidyapeetham, Coimbatore (Tamil Nadu), India.
Manuscript received on 03 June 2019 | Revised Manuscript received on 28 June 2019 | Manuscript Published on 04 July 2019 | PP: 383-386 | Volume-8 Issue-1S4 June 2019 | Retrieval Number: A10690681S419/2019©BEIESP
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: Medical image processing technologies have been experiencing rapid improvement and changes in the earlier few years to support the medical industry. In this work an automated cognitive system is developed in demand to classify the ASD from the Brain MRI. MRI analysis provides a dominant tool for studying brain structural variations in ASD affected individuals. After detecting the ASD, we will predict the causing neurotransmitter pattern which made easy to find Autistic region in the brain. In this research, we using PCA feature extraction technique and naïve Bayesian for classification of autism. First, we using PCA technique to extract feature and classify the MRI image into two labels. The steps involved are Using median and unsharp masking the image is pre-processed in order to remove noise and improve the image. The pre-processed image is segmented in order to extract feature, segmentation is executed using Otsu segmentation technique. The white matter region is segmented and the feature is extracted using PCA technique. The features like Mean, RMS, SD, energy, homogeneity features are extracted and classify the image based on the extracted feature using PCA technique. We conclude that the classification of ASD is capable to make clear some of the contradictions in the literature.
Keywords: ASD-Autism Spectrum Disorder, SVM-Support Vector Machine, ABIDE-Autism Brain Imaging Data Exchange.
Scope of the Article: Software Analysis, Design and Modelling