A Comparative Approach of CNN Versus Auto Encoders to Classify the Autistic Disorders from Brain MRI
B.J. Bipin Nair1, C. Adith2, S. Saikrishna3
1B.J. Bipin Nair, Department of Computer Science, Amrita School of Arts and Sciences, Mysore (Karnataka), India.
2C. Adith, Department of Computer Science, Amrita School of Arts and Sciences, Mysore (Karnataka), India.
3S. Saikrishna, Department of Computer Science, Amrita School of Arts and Sciences, Mysore (Karnataka), India.
Manuscript received on 22 April 2019 | Revised Manuscript received on 01 May 2019 | Manuscript Published on 08 May 2019 | PP: 144-149 | Volume-7 Issue-5S3 February 2019 | Retrieval Number: E11270275S19/19©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: In the present study, we are going to apply the deep learning methods to create a trained predictive model to recognize the neuro developmental diseases such as ASD, FASD. Deep learning is a dominant ML technique in classification. It will extract all kind (low to high) feature from a digital image. Classifying medical images tends to develop a predictive model in order to predict the neuro developmental disease. Classification of medical data for a medical condition for example ASD always a challenging task and selecting the important feature is also a difficult task. Using the deep learning techniques, we can successfully classify the MRI data of the neurodegenerative disease. Feature extracted by Convolution Neural Network and classification using the same advise a most robust method of classifying medical data especially MRI images. Stacked Auto encoders another better way for the feature extraction and prediction clinical data.
Keywords: ASD-Autism Spectrum Disorder, ABIDE-Autism Brain Imaging Data Exchange.
Scope of the Article: Data Mining and Warehousing