e-ISSN:0976-5166
p-ISSN:2231-3850


INDIAN JOURNAL OF COMPUTER SCIENCE AND ENGINEERING

Call for Papers 2021

Aug 2021 - Volume 12, Issue 4
Deadline: 15 Jul 2021
Publication: 20 Aug 2021

Oct 2021 - Volume 12, Issue 5
Deadline: 15 Sep 2021
Publication: 25 Oct 2021

More

Indexed in

IJCSE Indexed in Scopus

ABSTRACT

Title : DEVELOPMENT OF A CONVOLUTIONAL NEURAL NETWORK MODEL FOR BRAIN ABNORMALITY CLASSIFICATION
Authors : Hetal Barad, Atul Patel
Keywords : Brain Abnormality; Deep learning; CNN; Multi-class Classification
Issue Date : Nov-Dec 2020
Abstract :
Medical image includes various image modalities and procedures, which helps in patient treatment and diagnosis. Many abnormalities are occurred in human brain. Among them, with the frequent occurrence of the tumor disease and its complexity, brain tumor has become important research topic in the medical field. The brain tumor diagnosis is mainly based on the medical imaging data analysis from the images of brain tumor. However, the accurate analysis tumor medical image is highly dependent on the knowledge, experience and eye strain of a doctor. with the advancement in the field of artificial intelligence along with computer vision provided opportunity for brain tumor identification more precisely without explicit feature extraction and human intervention. Deep Learning, the recent advancement in the field of artificial intelligence, makes system enable learning features automatically for the given problem without human expertise. In deep learning, convolutional neural network (CNN) is a promising architecture that applied to visual imagery. The aim of this paper is to develop and propose a CNN model for classifying brain abnormalities by considering one of the abnormality i.e. brain tumor. For that, the different benchmark CNN models are applied and evaluated. The results obtained from these models are analysed that motivates us to develop the custom CNN model for the said problem. From the performance evaluation, it is found that, the proposed model outperforms with good accuracy.
Page(s) : 793-800
ISSN : 0976-5166
Source : Vol. 11, No.6
PDF : Download
DOI : 10.21817/indjcse/2020/v11i6/201106058