[This article belongs to Volume - 54, Issue - 02]
Gongcheng Kexue Yu Jishu/Advanced Engineering Science
Journal ID : AES-17-11-2022-447

Dr.K.Kalaivani#1 , Dr Ganapriya K*2 , Dr PoobalanA#3, Mr.N.Balamurugan *4

Abstract :

Brain Tumour analysis without human involvement is a crucial field of study. Convolutional neural networks, on the other hand, can help with this (CNNs). They have excelled at solving computer vision and other challenges such as visual object recognition, detection, and segmentation. It aids in the diagnosis of brain tumours by improving brain pictures utilising segmentation algorithms that are extremely resistant to noise and cluster size sensitivity issues, as well as the automated area of Interest (ROI) detection. One of the key arguments for using CNNs is that they have a high level of accuracy and do not require human feature extraction. Detecting a brain tumour and correctly identifying its kind is a difficult undertaking. Because of its widespread use in image recognition, CNN performs better than others. Because a human-assisted manual categorization might result in erroneous prediction and diagnosis, brain tumour segmentation is one of the most important and difficult challenges in the field of medical image processing. Furthermore, it is a difficult process when there is a huge amount of data to assist. Because brain tumours have such a wide range of appearances and because tumours and normal tissues are so comparable, extracting tumour areas from pictures becomes difficult.