Brain Tumor Detection and Classification Using Deep Learning Approaches

Brain tumors account for having the lowest survival rate and being the most fatal cancer in the world. This makes detection and early diagnosis of the same to be of utmost importance. Classification of tumors depends on the shape, size, texture, and location. Magnetic Resonance Images (MRI) prove to...

Full description

Saved in:
Bibliographic Details
Published in2023 4th International Conference for Emerging Technology (INCET) pp. 1 - 6
Main Authors G, Ankitha, J, Hafsa Tuba, J, Akhilesh, Bhanu, Archana, Ig, Naveen
Format Conference Proceeding
LanguageEnglish
Published IEEE 26.05.2023
Subjects
Brain modeling
Brain tumor
Classification
Classification algorithms
CNN
Convolutional neural networks
Deep learning
Magnetic resonance
Magnetic resonance imaging
MRI
Shape
Online AccessGet full text

Cover

More Information
Summary:Brain tumors account for having the lowest survival rate and being the most fatal cancer in the world. This makes detection and early diagnosis of the same to be of utmost importance. Classification of tumors depends on the shape, size, texture, and location. Magnetic Resonance Images (MRI) prove to be the most effective technique for distinguishing tumors. The main aim of the proposed work is to capture the distribution of unique features from the input MRI dataset images. These images are then synthesized using a generative model which classifies the dataset to detect the presence of a tumour in brain. Deep learning algorithms such as Convolutional Neural Network (CNN) help in classification of the different tumours. The proposed model is experimentally evaluated on three datasets. The suggested methods provide for the successful comparison and convincing performance. An accuracy of 98.02% was achieved with ResNet50 architecture and 98.32% with Xception architecture.
DOI:10.1109/INCET57972.2023.10169933