An Early Diagnosis of Brain Tumor Using Fused Transfer Learning

This study aims to develop a system that can classify brain tumors as either benign or malignant. The dataset used in this study consists of 253 MRI images of the brain. To achieve high accuracy in classification, the researchers employed a novel fusion architecture of two deep learning models: ResN...

Full description

Saved in:
Bibliographic Details
Published in2023 International Conference on Business Analytics for Technology and Security (ICBATS) pp. 1 - 5
Main Authors Sajjad, Ghulam, Shoaib Khan, Muhammad Bilal, Ghazal, Taher M., Saleem, Muhammad, Khan, Muhammad Farhan, Wannous, Muhammad
Format Conference Proceeding
LanguageEnglish
Published IEEE 07.03.2023
Subjects
Benign
Brain modeling
Brain Tumor
Deep learning
Inception-V3
Magnetic resonance imaging
Malignant
MRI
ResNet50
Security
Sensitivity
Transfer learning
Tumors
Online AccessGet full text

Cover

More Information
Summary:This study aims to develop a system that can classify brain tumors as either benign or malignant. The dataset used in this study consists of 253 MRI images of the brain. To achieve high accuracy in classification, the researchers employed a novel fusion architecture of two deep learning models: ResNet-50 and Inception-V3. The proposed system was developed using MATLAB, and its performance was evaluated using various metrics such as accuracy, specificity, and sensitivity. The results showed that the proposed system achieved an accuracy of 98.67% on the Kaggle dataset using two different optimizers: ADAM and RMSProp. The system was trained for 10 epochs, and the elapse time for each optimizer was 62.52 and 65.58 minutes, respectively. Overall, the study demonstrates the effectiveness of the proposed fusion architecture in accurately classifying brain tumors. The high accuracy achieved by the proposed system suggests that it could be a valuable tool for clinicians in the diagnosis and treatment of brain tumors.
DOI:10.1109/ICBATS57792.2023.10111263