Scopus Indexed Publications

Paper Details


Title
Comparative evaluation of transfer learning models and Grad-CAM interpretability for brain tumor detection from MRI

Author
Md. Firoz Hasan, Dewan Mamun Raza, Md. Atik Shahriar, Md. Awal Hadi, Md. Hasan Moon, Md. Raisul Islam, Sumaiya Nasrin, Tanvir Ahmed Momin,

Email

Abstract
Brain tumor classification plays an important role in early diagnosis and treatment planning. The current study aims to evaluate and compare the performance of five pre-trained convolutional neural network (CNN) models, namely VGG16, VGG19, MobileNet, Xception, and InceptionV3 using magnetic resonance imaging (MRI) images categorized into glioma, meningioma, pituitary tumor, and no tumor classes. To enhance model performance and address class imbalance, transfer learning and data augmentation techniques were employed. To boost model interpretability, heatmaps of important areas in tumor classification were produced through gradient-weighted class activation mapping (Grad-CAM). MobileNet was the most accurate with 97% and was more precise and more sensitive. The Grad-CAM visualizations showed the models were attending to clinically relevant features, which increased the interpretability. This comparative study demonstrates the effectiveness of the integration of explainable artificial intelligence (XAI) in deep learning pipelines for reliable brain tumor diagnosis.


Keywords
Brain tumor classification; Explainable artificial intelligence; Healthcare artificial intelligence; Medical imaging; Transfer learning

Journal or Conference Name
IAES International Journal of Artificial Intelligence

Publication Year
2026

Indexing
scopus