A Comparative Analysis of Vision Transformers and Transfer Learning for Brain Tumor Classification

Authors : Ahmet Solak

Pages : 558-572

Doi:10.29130/dubited.1521340

View : 38 | Download : 26

Publication Date : 2025-01-30

Article Type : Research Paper

Abstract :Accurate brain tumor classification is crucial in neuro-oncology for guiding treatment plans and improving patient outcomes. Leveraging the potential of Vision Transformers (ViTs), this study investigates their efficacy in binary classification of brain tumors using magnetic resonance (MR) images, comparing them to CNN-based models such as VGG16, VGG19, and ResNet50. Comprehensive evaluation using accuracy, precision, recall, and F1-score reveals ViTs’ superior performance, achieving 92.59% accuracy, surpassing VGG16 (85.19%), VGG19 (74.04%), and ResNet50 (88.89%). These findings highlight ViTs as a transformative tool for clinical adoption, enhancing diagnostic accuracy and patient care in neuro-oncology.
Keywords : Beyin tümörü sınıflandırması, Görü dönüştürücü, Derin öğrenme, Transfer öğrenme, Tıbbi görüntüleme