Brain tumor detection using deep learning approaches

Computer Engineering

Brain tumors are one of the most critical neurological disorders requiring early diagnosis for effective treatment. The advancement in deep learning has significantly enhanced medical image analysis, providing automated and accurate detection solutions. This paper proposes a novel approach utilizing Graph Neural Networks (GNNs) and Multi-Scale Hybrid Transformers to detect and classify brain tumors from MRI scans. The GNNs capture spatial and relational dependencies within MRI images, while the hybrid transformer network extracts multi-scale features for improved classification accuracy. When compared to traditional deep learning methods, the suggested approach significantly improves accuracy, precision, recall, and F1-score when tested on a common benchmark dataset. The real-time implementation of the developed model in a clinical setting is also discussed, highlighting its potential for aiding radiologists in tumor detection and classification.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.