This research introduces a novel deep learning architecture that significantly improves brain tumor segmentation accuracy by combining Xception CNN with self-attention mechanisms.

Key Innovations:

• Integration of dot-product self-attention modules with Xception backbone for superior feature extraction
• Utilizes multiple MRI sequences (T1CE, T2, FLAIR) for comprehensive tumor analysis
• Trained on the BraTS 2021 dataset, incorporating diverse tumor presentations
• Architecture specifically optimized for medical imaging challenges

Clinical Impact: Accurate tumor segmentation directly impacts treatment planning and patient outcomes by enabling precise tumor boundary identification, potentially reducing surgical complications and improving targeted therapies.

Attention Xception UNet (AXUNet): A Novel Combination of CNN and Self-Attention for Brain Tumor Segmentation