Transformer-Driven Deep Learning Framework for Brain Tumour Detection

Abstract

Brain tumours are among the most life-threatening neurological disorders, requiring accurate and timely diagnosis for effective treatment planning and improved patient outcomes. Magnetic Resonance Imaging (MRI) has become the primary imaging modality for detecting and characterizing brain tumours due to its superior soft-tissue contrast and non-invasive nature. However, manual analysis of MRI scans is time-consuming, subjective, and highly dependent on radiological expertise. Conventional machine learning and convolutional neural network (CNN)-based approaches have demonstrated promising performance in automated brain tumour detection, yet they often struggle to capture long-range spatial dependencies and complex tumour characteristics. Recent advancements in Transformer architectures have introduced powerful self-attention mechanisms capable of learning global contextual information from medical images. This research proposes a Transformer-Driven Deep Learning Framework for Brain Tumour Detection (TDDLF-BTD) that integrates MRI preprocessing, intelligent feature extraction, transformer-based attention learning, tumour classification, and decision optimization