eng Oriental Scientific Publishing Company Biomedical and Pharmacology Journal 0974-6242 2025-02-20 18 March Spl Edition 99 119 10.13005/bpj/3076 63926 article Simran 1 Shiva Mehta 1 Vinay Kukreja 1 Ayush Dogra 1 Tejinder Pal Singh Brar 2 Centre for Research Impact and Outcome, Chitkara University Institute of Engineering and Technology, Chitkara University, Rajpura, Punjab, India M.M. Institute of Computer Technology and Business Management, Maharishi Markandeshwar (Deemed to be University), Mullana, Ambala, Haryana Brain tumor identification through Bio-medical magnetic resonance imaging (MRI) presents a critical challenge in diagnostic imaging, where high accuracy is essential for informed treatment planning. Traditional methods face limitations in segmentation precision, leading to increased misdiagnosis risks. This study introduces a hybrid deep-learning model integrating a Vision Transformer (ViT) and Capsule Network (CapsNet) to improve brain tumor classification and segmentation accuracy. The model aims to enhance sensitivity and specificity in tumor categorization. Utilising the BRATS2020 dataset, which comprises 6,000 MRI scans across four classes (meningioma, glioma, pituitary tumor, and no tumor), the dataset was divided into an 80-20 training-testing split. Data pre-processing included scaling, normalization, and feature augmentation to improve model robustness. The hybrid ViT-CapsNet model was assessed alongside individual ViT and CapsNet performance using accuracy, precision, recall, F1-score, and AUC-ROC metrics. The hybrid model achieved an accuracy of 90%, precision of 90%, recall of 89%, and an F1-score of 89.5%, outperforming individual models. Data augmentation yielded a 4-5% improvement in accuracy across tumor types, with notable gains for gliomas and pituitary tumors. Unlike prior methods, achieving 88% accuracy, our hybrid model demonstrates superior performance with an accuracy of 90%. This hybrid approach offers a promising solution for more accurate brain tumor detection. Future research could explore refining feature fusion techniques, integrating advanced interpretability methods, and expanding the model’s application across various clinical imaging environments. https://biomedpharmajournal.org/vol18marchspledition/hybrid-vit-capsnet-framework-for-brain-tumor-diagnosis-using-biomedical-mri/ Brain Tumor Segmentation Bio-Medical MRI Imaging Classification Capsule Network (CapsNet) Vision Transformer (ViT) Hybrid Model