Fed-Brain: Privacy-Preserving Brain Tumour Classification Using Federated Learning
Abstract
Brain tumour detection is a critical challenge in medical imaging, requiring accurate and efficient classification methods. Traditional deep learning approaches rely on centralized data storage, which poses significant privacy risks, especially in the healthcare domain. In this study, we propose a federated learning-based brain tumour classification system that enables collaborative model training across multiple healthcare institutions while ensuring data privacy. The model is trained on MRI scan datasets distributed across different clients without the need to share patient data. The proposed system utilizes a Convolutional Neural Network (CNN) architecture, trained in a federated learning framework, where each hospital performs local training and contributes to a globally aggregated model. The experiments were conducted over 50 communication rounds, achieving a high classification accuracy of 91.5%, outperforming traditional centralized deep learning methods. The model's performance was evaluated using key metrics such as precision, recall, and F1-score, demonstrating robust classification across tumour types, including Pituitary tumour, Meningioma, Glioma, and No tumour cases. The results indicate that federated learning enhances model generalization by leveraging diverse MRI scans from multiple sources while preserving patient confidentiality. Compared to centralized learning, federated learning not only improves classification accuracy but also addresses critical…
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.