Federated learning framework improves bladder cancer segmentation and classification

Researchers have developed a federated multi-task learning framework to improve bladder tumor segmentation and muscle-invasive bladder cancer (MIBC) classification using T2-weighted MRI. The proposed Swin Hybrid model, which combines ResNet-34 and Swin-Tiny Transformer branches, aims to overcome challenges posed by data privacy and imaging variability across institutions. Experiments on the FedBCa dataset demonstrated that the Swin Hybrid architecture, particularly with Geo+Elastic augmentation under federated training, achieved a Dice Similarity Coefficient (DSC) of 0.8100 for segmentation and a patient-level Area Under the Curve (AUC) of 0.8931 for classification. AI

IMPACT This research demonstrates a viable approach for collaborative medical image analysis across institutions without compromising patient data privacy.

RANK_REASON The cluster contains an academic paper detailing a new machine learning architecture and experimental results. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• FedBCa dataset
• Geo+Elastic augmentation
• muscle-invasive bladder cancer
• non-muscle invasive bladder neoplasm
• ResNet-34
• Sachin Dudda Nagaraju
• Swin Hybrid
• Swin-Tiny Transformer

• paper
• model release
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →