Federated Multi-Task Learning for Bladder Tumor Segmentation and MIBC Classification Using a Hybrid CNN-Transformer Architecture
2026-08-31 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors developed a way for computers to find bladder tumors and decide if they have grown into muscle using MRI images from different hospitals without sharing patient data. They combined two types of computer models—one that focuses on local details and another that looks at the whole picture—to improve accuracy. Their method also uses tumor location info to help with classification. They tested different ways to make the model stronger against differences in images from various hospitals, and the best results came from a specific training setup. This shows it’s possible to analyze bladder cancer images from multiple centers while keeping patient data private.
bladder tumor segmentationmuscle invasion bladder cancer (MIBC)T2-weighted MRIfederated learningmulti-task learningSwin TransformerResNet-34data augmentationdice similarity coefficient (DSC)area under curve (AUC)
Authors
Malhar Udmale, Divyanshu Dwivedi, Aarohi Dhand, Sachin Dudda Nagaraju, Mayank Rai, Bagesh Kumar
Abstract
Accurate bladder tumor segmentation and assessment of mus- cle invasion from T2-weighted MRI are important for treatment plan- ning, but developing robust models across institutions is challenging be- cause patient data cannot be centrally pooled and imaging characteristics vary across scanners and acquisition protocols. We propose a federated multi-task learning framework for joint bladder tumor segmentation and MIBC/NMIBC classification across four clinical centers. The proposed Swin Hybrid model combines a ResNet-34 branch for local texture and boundary information with a Swin-Tiny Transformer for global anatomi- cal context. A segmentation-guided classification mechanism further uses tumor localization information to support MIBC prediction. We also investigate several augmentation strategies under both centralized and federated training to improve robustness to multi-center variability. Ex- periments on the FedBCa dataset show that the Swin Hybrid provides the best overall balance between segmentation and classification among the evaluated architectures. Under federated training, Geo+Elastic aug- mentation achieved a DSC of 0.8100 and a patient-level AUC of 0.8931, yielding the highest combined score of 0.8474. These results demonstrate that joint segmentation and classification can be effectively performed across multiple institutions using federated training without centralizing patient data.