The Federated Tumor Segmentation Challenge 2027

International challenges have become the standard for validation of biomedical image analysis methods. We argue, though, that the actual performance even of the winning algorithms on ``real-world`` clinical data often remains unclear, as the data included in these challenges are usually acquired in very controlled settings at few institutions. The seemingly obvious solution of just collecting increasingly more data from more institutions in such challenges does not scale well due to privacy and ownership hurdles. We build upon the first-ever proposed federated learning challenge, Federated Tumor Segmentation (FeTS) 2021 and its follow-up in 2022 (published in Nature Communications), and in 2024 (published in MELBA), intending to address these hurdles, for the creation of tumor segmentation models. Specifically, the FeTS 2027 challenge will use clinically acquired, multi-institutional multi-parametric magnetic resonance imaging (mpMRI) scans from the BraTS 2025 Lighthouse challenge. The FeTS 2027 challenge focuses on innovating at the level of federated aggregation where locally trained models combine to form the consensus model for the segmentation of intrinsically heterogeneous (in appearance, shape, and histology) brain tumors, namely gliomas and meningiomas both in the pre-operative and post-operative setting. Compared to the BraTS 2025 Lighthouse challenge, the ultimate goal of the FeTS challenge is the creation of a consensus segmentation model that has gained knowledge
from data of multiple institutions without pooling their data together (i.e., by retaining the data within each institution).

Since the conception of the FeTS 2021 and its conduction of FeTS 2022 and 2024, Federated Learning has matured to a more active research field in biomedical AI. What separates FeTS 2027 challenge from those of previous years is that in 2027 we plan to further broaden the challenge in 3 ways: 1) We completely change the software infrastructure of the  challenge and move from the previously custom code and OpenFL to NVIDIA FLARE, an enterprise-grade federated learning framework that streamlines development and supports an efficient transition from research prototypes to realworld deployment; 2) following the success of the BraTS 2025 lighthouse challenge, FeTS 2027 moves from purely preoperative MRI brain tumor scans to a combination of both pre-operative and post-operative settings, including resection
cavities; 3) the generalizability of the participants' aggregation methods will be evaluated beyond the challenge's segmentation task that the participants have access to, to a hidden (to the participants) task. This added evaluation will be a significant part of the final challenge paper which will provide detailed meta-analysis and provide further
insights about the developed aggregation methods. For fairness, since the participants will only have access to the segmentation data, this added evaluation on a new task will not be considered for the ranking. We will however announce performance on this hidden task during the challenge results presentation at MICCAI.