Table 3 Overview of the number of cases and institutions in the training, validation, and test sets

From: Towards fair decentralized benchmarking of healthcare AI algorithms with the Federated Tumor Segmentation (FeTS) challenge

 

Training

Validation

Test (Task 1)

Test (Task 2)

Source

BraTS21

BraTS21

BraTS21

BraTS21 + FeTS

No. cases

1251

219

570

2625

No. sites

23a

n/a

n/a

32

Access

Public (img, seg)

Public (img)

Organizers

Data owners

  1. The centralized, multi-centric data from the Brain Tumor Segmentation Challenge 2021 (BraTS21)3 is used for benchmarking FL methods (Task 1). Additionally, for Task 2 the testing data is augmented with distributed data from the FeTS initiative28, increasing size and geographical diversity drastically. img imaging data, seg reference segmentations.
  2. abased on partitioning 1.
Back to article page