FLARE-MedFM/FLARE-Task4-CT-FM

MICCAI FLARE25

Task 4: Foundation Models for 3D CT and MRI Scans (Homepage)

This is the official dataset for CT image foundation model development. We provide 10,000+ CT scans for model pretraining. Downstream tasks include:

• Abdominal disease classification
• Abdominal lesion segmentation
• Abdominal organ segmentation
• Lung lesion segmentation

Dataset

Dataset Name	Task	Metric	Source	License
Abdominal Disease Classification	multi-label Classification	mAP	https://era-ai-biomed.github.io/amos/	CC 4.0
Abdominal Lesion Segmentation	Abdominal Lesion Segmentation	DSC, NSD	https://codalab.lisn.upsaclay.fr/competitions/12239	CC-BY-NC 4.0
Abdominal Organ Segmentation	Abdominal Organ Segmentation	DSC, NSD	https://flare22.grand-challenge.org/	CC-BY-NC 4.0
Lung Lesion Segmentation	Lung Lesion Segmentation	DSC, NSD	https://luna25.grand-challenge.org/	CC 4.0

Folder structure

FLARE-Task4-CT-FM/
├── README.md
├── train_part1 # training set part 1
└── train_part2/ # training set part 2
├── val_downstream/ # validation sets for downstream tasks
│   ├── abdomen_disease_classify
│   ├── abdomen_lesion_seg
│   ├── abdomen_organ_seg
│   └── lung_lesion_seg
# To avoid potential data leakage, the testing set will not be released. 

Remarks

1. During training, please don't use any of the (public) annotations. Only raw images are allowed.

References

TBD