A-Eval

🤖 Trained Model

We provide pre-trained model that has been jointly trained on A-Eval. You can download it from our Google Drive repository: Google Drive Download Link

📚 Datasets

Dataset	Modality	# Train	# Test	# Organs	# Organs (Test)	Region
FLARE22	CT	50 labeled 2000 unlabeled	50	13	10	North America Europe
AMOS CT	CT	200	40	15	10	Asia
WORD	CT	100	30	16	10	Asia
TotalSegmentator v2	CT	1082	89	117	10	Europe
BTCV	CT	-	30	13	10	North America
AMOS MR	MR	40	20	15	10	Asia
TotalSegmentator MR	MR	268	30	56	10	Europe
A-Eval Totals	CT & MR	1432 labeled CT 2000 unlabeled CT 308 MR	239 CT 50 MR	10	10	North America Europe Asia

To ensure a meaningful and fair comparison across datasets, we evaluate the models' performance based on a set of ten organ classes shared by all datasets. We unify these labels using an overlapped label system. The corresponding code for label systems and label conversion can be found in the repository: label_systems.py and convert_label.py.

Organ Class	FLARE22	AMOS CT	WORD*	TotalSegmentator v2	AMOS MR	TotalSegmentator MR	A-Eval
Liver	✓	✓	✓	✓	✓	✓	✓
Kidney Right	✓	✓	✓	✓	✓	✓	✓
Kidney Left	✓	✓	✓	✓	✓	✓	✓
Spleen	✓	✓	✓	✓	✓	✓	✓
Pancreas	✓	✓	✓	✓	✓	✓	✓
Aorta	✓	✓	✗	✓	✓	✓	✗
Inferior Vena Cava	✓	✓	✗	✓	✓	✓	✗
Adrenal Gland Right	✓	✓	✓	✓	✓	✓	✓
Adrenal Gland Left	✓	✓	✓	✓	✓	✓	✓
Gallbladder	✓	✓	✓	✓	✓	✓	✓
Esophagus	✓	✓	✓	✓	✓	✓	✓
Stomach	✓	✓	✓	✓	✓	✓	✓
Duodenum	✓	✓	✓	✓	✓	✓	✗

*Note: The WORD dataset has been post-processed to distinguish between left and right adrenal glands.

Results

🎫 License

This project is released under the Apache 2.0 license.

🙏 Acknowledgement

• Special thanks go to the creators and maintainers of the public datasets that made our research possible:
  • FLARE22
  • AMOS
  • WORD
  • TotalSegmentator
  • BTCV
• Thanks to the SOTA framework of: nnUNet

👋 Hiring & Global Collaboration

• Hiring: We are hiring researchers, engineers, and interns in General Vision Group, Shanghai AI Lab. If you are interested in Medical Foundation Models and General Medical AI, including designing benchmark datasets, general models, evaluation systems, and efficient tools, please contact us.
• Global Collaboration: We're on a mission to redefine medical research, aiming for a more universally adaptable model. Our passionate team is delving into foundational healthcare models, promoting the development of the medical community. Collaborate with us to increase competitiveness, reduce risk, and expand markets.
• Contact: Junjun He([email protected]), Jin Ye([email protected]), and Tianbin Li ([email protected]).