layout	title	description	permalink	bibliography	toc	_styles
distill	TFM4MedIA	Transferring Foundation Models for Multi-center Real-World Medical Image Analysis	/track1/	reference.bib	name Important Dates name Motivation name Task name Registration name Data Overview name Metrics & Ranking name Rules	d-article { contain: layout style; overflow-x: hidden; border-top: 1px solid rgba(0, 0, 0, 0.1); padding-top: 2rem; color: rgba(0, 0, 0, 0.8); } d-article > * { grid-column: text; }

Important Dates

The schedule for this track is as follows. All deadlines (DDL) are in Pacific Standard Time.

Training Data Release	TBD
Validation Phase	TBD
Test Phase	TBD
Abstract Submission	TBD
Paper Submission	TBD
Notification	TBD
Camera Ready	TBD
Workshop (Half-Day)	TBD

Motivation

{% include figure.liquid loading="eager" path="/assets/img/tfm4media.png" class="img-fluid" zoomable=true %}

While foundation models like the Segment Anything Model (SAM) have demonstrated efficacy in various medical image analysis tasks, their performance on real-world data remains underexplored. Specifically, these models are typically trained for normal and large targets such as the liver and lungs, yet real-world data often originates from different centers with diverse modalities. Furthermore, the application of foundation models to complicated Regions of Interest (ROIs), like lesions or scars, poses additional challenges due to their small size and irregular shapes. Hence, developing effective and efficient transfer learning approaches to fully utilize those foundation models for real world medical image segmentation is of great values.

Task

In this track, we encourage participants to design effective transfer learning approaches to exploit the knowledge from existed foundation models efficiently and make them more effective in four segmentation tasks, including Myocardial Pathology Segmentation in MyoPS++ track, Liver Segmentation in LiFS track, Whole Heart Segmentation in WHS++ track and Left Atrial and Scar Segmentation in LAScarQS++ track.

Note: To address this task, participants are encouraged to leverage external data.

Registration

Please register here to participate in the challenge and get access to the dataset!

Data Overview

Multi-center datasets are provided for four sub-tasks. More detailed data information can be found here for Myocardial Pathology Segmentation, Liver Segmentation, Whole Heart Segmentation and Left Atrial and Scar Segmentation.

Training Dataset

Note: Only points or rectangular frames are acceptable as Prompts.

1. Myocardial Pathology Segmentation

Center	Num. patients	Sequences	Manual labels
1	81	LGE	Scar, left ventricle and myocardium
2	50	LGE, T2 and bSSFP	Scar, edema, left ventricle, myocardium and right ventricle
3	45	LGE, T2 and bSSFP	Scar, edema, left ventricle, myocardium and right ventricle
5	07	LGE and bSSFP	Scar, left ventricle, myocardium and and right ventricle
6	09	LGE and bSSFP	Scar, left ventricle, myocardium and and right ventricle
7	08	LGE and bSSFP	Scar, left ventricle, myocardium and and right ventricle

2. Liver Segmentation

Vendor	Center	Num. studies	Num. Annotations
A	1	100	10
A	2	100	10
B	3	50	10

3. Whole Heart Segmentation

Center	Num. patients	Modalities
A	20	CT
B	20	CT
C/D	20	MRI
E	26	MRI

4. Left Atrial and Scar Segmentation

Center	Modality	Num. task1	Num. task2
3	LGE MRI	60	130

Note: points or rectangular frames for each class are provided as Prompts. Participants can generate prompts based on the segmentation ground truth by themselves.

Validation Dataset

1. Myocardial Pathology Segmentation

Center	Num. patients	Sequences	Manual labels
4	25	LGE, T2 and bSSFP	Scar, edema, left ventricle, myocardium and right ventricle

2. [Liver Segmentation](http://zmic.org.cn/care_2024/track3/)

Vendor	Center	Num. studies	Num. Annotations
A	1	10	10
B	2	10	10
B	3	10	10

3. Whole Heart Segmentation

Center	Num. patients	Modalities
A	20	CT
B	10	CT
C/D	20	MRI

4. Left Atrial and Scar Segmentation

Center	Modality	Num. task1	Num. task2
2.1	LGE MRI	0	10
3	LGE MRI	10	10

*Note: **points** or **rectangular frames** for each class are provided as **Prompts**. Participants can generate prompts by themselves*. ### Test Dataset

1. Myocardial Pathology Segmentation

Center	Num. patients	Sequences	Manual labels
4	25	LGE, T2 and bSSFP	Scar, edema, left ventricle, myocardium and right ventricle

2. [Liver Segmentation](http://zmic.org.cn/care_2024/track3/)

The 160 test cases corresponded to 120 new cases from the vendors provided in the training set and 40 additional cases from a third unseen centre, that were tested for model generalizability.

Vendor	Center	Num. studies
A	1	40
B	2	40
B	3	40
C (new)	4	40

3. Whole Heart Segmentation

Center	Num. patients	Modalities
A	20	CT
B	14	CT
C/D	20	MRI
F	16	MRI

4. Left Atrial and Scar Segmentation

Center	Modality	Num. task1	Num. task2
1	LGE MRI	0	20
2.1	LGE MRI	0	10
2.2	LGE MRI	0	40
3	LGE MRI	24	14

*Note: for testing, no more than 5 points and 1 rectangular frame are used by organizers for each class as prompts*. ## Metrics & Ranking

Metrics

Dice Similarity Coefficient (DSC), Hausdorff Distance

Rank methods

The overall performance across all sub-tasks are considered for ranking:

1. Firstly, for each sub-task, the results will be ranked according to the Dice score on in-distribution (seen centre) and out-of-distribution (OOD, unseen centre) dataset, respectively.
2. Then the ranking results of all sub-tasks are averaged as the final rank.

This ranking approach encourage the participants to develop methods from foundation models to be consistently effective across all tasks as well as on OOD datasets.

Rules

1. Publicly available data and pretrained model are allowed.

Citations

Please cite these papers when you use the data for publications:

@article{Wu2023SemiSL,
  author={Wu, Fuping and Zhuang, Xiahai},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, 
  title={Minimizing Estimated Risks on Unlabeled Data: A New Formulation for Semi-Supervised Medical Image Segmentation}, 
  year={2023},
  volume={45},
  number={5},
  pages={6021-6036},
}

@article{GAO2023BayeSeg,
  title = {BayeSeg: Bayesian modeling for medical image segmentation with interpretable generalizability},
  journal = {Medical Image Analysis},
  volume = {89},
  pages = {102889},
  year = {2023},
  author = {Shangqi Gao and Hangqi Zhou and Yibo Gao and Xiahai Zhuang},
}

Contact

If you have any problems about this track, please contact Dr. Fuping Wu or Dr. Shangqi Gao.