Multi-Modality Abdominal Multi-Organ Segmentation Challenge 2022

News

  • [2024.6.17] The MICCAI 2024 AMOS-MM competition brings two new tasks: medica report generation and medical visual question answering, please check the link for more information!
  • [2022.10.10] The whole set of AMOS (train set, validation set, and test data (wo gt label)) has been released on the zenodo platform, please check the zenodo. You can submit your test set performance via the submission entry.
  • [2022.09.10] The event schedules of AMOS in MICCAI22 have been released, please see here.
  • [2022.08.21] The final ranking results are available here. Cheers everyone!
  • [2022.07.19] We have placed a suggested limitation on the model run time, see the forum, we have extended the submission deadline by two days (7.22 AOE) to reduce disruption to the teams. Sorry for the inconvenience.
  • [2022.06.27] The deadline for the first stage is extended to 7.17. Besides, the latex template is provided in the instruction.
  • [2022.06.27] The guidance on docker submission has been provided, please see the instruction.
  • [2022.05.10] The first stage of validation is now Opening, please see this for more details.
  • [2022.05.06] Due to technical problems, the opening of the first phase of validation will be delayed until the 10th.
  • [2022.05.06] We offer additional cash awards. please see the Cash award.
  • [2022.05.01] The dataset is now available for download, please follow the instruction to register and download data.
  • [2022.04.27] AWS will offer a total of $107,500 in credit awards to the top ten teams, please see AWS credit award.
  • [2022.04.15] AMOS22 website is now fully open. Please check the timeline.
  • [2022.04.05] AMOS22 will be present at the DALI workshop and the award winners of the competition will be invited to present their solutions. More details will be updated later.

Motivation

Abdominal multi-organ segmentation is one of the most attractive topics in the field of medical image analysis, which plays an important role in supporting clinical workflows such as disease diagnosis and treatment planning. 

The recent success of deep learning methods applied for abdominal multi-organ segmentation exposes the lack of large-scale comprehensive benchmarks for developing and comparing such methods. While several benchmark datasets for abdominal organ segmentation are available, the limited number of organs of interest and training samples still limits the power of modern deep models and makes it difficult to provide a fully comprehensive and fair estimate of different methods.

And, most research in medical image analysis today focuses on building bespoke systems to handle stereotypical inputs and outputs associated with a single task, the complexity of systems like this can grow dramatically as the inputs or outputs grow more diverse. If a single algorithm could handle a wide variety of input patterns and outputs, the actual deployment would be greatly simplified, and it would be consistent with human capabilities: the ability to absorb data from many sources, learn common rules, integrate them seamlessly, and serve a variety of tasks.

To address the above drawbacks and further promote the development of medical image segmentation technology, we present AMOS, a large-scale, clinical and diverse abdominal multi-organ segmentation benchmark. It provides 500 CT and 100 MRI scans collected from multi-center, multi-vendor, multi-modality, multi-phase, multi-disease patients, each with voxel-level annotations of 15 abdominal organs. 

Figure 1. The example illustration of the AMOS dataset, AMOS provides 500 CT and 100 MRI scans with voxel-level annotations of 15 abdominal organs, including the spleen, right kidney, left kidney, gallbladder, esophagus, liver, stomach, aorta, inferior vena cava, pancreas, right adrenal gland, left adrenal gland, duodenum, bladder, prostate/uterus.

Specifically, AMOS 2022 contains two tasks in which the participating teams can take place and submit their result(s):

a) Task 1 - Segmentation of abdominal organs (CT only):  as a mostly regular task, Task 1 aims to comprehensively evaluate the performance of different segmentation methods across large-scale and great diversity CT scans, a total of 500 cases with annotations of 15 organs (spleen, right kidney, left kidney, gallbladder, esophagus, liver, stomach, aorta, inferior vena cava, pancreas, right adrenal gland, left adrenal gland, duodenum, bladder, prostate/uterus) are presented.

b) Task 2 - Segmentation of abdominal organs (CT & MRI): this task extends the image modality target of Task 1 to the MRI modality. Under such a “Cross Modality” setting, a single algorithm is required to segment abdominal organs from both CT and MRI. Specifically, additional 100 MRI scans with the same type of annotation will be provided. 

The AMOS 2022 challenge has the following main features:

  1. Large-Scale: we collect the most comprehensive abdominal multi-organ data to date.
  2. Multi-Modality: we build the dataset with two modalities CT and MRI.
  3. Multi-Tasks: we design two tasks including the regular CT segmentation task and the extended cross-modality CT & MRI segmentation task. 

Timeline

Expected Website Full Open: April 15, 2022
Expected Training Set Released: May 1, 2022
Expected Validation Set Released: May 1, 2022
Expected First Stage Submission Open: May 7, 2022  May 10, 2022
Expected First stage Deadline: July 15, 2022 July 17, 2022
Expected Submission of Short Papers,  Docker Alogs Deadline: July 20, 2022 July 22, 2022 (anywhere on the earth)
Expected Announcement of the final Top 10 ranked teams: Aug 15, 2022

Citation

if you found this dataset useful for your research, please cite:

@article{ji2022amos,
  title={AMOS: A Large-Scale Abdominal Multi-Organ Benchmark for Versatile Medical Image Segmentation},
  author={Ji, Yuanfeng and Bai, Haotian and Yang, Jie and Ge, Chongjian and Zhu, Ye and Zhang, Ruimao and Li, Zhen and Zhang, Lingyan and Ma, Wanling and Wan, Xiang and others},
  journal={arXiv preprint arXiv:2206.08023},
  year={2022}
}