Pulmonary Artery Segmentation Challenge 2022


[Notice]  Anyone using this dataset, please cite the following challenge overview paper on arXiv.

Luo, Gongning, et al. "Efficient automatic segmentation for multi-level pulmonary arteries: The PARSE challenge." arXiv preprint arXiv:2304.03708 (2023).

[NEWS] 2022/07/19 Extended the evaluation phase to July 31st.

You can submit your validation results on the submission page until July 31st. And you can submit your docker and paper from July 20th to August 7th.


[NEWS] 2022/07/15 Validation dataset release.


[NEWS] 2022/04/06 Training dataset release.


[NEWS] 2022/03/28-Parse challenge 2022  is now open for registration! Remember to send the signed document to PARSE2022@hotmail.com for participation.

About

It is of significant clinical interest to study pulmonary artery structures in the field of medical image analysis. One prerequisite step is to segment pulmonary artery structures from CT with high accuracy and low time-consuming. The segmentation of pulmonary artery structures benefits the quantification of its morphological changes for diagnosis of pulmonary hypertension and thoracic surgery. However, due to the complexity of pulmonary artery topology, automated segmentation of pulmonary artery topology is a challenging task.

Besides, the open accessible large-scale CT data with well labeled pulmonary artery are scarce (The large variations of the topological structures from different patients make the annotation an extremely challenging process). The lack of well labeled pulmonary artery hinders the development of automatic pulmonary artery segmentation algorithm. Hence, we try to host the first Pulmonary ARtery SEgmentation challenge in MICCAI 2022 (Named Parse2022) to start a new research topic and make a solid benchmark for pulmonary artery segmentation task.

We have collected 200 3D volumes with refined pulmonary artery labeling from 10 clinicians, 100 for the training dataset, 70 for the closed testing dataset and 30 for the opened validated dataset.  Multi-level Dice, Multi-level HD95, Maximum used GPU memory, and time-cost are adopted as evaluation metrics. This challenge will also promote the pulmonary disease treatment, interactions between researchers and interdisciplinary communication.

Task

Participants are required to segment artery in 3D pulmonary CT image.

Schedule

  • Registration: March 28 (11:59PM GMT), 2022
  • Training dataset release: April 6 (11:59PM GMT), 2022
  • Validation dataset release, open validation leaderboard submission: July 15 (11:59PM GMT), 2022
  • Deadline for the validation leaderboard submission: July 20 (11:59PM GMT), 2022
  • Opening docker and short paper submission for testing dataset: July 20 July 31 (11:59PM GMT), 2022
  • Deadline for docker and short paper submission: July 30 August 7 (11:59PM GMT), 2022
  • Winner and invitation speakers: September 18 (11:59PM GMT), 2022


Registration

Individuals or team members interested in participating in this challenge should carefully study the challenge rules and  follow the instructions to join challenge.

*Please note that by participating in this challenge you are agreeing to all its rules and policies.

Award

  1. Successful participation awards, which are electronic certificates, will be awarded to all teams that obtain valid test scores in the challenge leaderboard and complete technical paper submissions reviewed by the organizing committee.
  2. The top-1 team will receive 500 dollars or electronic products with similar prices. The exquisite certificates will be awarded to all members of the Top-1 team.
  3. The team that win the second place will receive 300 dollars or electronic products with similar prices. The exquisite certificates will be awarded to all members of the Top-2 team.
  4. The team that win the third place will receive 200 dollars or electronic products with similar prices. The exquisite certificates will be awarded to all members of the Top-3 team.
  5. The team achieving the first place in the single index (such as Dice, HD95, Maximum used GPU memory or time consuming) will be awarded to all members with electronic certificates.

Citation

If using our dataset, you must cite the following paper:

[1] G. Luo, K. Wang, J. Liu, S. Li, X. Liang, X. Li, S. Gan, W. Wang, S. Dong, W. Wang, P. Yu, E. Liu, H. Wei, N. Wang, J. Guo, H. Li, Z. Zhang, Z. Zhao, N. Gao, N. An, A. Pakzad, B. Rangelov, J. Dou, S. Tian, Z. Liu, Y. Wang, A. Sivalingam, K. Punithakumar, Z. Qiu and X. Gao (2023). Efficient automatic segmentation for multi-level pulmonary arteries: The PARSE challenge. ArXiv, abs/2304.03708. [ArXiv]