Pulmonary Artery Segmentation Challenge 2022

It is of significant clinical interest to study pulmonary artery structures in the field of medical image analysis. The pulmonary arteries used in this challenge are located in the chest cavity of the human body, including the pulmonary trunk coming out of the pulmonary valve of the right ventricle, the left and right pulmonary arteries, and their main branches in the lungs. The primary function is to transport the human blood that's low in oxygen and high in carbon dioxide to the pulmonary capillaries of the lungs for the exchange of oxygen and carbon dioxide, which is extremely important. One prerequisite step is to segment pulmonary artery structures from CT with high accuracy and low time-consumption. 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 memory, and time-cost are adopted as evaluation metrics. This challenge will also promote the pulmonary disease treatment, interactions between researchers and interdisciplinary communication.