PENGWIN Task 1: Pelvic Fracture Segmentation on CT
Creators
- 1. Rossum Robot
-
2.
Beihang University
-
3.
Johns Hopkins University
-
4.
Beijing Jishuitan Hospital
Description
The CT segmentation task (Task 1) of the PENGWIN segmentation challenge is designed to advance the development of automated fracture segmentation methods for pelvic CT scans, with a focus on enhancing their accuracy and efficiency. Our dataset comprises CT scans from 150 patients scheduled for pelvic reduction surgery, collected from multiple institutions using a variety of scanning equipment. This dataset represents a diverse range of patient cohorts and fracture types. Ground-truth segmentations for sacrum and hipbone fragments have been semi-automatically annotated and subsequently validated by medical experts.
This repository contains the training set of 100 CT scans with pelvic fractures and their ground-truth segmentation labels. The images and labels are stored in mha format. Each bone anatomy (sacrum, left hipbone, right hipbone) has up to 10 fragments. Bone that does not present any fracuture has only one fragment, which is itself. Label assignment: 0 = background, 1-10 = sacrum fragment, 11-20 = left hipbone fragment, 21-30 = right hipbone fragment.
For more information, please visit the challenge webpage. For the PENGWIN simulated X-ray training dataset (Task 2), please visit the separate repository (10.5281/zenodo.10913196).
Files
PENGWIN_CT_train_images_part1.zip
Additional details
Related works
- Is new version of
- Conference proceeding: 10.1007/978-3-031-43996-4_30 (DOI)
Dates
- Created
-
2024-03-20