Automated segmentation of intracranial hemorrhages from 3D CT
This work addresses automated segmentation for stroke diagnosis, but it is incremental as it applies existing methods to a specific challenge.
The authors tackled the problem of segmenting intracranial hemorrhages from 3D CT scans by using a 2D SegResNet network slice-wise without resampling and an ensemble of 18 models, achieving a top Dice score of 0.721 and overall rank 2 in the INSTANCE 2022 challenge.
Intracranial hemorrhage segmentation challenge (INSTANCE 2022) offers a platform for researchers to compare their solutions to segmentation of hemorrhage stroke regions from 3D CTs. In this work, we describe our solution to INSTANCE 2022. We use a 2D segmentation network, SegResNet from MONAI, operating slice-wise without resampling. The final submission is an ensemble of 18 models. Our solution (team name NVAUTO) achieves the top place in terms of Dice metric (0.721), and overall rank 2. It is implemented with Auto3DSeg.