Deep learning for automatic volumetric bowel segmentation on body CT images
Questions Automated bowel segmentation is a first step for algorithms, including bowel tracing and length measurement, but the complexity of the gastrointestinal tract limits its accuracy. Findings Our 3D nnU-Net model showed high performance in segmentation and four-part separation of the GI tract (DSC > 0.95), except for the esophagus. Clinical relevance Our model accurately segments the gastrointestinal tract and separates it into major compartments. Our model potentially has use in various clinical applications, including semi-automated measurement of LBL in individuals with constipation.
Authors
- Han-Jae Chung (ORCID: https://orcid.org/0000-0002-3410-0179)
- Junghoan Park (ORCID: https://orcid.org/0000-0002-0636-3756)
- Soon Ho Yoon (ORCID: https://orcid.org/0000-0002-3700-0165)
- Eun Kyung Choe (ORCID: https://orcid.org/0000-0002-7222-1772)
- Jongmin Kim (ORCID: https://orcid.org/0000-0001-7020-0030)
- Da In Lee
- Kyu Joo Park (ORCID: https://orcid.org/0000-0002-9134-7426)
- Se‐Hyung Kim (ORCID: https://orcid.org/0000-0002-1694-9435)
- Sungeun Park (ORCID: https://orcid.org/0000-0003-1462-9689)
Institutions
- Seoul National University (KR)
- Seoul National University Hospital (KR)
- Seoul Medical Center (KR)
- Konkuk University Medical Center (KR)
Publication Details
- Journal
- European Radiology
- Published
- 2025-05-02
- DOI
- https://doi.org/10.1007/s00330-025-11623-z
- Citations
- 1
- Primary Topic
- Colorectal Cancer Screening and Detection
- Type
- article
- Field-Weighted Citation Impact
- 1.05
Funders
- Seoul National University