Subcutaneous edema segmentation on abdominal CT using multi-class labels and iterative annotation
Weakly supervised learning using 3D multi-class labels and iterative annotation is an efficient way to perform high-quality edema segmentation with minimal false positives. Automated edema segmentation can produce edema volume estimates that are highly correlated with manual annotation. The proposed approach is promising for clinical applications to monitor anasarca using estimated edema volumes.
Authors
- Ronald M. Summers (ORCID: https://orcid.org/0000-0001-8081-7376)
- Sayantan Bhadra (ORCID: https://orcid.org/0000-0001-8321-1142)
- Jianfei Liu (ORCID: https://orcid.org/0000-0001-9885-1695)
Institutions
- National Institutes of Health Clinical Center (US)
Publication Details
- Journal
- International Journal of Computer Assisted Radiology and Surgery
- Published
- 2024-09-14
- DOI
- https://doi.org/10.1007/s11548-024-03262-4
- Citations
- 1
- Primary Topic
- Thermal Regulation in Medicine
- Type
- article
- Field-Weighted Citation Impact
- 0.42
Funders
- National Institutes of Health
- NIH Clinical Center