Benchmarking robustness of automated CT pancreas segmentation: achieving human-level reliability through human-in-the-loop optimization
Automated pancreas segmentation reduces workload but remains constrained by tail-case failures. Active learning enhances model reliability, bridging the gap between artificial intelligence and human-level performance.
Authors
- Ralph H. Hruban (ORCID: https://orcid.org/0000-0003-4554-5672)
- Alejandra Blanco (ORCID: https://orcid.org/0000-0003-1856-5184)
- Elliot K. Fishman (ORCID: https://orcid.org/0000-0002-2567-1658)
- Rahul Dodhia (ORCID: https://orcid.org/0000-0003-3812-3906)
- Felipe Oviedo (ORCID: https://orcid.org/0000-0001-9480-6539)
- Juan Lavista Ferres (ORCID: https://orcid.org/0000-0002-9654-3178)
- William B. Weeks (ORCID: https://orcid.org/0000-0002-9918-1360)
- Florent Tixier (ORCID: https://orcid.org/0000-0002-9668-6574)
- Felipe Lopez‐Ramirez (ORCID: https://orcid.org/0000-0002-1560-9172)
- Linda Chi Hang Chu (ORCID: https://orcid.org/0000-0001-9729-2756)
- Satomi Kawamoto (ORCID: https://orcid.org/0000-0002-3577-1388)
Institutions
- Microsoft (United States) (US)
- Johns Hopkins University (US)
- Johns Hopkins Medicine (US)
Publication Details
- Journal
- Radiology Advances
- Published
- 2025-11-01
- DOI
- https://doi.org/10.1093/radadv/umaf040
- Citations
- 2
- Primary Topic
- Pancreatic and Hepatic Oncology Research
- Type
- article
- Field-Weighted Citation Impact
- 2.09