Deep Learning for Automated Segmentation of Basal Cell Carcinoma on Mohs Micrographic Surgery Frozen Section Slides
This study trained a segmentation model to localize BCC on MMS frozen section slides with reasonably high sensitivity and specificity, and this varied by BCC subtype. More accurate and clinically relevant performance metrics for segmentation studies are needed.
Authors
- Trent D. Walker (ORCID: https://orcid.org/0000-0002-2632-2126)
- Kirsten M. Johnson (ORCID: https://orcid.org/0000-0003-2388-0621)
- Kathryn T. Shahwan (ORCID: https://orcid.org/0000-0002-8744-1962)
- Daan Geijs (ORCID: https://orcid.org/0000-0003-3984-1766)
- Geert Litjens (ORCID: https://orcid.org/0000-0003-1554-1291)
- Vamsi Varra (ORCID: https://orcid.org/0000-0002-0385-4189)
- Rachel M. Kirven (ORCID: https://orcid.org/0000-0003-4312-0670)
- David R. Carr
Institutions
- Radboud University Nijmegen (NL)
- University of North Dakota (US)
- Altru Health System (US)
- The Ohio State University Wexner Medical Center (US)
- Radboud University Medical Center (NL)
Publication Details
- Journal
- Dermatologic Surgery
- Published
- 2024-12-03
- DOI
- https://doi.org/10.1097/dss.0000000000004501
- Primary Topic
- Nonmelanoma Skin Cancer Studies
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Regeneron Pharmaceuticals