Skip to main content

Deep Learning for Automated Segmentation of Basal Cell Carcinoma on Mohs Frozen Section Slides

Varra V, Shahwan KT, Johnson K, et al.
Dermatologic Surgery (2025)

Open AccessJC: May 2025

This study developed and evaluated a deep learning algorithm for automated segmentation of basal cell carcinoma on Mohs frozen section slides. The AI model demonstrated high accuracy in identifying tumor regions, showing potential for assisting Mohs surgeons in margin assessment and reducing interpretation time.

Take-Home Messages

  • Deep learning algorithms can accurately segment BCC on Mohs frozen sections, offering potential as a computer-aided detection tool.
  • AI-assisted margin assessment may improve efficiency and reduce diagnostic errors during Mohs surgery for BCC.

Topic

Margin Assessment & IHC

Frozen section interpretation, MART-1, immunohistochemistry stains

Related MohsPedia Articles

Abstract

Deep learning has been used to classify basal cell carcinoma (BCC) on histopathologic images. Segmentation models, required for localization of tumor on Mohs surgery (MMS) frozen section slides, have yet to reach clinical utility. To train a segmentation model to localize BCC on MMS frozen section slides and to evaluate performance by BCC subtype. The study included 348 fresh frozen tissue slides, scanned as whole slide images, from patients treated with MMS for BCC. BCC foci were manually ou...

Literature review only. This summary is an editorial interpretation and may not reflect the complete findings of the original publication. Always refer to the full-text article for clinical decision-making.