English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Background: Margin assessment of basal cell carcinoma using the frozensection is a common task of pathology intraoperative consultation. Althoughfrequently straight-forward, the determination of the presence or absence ofbasal cell carcinoma on the tissue sections can sometimes be challenging. Weexplore if a deep learning model trained on mobile phone-acquired frozensection images can have adequate performance for future deployment. Materialsand Methods: One thousand two hundred and forty-one (1241) images of frozensections performed for basal cell carcinoma margin status were acquired usingmobile phones. The photos were taken at 100x magnification (10x objective). Theimages were downscaled from a 4032 x 3024 pixel resolution to 576 x 432 pixelresolution. Semantic segmentation algorithm Deeplab V3 with Xception backbonewas used for model training. Results: The model uses an image as input andproduces a 2-dimensional black and white output of prediction of the samedimension; the areas determined to be basal cell carcinoma were displayed withwhite color, in a black background. Any output with the number of white pixelsexceeding 0.5% of the total number of pixels is deemed positive for basal cellcarcinoma. On the test set, the model achieves area under curve of 0.99 forreceiver operator curve and 0.97 for precision-recall curve at the pixel level.The accuracy of classification at the slide level is 96%. Conclusions: The deeplearning model trained with mobile phone images shows satisfactory performancecharacteristics, and thus demonstrates the potential for deploying as a mobilephone app to assist in frozen section interpretation in real time.

Deep learning model trained on mobile phone-acquired frozen section images effectively detects basal cell carcinoma