Transformer-based Model for Oral Epithelial Dysplasia Segmentation

TL;DR

This study introduces a Transformer-based model for segmenting oral epithelial dysplasia in histopathology images, demonstrating state-of-the-art performance and good generalisability across multiple external datasets.

Contribution

First externally validated Transformer-based segmentation model for oral epithelial dysplasia in histopathology images, improving detection accuracy and generalisability.

Findings

Mean F1-score of 0.81 internally

F1-score of 0.71 on external data

State-of-the-art segmentation results

Abstract

Oral epithelial dysplasia (OED) is a premalignant histopathological diagnosis given to lesions of the oral cavity. OED grading is subject to large inter/intra-rater variability, resulting in the under/over-treatment of patients. We developed a new Transformer-based pipeline to improve detection and segmentation of OED in haematoxylin and eosin (H&E) stained whole slide images (WSIs). Our model was trained on OED cases (n = 260) and controls (n = 105) collected using three different scanners, and validated on test data from three external centres in the United Kingdom and Brazil (n = 78). Our internal experiments yield a mean F1-score of 0.81 for OED segmentation, which reduced slightly to 0.71 on external testing, showing good generalisability, and gaining state-of-the-art results. This is the first externally validated study to use Transformers for segmentation in precancerous histology images. Our publicly available model shows great promise to be the first step of a fully-integrated pipeline, allowing earlier and more efficient OED diagnosis, ultimately benefiting patient outcomes.