Abstract:90 Background: Premalignant lesions within the oral cavity demonstrate varied degrees of dysplasia. Oral stratified squamous epithelium is sensitive to carcinogenic insults and lesions progress to cancer at variable rates. The current diagnostic standard involves visual and tactile inspection followed by a biopsy. However, thus far, there is clinical ambiguity about which lesions may progress to carcinoma and there are insufficient prognostic tools available to guide clinical decision-making for follow-up. Deep learning is a form of artificial intelligence which can be applied to images. Self-supervised learning (SSL) and attention-based multiple instance learning (aMIL), are deep learning extensions which eliminate the need for pathologist annotations of histology and are more generalizable as the model learns general representations of unlabeled data, compared to traditional tile-based methods. We developed a novel artificial intelligence pipeline to predict progression to oral carcinoma using histological images from oral premalignant lesions. Methods: Digital histopathology images and clinical data cohorts were obtained from the US National Cancer Institute (n=561) and the University of Iowa (n=193) for training, and from the University of Chicago (n=214) for external validation. Patients were defined as progressing if initially non-malignant lesions had documented progression to carcinoma within 12 months of follow-up. Model training and evaluation was done in Python using our publicly available open-source Slideflow platform. We developed a deep learning pipeline utilizing SimCLR as an SSL method for feature extraction paired with an aMIL head for classification. The SimCLR feature generator was trained on only the training cohort, then used to generate features for both training and validation data. The aMIL model was trained on SimCLR-generated features. Trained model performance was evaluated on the external University of Chicago cohort and measured via areas under the receiver operating characteristic curve and precision recall curve (AUROC and AUPRC, respectively). During training and validation, model reliability was interpreted using Class Activation Mapping (Grad-CAM) and generative adversarial network (GAN) methods. Results: On training data, the model achieved average 3-fold cross validation AUROC of 0.86 with AUPRC of 0.83. In the external validation data from UC, the model achieved an AUROC of 0.80 with an AUPRC of 0.79. Deployed as a “rule-in” test for early progression prediction, the model achieved a validation data set specificity of 91% with a sensitivity of 40%. Conclusions: We developed a deep learning model to automatically predict cancer progression from premalignant lesions histopathology using a multi-institutional cohort, with promising external validation performance.

Machine Learning Algorithms Enhance Prediction Of Malignant Transformation Using Individual Dysplasia Features.

Artificial Intelligence for The Prediction of Malignant Transformation of Oral Leukoplakia

A Fully Automated and Explainable Algorithm for the Prediction of Malignant Transformation in Oral Epithelial Dysplasia

Artificial Intelligence Applications in Oral Cancer and Oral Dysplasia

A fully automated and explainable algorithm for predicting malignant transformation in oral epithelial dysplasia

A Deep-Learning Model to Predict Risk of Malignant Transformation in Oral Epithelial Dysplasia

A Deep Learning System to Predict Epithelial Dysplasia in Oral Leukoplakia

Development and Validation of an Artificial Intelligence-based Pipeline for Predicting Oral Epithelial Dysplasia Malignant Transformation

A pathology-based machine learning method to assist in epithelial dysplasia diagnosis

Prediction of the risk of cancer and the grade of dysplasia in leukoplakia lesions using deep learning

An Update on the Use of Artificial Intelligence in Digital Pathology for Oral Epithelial Dysplasia Research

Development of a Pathomics-Based Model for the Prediction of Malignant Transformation in Oral Leukoplakia

Machine learning in oral squamous cell carcinoma: Current status, clinical concerns and prospects for future-A systematic review

Machine Learning in the Detection of Oral Lesions With Clinical Intraoral Images

Deep learning-based pathology image analysis predicts cancer progression risk in patients with oral leukoplakia

Large Language Model Assistance In Diagnosing Oral Squamous Cell Carcinoma: Integrating Clinical And Histopathological Features

Automated Analysis of Nuclear Parameters in Oral Exfoliative Cytology Using Machine Learning

Architectural and Cytological Features of Epithelial Dysplasia Associated with Transformation Risk.

Machine learning based on magnetic resonance imaging and clinical parameters helps predict mesenchymal-epithelial transition factor expression in oral tongue squamous cell carcinoma: a pilot study

Integrating Machine Learning with Web-Based Tools for Personalized Prognosis in Oral Adenoid Cystic Carcinoma

Artificial intelligence-based cancer progression prediction of oral premalignant lesions via self-supervised deep learning on histopathology.