Related Experiment Video
Updated: Jul 9, 2026

05:33
Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
Published on: July 11, 2025
An Interpretable AI System for Oral Leukoplakia Progression: From Early Screening to Lesion Delineation
Linfei Feng1,2, Guanyu Chen2,3, Huabao Chen2,4
1Department of Oral and Maxillofacial Surgery, The First Affiliated Hospital of Anhui Medical University, Hefei, Anhui, China.
NPJ Digital Medicine
|July 7, 2026
Summary
A new deep learning system accurately diagnoses oral potentially malignant disorders (OPMDs) from clinical images. This AI tool offers visual explanations, improving early detection and monitoring of OPMDs, including leukoplakia.
Area of Science:
- Artificial Intelligence in Medicine
- Oral Pathology
- Digital Diagnostics
Background:
- Oral potentially malignant disorders (OPMDs) are precursors to oral cancer.
- Current diagnostics involve invasive biopsies and subjective assessments, limiting accessibility.
- Systematic monitoring of OPMDs like leukoplakia is crucial for early intervention.
Purpose of the Study:
- To develop and validate a deep learning system for OPMD diagnosis and visual explainability.
- To improve the accuracy and accessibility of OPMD screening and monitoring.
- To address barriers in clinical AI adoption for oral cancer detection.
Main Methods:
- A deep learning framework trained on 778 annotated clinical images.
- The system includes a diagnostic classifier and an interpretable segmentation module.
- External validation performed on 193 independent clinical cases.
Main Results:
- Diagnostic accuracy of 91.1% for distinguishing normal mucosa, leukoplakia, and malignant transformation.
- Pixel-level lesion delineation achieved a mean average precision of 72.2%.
- External validation confirmed high diagnostic accuracy (90.5%) and outperformed baseline models.
Conclusions:
- The deep learning system provides reliable OPMD classification with visual explainability.
- The AI tool demonstrates high accuracy and speed, enabling point-of-care deployment.
- This framework offers a scalable solution for OPMD screening and monitoring in various healthcare settings.