Next-generation AI framework for comprehensive oral leukoplakia evaluation and management

Abstract Oral potentially malignant disorder poses a significant risk of malignant transformation, particularly in cases with epithelial dysplasia (OED). Current OED assessment methods are invasive and lack reliable decision-support tools for cancer risk evaluation and follow-up optimization. This s...

Full description

Saved in:
Bibliographic Details
Main Authors: JingWen Li, YaFang Zhou, MengJing Zhang, John Adeoye, Jane JingYa Pu, MiMi Zhou, ChuanXia Liu, LiJie Fan, Colman McGrath, Dian Zhang, LiWu Zheng
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:npj Digital Medicine
Online Access:https://doi.org/10.1038/s41746-025-01885-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849761115779629056
author JingWen Li
YaFang Zhou
MengJing Zhang
John Adeoye
Jane JingYa Pu
MiMi Zhou
ChuanXia Liu
LiJie Fan
Colman McGrath
Dian Zhang
LiWu Zheng
author_facet JingWen Li
YaFang Zhou
MengJing Zhang
John Adeoye
Jane JingYa Pu
MiMi Zhou
ChuanXia Liu
LiJie Fan
Colman McGrath
Dian Zhang
LiWu Zheng
author_sort JingWen Li
collection DOAJ
description Abstract Oral potentially malignant disorder poses a significant risk of malignant transformation, particularly in cases with epithelial dysplasia (OED). Current OED assessment methods are invasive and lack reliable decision-support tools for cancer risk evaluation and follow-up optimization. This study developed and validated OMMT-PredNet, a fully automated multimodal deep learning framework requiring no manual ROI annotation, for non-invasive OED identification and time-dependent cancer risk prediction. Utilizing data from 649 histopathologically confirmed leukoplakia cases across multiple institutions (2003–2024), including 598 cases in the primary cohort and 51 in the external validation set, the model integrated paired high-resolution clinical images and medical records. OMMT-PredNet achieved an AUC of 0.9592 (95% CI: 0.9491–0.9693) for cancer risk prediction and 0.9219 (95% CI: 0.9088–0.9349) for OED identification, with high specificity (MT: 0.9490; OED: 0.9182) and precision (MT: 0.9442; OED: 0.9303). Calibration and decision curve analyses confirmed clinical applicability, while external validation demonstrated robustness. This multidimensional model effectively predicts OED and cancer risk, highlighting its global applicability in enhancing oral cancer screening and improving patient outcomes.
format Article
id doaj-art-5b0abcc6757d48b19d58ec1d17b69bf7
institution DOAJ
issn 2398-6352
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series npj Digital Medicine
spelling doaj-art-5b0abcc6757d48b19d58ec1d17b69bf72025-08-20T03:06:08ZengNature Portfolionpj Digital Medicine2398-63522025-08-018111010.1038/s41746-025-01885-8Next-generation AI framework for comprehensive oral leukoplakia evaluation and managementJingWen Li0YaFang Zhou1MengJing Zhang2John Adeoye3Jane JingYa Pu4MiMi Zhou5ChuanXia Liu6LiJie Fan7Colman McGrath8Dian Zhang9LiWu Zheng10Division of Oral & Maxillofacial Surgery, Faculty of Dentistry, University of Hong KongDepartment of Computer Science and Software Engineering, Shenzhen UniversityDepartment of Computer Science and Software Engineering, Shenzhen UniversityDivision of Applied Oral Sciences & Community Dental Care, Faculty of Dentistry, The University of Hong KongDivision of Oral & Maxillofacial Surgery, Faculty of Dentistry, University of Hong KongStomatology Hospital, Zhejiang University School of MedicineStomatology Hospital, Zhejiang University School of MedicineStomatology Hospital, Zhejiang University School of MedicineDivision of Applied Oral Sciences & Community Dental Care, Faculty of Dentistry, The University of Hong KongDepartment of Computer Science and Software Engineering, Shenzhen UniversityDivision of Oral & Maxillofacial Surgery, Faculty of Dentistry, University of Hong KongAbstract Oral potentially malignant disorder poses a significant risk of malignant transformation, particularly in cases with epithelial dysplasia (OED). Current OED assessment methods are invasive and lack reliable decision-support tools for cancer risk evaluation and follow-up optimization. This study developed and validated OMMT-PredNet, a fully automated multimodal deep learning framework requiring no manual ROI annotation, for non-invasive OED identification and time-dependent cancer risk prediction. Utilizing data from 649 histopathologically confirmed leukoplakia cases across multiple institutions (2003–2024), including 598 cases in the primary cohort and 51 in the external validation set, the model integrated paired high-resolution clinical images and medical records. OMMT-PredNet achieved an AUC of 0.9592 (95% CI: 0.9491–0.9693) for cancer risk prediction and 0.9219 (95% CI: 0.9088–0.9349) for OED identification, with high specificity (MT: 0.9490; OED: 0.9182) and precision (MT: 0.9442; OED: 0.9303). Calibration and decision curve analyses confirmed clinical applicability, while external validation demonstrated robustness. This multidimensional model effectively predicts OED and cancer risk, highlighting its global applicability in enhancing oral cancer screening and improving patient outcomes.https://doi.org/10.1038/s41746-025-01885-8
spellingShingle JingWen Li
YaFang Zhou
MengJing Zhang
John Adeoye
Jane JingYa Pu
MiMi Zhou
ChuanXia Liu
LiJie Fan
Colman McGrath
Dian Zhang
LiWu Zheng
Next-generation AI framework for comprehensive oral leukoplakia evaluation and management
npj Digital Medicine
title Next-generation AI framework for comprehensive oral leukoplakia evaluation and management
title_full Next-generation AI framework for comprehensive oral leukoplakia evaluation and management
title_fullStr Next-generation AI framework for comprehensive oral leukoplakia evaluation and management
title_full_unstemmed Next-generation AI framework for comprehensive oral leukoplakia evaluation and management
title_short Next-generation AI framework for comprehensive oral leukoplakia evaluation and management
title_sort next generation ai framework for comprehensive oral leukoplakia evaluation and management
url https://doi.org/10.1038/s41746-025-01885-8
work_keys_str_mv AT jingwenli nextgenerationaiframeworkforcomprehensiveoralleukoplakiaevaluationandmanagement
AT yafangzhou nextgenerationaiframeworkforcomprehensiveoralleukoplakiaevaluationandmanagement
AT mengjingzhang nextgenerationaiframeworkforcomprehensiveoralleukoplakiaevaluationandmanagement
AT johnadeoye nextgenerationaiframeworkforcomprehensiveoralleukoplakiaevaluationandmanagement
AT janejingyapu nextgenerationaiframeworkforcomprehensiveoralleukoplakiaevaluationandmanagement
AT mimizhou nextgenerationaiframeworkforcomprehensiveoralleukoplakiaevaluationandmanagement
AT chuanxialiu nextgenerationaiframeworkforcomprehensiveoralleukoplakiaevaluationandmanagement
AT lijiefan nextgenerationaiframeworkforcomprehensiveoralleukoplakiaevaluationandmanagement
AT colmanmcgrath nextgenerationaiframeworkforcomprehensiveoralleukoplakiaevaluationandmanagement
AT dianzhang nextgenerationaiframeworkforcomprehensiveoralleukoplakiaevaluationandmanagement
AT liwuzheng nextgenerationaiframeworkforcomprehensiveoralleukoplakiaevaluationandmanagement