Development and multicenter validation of an AI driven model for quantitative meibomian gland evaluation

Abstract This multicenter retrospective study developed and validated an AI driven model for automated segmentation and quantitative evaluation of meibomian glands using infrared meibography images acquired by the Keratograph 5M device. A total of 1350 infrared meibography images were collected and...

Full description

Saved in:
Bibliographic Details
Main Authors: Li Li, Kunhong Xiao, Taichen Lai, Kunfeng Lai, Jiawen Lin, Zongyuan Ge, Lingyi Liang, Hao Huang, Xiaoshan Zhang, Li Liu, Yujie Wang, Xianwen Shang, Mingguang He, Ying Xue, Zhuoting Zhu
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:npj Digital Medicine
Online Access:https://doi.org/10.1038/s41746-025-01753-5
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849389285901336576
author Li Li
Kunhong Xiao
Taichen Lai
Kunfeng Lai
Jiawen Lin
Zongyuan Ge
Lingyi Liang
Hao Huang
Xiaoshan Zhang
Li Liu
Yujie Wang
Xianwen Shang
Mingguang He
Ying Xue
Zhuoting Zhu
author_facet Li Li
Kunhong Xiao
Taichen Lai
Kunfeng Lai
Jiawen Lin
Zongyuan Ge
Lingyi Liang
Hao Huang
Xiaoshan Zhang
Li Liu
Yujie Wang
Xianwen Shang
Mingguang He
Ying Xue
Zhuoting Zhu
author_sort Li Li
collection DOAJ
description Abstract This multicenter retrospective study developed and validated an AI driven model for automated segmentation and quantitative evaluation of meibomian glands using infrared meibography images acquired by the Keratograph 5M device. A total of 1350 infrared meibography images were collected and annotated for model training and validation. The model demonstrated high segmentation performance, with an Intersection over Union of 81.67% (95% Confidence Interval [CI]: 81.03–82.31) and accuracy of 97.49% (95% CI: 97.38–97.62), outperforming conventional algorithms. The agreement was observed between AI-based and manual gland grading (Kappa = 0.93) and gland counting (Spearman r = 0.9334). Repeatability analysis confirmed the model’s stability, and external validation across four independent centers yielded consistent results with AUCs exceeding 0.99. This AI tool offers a standardized, efficient, and objective method for meibography image analysis, which may improve diagnostic precision and assist in the clinical management of meibomian gland dysfunction across diverse populations.
format Article
id doaj-art-c52e20654e064943b8e1fc79bbd94a40
institution Kabale University
issn 2398-6352
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series npj Digital Medicine
spelling doaj-art-c52e20654e064943b8e1fc79bbd94a402025-08-20T03:42:00ZengNature Portfolionpj Digital Medicine2398-63522025-07-01811910.1038/s41746-025-01753-5Development and multicenter validation of an AI driven model for quantitative meibomian gland evaluationLi Li0Kunhong Xiao1Taichen Lai2Kunfeng Lai3Jiawen Lin4Zongyuan Ge5Lingyi Liang6Hao Huang7Xiaoshan Zhang8Li Liu9Yujie Wang10Xianwen Shang11Mingguang He12Ying Xue13Zhuoting Zhu14Centre for Eye Research Australia, Royal Victorian Eye and Ear HospitalDepartment of Ophthalmology and Optometry, Fujian Medical UniversityDepartment of Ophthalmology, Shengli Clinical Medical College of Fujian Medical University, Fujian Provincial Hospital, Fuzhou University Affiliated Provincial HospitalCollege of Computer and Data Science, Fuzhou UniversityCollege of Computer and Data Science, Fuzhou UniversityMonash e-Research Centre, Monash UniversityState Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangdong Provincial Key Laboratory of Ophthalmology and Visual ScienceDepartment of Ophthalmology, Zhuzhou Hospital Affiliated to Xiangya School of Medicine, Central South UniversityDongguan Huaxia Eye HospitalHuaxia Eye Hospital Group Putian Eye HospitalCentre for Eye Research Australia, Royal Victorian Eye and Ear HospitalSchool of Optometry, The Hong Kong Polytechnic UniversitySchool of Optometry, The Hong Kong Polytechnic UniversityDepartment of Ophthalmology, Shengli Clinical Medical College of Fujian Medical University, Fujian Provincial Hospital, Fuzhou University Affiliated Provincial HospitalCentre for Eye Research Australia, Royal Victorian Eye and Ear HospitalAbstract This multicenter retrospective study developed and validated an AI driven model for automated segmentation and quantitative evaluation of meibomian glands using infrared meibography images acquired by the Keratograph 5M device. A total of 1350 infrared meibography images were collected and annotated for model training and validation. The model demonstrated high segmentation performance, with an Intersection over Union of 81.67% (95% Confidence Interval [CI]: 81.03–82.31) and accuracy of 97.49% (95% CI: 97.38–97.62), outperforming conventional algorithms. The agreement was observed between AI-based and manual gland grading (Kappa = 0.93) and gland counting (Spearman r = 0.9334). Repeatability analysis confirmed the model’s stability, and external validation across four independent centers yielded consistent results with AUCs exceeding 0.99. This AI tool offers a standardized, efficient, and objective method for meibography image analysis, which may improve diagnostic precision and assist in the clinical management of meibomian gland dysfunction across diverse populations.https://doi.org/10.1038/s41746-025-01753-5
spellingShingle Li Li
Kunhong Xiao
Taichen Lai
Kunfeng Lai
Jiawen Lin
Zongyuan Ge
Lingyi Liang
Hao Huang
Xiaoshan Zhang
Li Liu
Yujie Wang
Xianwen Shang
Mingguang He
Ying Xue
Zhuoting Zhu
Development and multicenter validation of an AI driven model for quantitative meibomian gland evaluation
npj Digital Medicine
title Development and multicenter validation of an AI driven model for quantitative meibomian gland evaluation
title_full Development and multicenter validation of an AI driven model for quantitative meibomian gland evaluation
title_fullStr Development and multicenter validation of an AI driven model for quantitative meibomian gland evaluation
title_full_unstemmed Development and multicenter validation of an AI driven model for quantitative meibomian gland evaluation
title_short Development and multicenter validation of an AI driven model for quantitative meibomian gland evaluation
title_sort development and multicenter validation of an ai driven model for quantitative meibomian gland evaluation
url https://doi.org/10.1038/s41746-025-01753-5
work_keys_str_mv AT lili developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT kunhongxiao developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT taichenlai developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT kunfenglai developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT jiawenlin developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT zongyuange developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT lingyiliang developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT haohuang developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT xiaoshanzhang developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT liliu developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT yujiewang developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT xianwenshang developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT mingguanghe developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT yingxue developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation
AT zhuotingzhu developmentandmulticentervalidationofanaidrivenmodelforquantitativemeibomianglandevaluation