{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,29]],"date-time":"2025-10-29T12:39:50Z","timestamp":1761741590730,"version":"build-2065373602"},"reference-count":27,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,10,29]],"date-time":"2025-10-29T00:00:00Z","timestamp":1761696000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>\n                    In recent years, numerous advanced image segmentation algorithms have been employed in the analysis of meibomian glands (MG). However, their clinical utility remains limited due to insufficient integration with the diagnostic and grading processes of meibomian gland dysfunction (MGD). To bridge this gap, the present study leverages three state-of-the-art deep learning models\u2014DeepLabV3+, U-Net, and U-Net++\u2014to segment infrared MG images and extract quantitative features for MGD diagnosis and severity assessment. A comprehensive set of morphological (e.g., gland area, width, length, and distortion) and distributional (e.g., gland density, count, inter-gland distance, disorder degree, and loss ratio) indicators were derived from the segmentation outcomes. Spearman correlation analysis revealed significant positive associations between most indicators and MGD severity (correlation coefficients ranging from 0.26 to 0.58;\n                    <jats:italic>p<\/jats:italic>\n                    \u202f&amp;lt;\u202f0.001), indicating their potential diagnostic value. Furthermore, Box plot analysis highlighted clear distribution differences in the majority of indicators across all grades, with medians shifting progressively, interquartile ranges widening, and an increase in outliers, reflecting morphological changes associated with disease progression. Logistic regression models trained on these quantitative features yielded area under the receiver operating characteristic curve (AUC) values of 0.89\u202f\u00b1\u202f0.02, 0.76\u202f\u00b1\u202f0.03, 0.85\u202f\u00b1\u202f0.02, and 0.94\u202f\u00b1\u202f0.01 for MGD grades 0, 1, 2, and 3, respectively. The models demonstrated strong classification performance, with micro-average and macro-average AUCs of 0.87\u202f\u00b1\u202f0.02 and 0.86\u202f\u00b1\u202f0.03, respectively. Model stability and generalizability were validated through 5-fold cross-validation. Collectively, these findings underscore the clinical relevance and robustness of deep learning-assisted quantitative analysis for the objective diagnosis and grading of MGD, offering a promising framework for automated medical image interpretation in ophthalmology.\n                  <\/jats:p>","DOI":"10.3389\/frai.2025.1642361","type":"journal-article","created":{"date-parts":[[2025,10,29]],"date-time":"2025-10-29T07:06:57Z","timestamp":1761721617000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Quantitative evaluation of meibomian gland dysfunction via deep learning-based infrared image segmentation"],"prefix":"10.3389","volume":"8","author":[{"given":"Ziyang","family":"Yu","sequence":"first","affiliation":[]},{"given":"Zhijun","family":"Wei","sequence":"additional","affiliation":[]},{"given":"Mini Han","family":"Wang","sequence":"additional","affiliation":[]},{"given":"Jiazheng","family":"Cui","sequence":"additional","affiliation":[]},{"given":"Jiaxiang","family":"Tan","sequence":"additional","affiliation":[]},{"given":"Yang","family":"Xu","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2025,10,29]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1016\/j.jtos.2012.09.005","article-title":"Morphological evaluation of Meibomian glands using noncontact infrared Meibography","volume":"11","author":"Ban","year":"2013","journal-title":"Ocular Surface"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"300","DOI":"10.1136\/bjophthalmol-2015-307415","article-title":"Revisiting the vicious circle of dry eye disease: a focus on the pathophysiology of meibomian gland dysfunction","volume":"100","author":"Baudouin","year":"2016","journal-title":"Br. J. Ophthalmol."},{"key":"ref3","doi-asserted-by":"publisher","first-page":"834","DOI":"10.1109\/TPAMI.2017.2699184","article-title":"DeepLab: semantic image segmentation with deep convolutional nets, Atrous convolution, and fully connected Crfs","volume":"40","author":"Chen","year":"","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref4","first-page":"833","volume-title":"Encoder-decoder with Atrous separable convolution for semantic image segmentation","author":"Chen","year":""},{"key":"ref5","doi-asserted-by":"publisher","first-page":"276","DOI":"10.1016\/j.jtos.2017.05.008","article-title":"Tfos dews ii definition and classification report","volume":"15","author":"Craig","year":"2017","journal-title":"Ocul. Surf."},{"key":"ref6","doi-asserted-by":"publisher","first-page":"101132","DOI":"10.1016\/j.eclinm.2021.101132","article-title":"Quantitative analysis of morphological and functional features in Meibography for Meibomian gland dysfunction: diagnosis and grading","volume":"40","author":"Deng","year":"2021","journal-title":"Eclinical Med"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"1527010","DOI":"10.3389\/frai.2025.1527010","article-title":"Applied machine learning in intelligent systems: knowledge graph-enhanced ophthalmic contrastive learning with \"clinical profile\" prompts","volume":"8","author":"Han Wang","year":"2025","journal-title":"Front. Artif. Intell."},{"key":"ref8","doi-asserted-by":"publisher","first-page":"e000436","DOI":"10.1136\/bmjophth-2020-000436","article-title":"Image based analysis of meibomian gland dysfunction using conditional generative adversarial neural network","volume":"6","author":"Khan","year":"2021","journal-title":"Bmj Open Ophthalmol"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"086008","DOI":"10.1117\/1.JBO.17.8.086008","article-title":"Detection of meibomian glands and classification of meibography images","volume":"17","author":"Koh","year":"2012","journal-title":"J. Biomed. Opt."},{"key":"ref10","first-page":"154","volume-title":"Automatic Meibomian gland segmentation and assessment based on TransUnet with data augmentation","author":"Lai","year":"2024"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"34","DOI":"10.1167\/tvst.9.9.34","article-title":"A novel quantitative index of Meibomian gland dysfunction, the Meibomian gland tortuosity","volume":"9","author":"Lin","year":"2020","journal-title":"Transl. Vis. Sci. Technol."},{"key":"ref12","doi-asserted-by":"publisher","first-page":"17","DOI":"10.1167\/tvst.8.4.17","article-title":"A novel automated approach for infrared-based assessment of Meibomian gland morphology","volume":"8","author":"Llorens-Quintana","year":"2019","journal-title":"Transl. Vis. Sci. Technol."},{"key":"ref13","doi-asserted-by":"publisher","first-page":"102","DOI":"10.1016\/j.zemedi.2018.11.002","article-title":"An overview of deep learning in medical imaging focusing on Mri","volume":"29","author":"Lundervold","year":"2019","journal-title":"Z. Med. Phys."},{"key":"ref14","first-page":"234","volume-title":"U-net: Convolutional networks for biomedical image segmentation","author":"Ronneberger","year":"2015"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"283","DOI":"10.1016\/j.jtos.2022.06.006","article-title":"Automated quantification of meibomian gland dropout in infrared meibography using deep learning","volume":"26","author":"Saha","year":"2022","journal-title":"Ocul. Surf."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"7649","DOI":"10.1038\/s41598-021-87314-8","article-title":"Deep learning-based automatic meibomian gland segmentation and morphology assessment in infrared meibography","volume":"11","author":"Setu","year":"2021","journal-title":"Sci. Rep."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"334","DOI":"10.1016\/j.jtos.2017.05.003","article-title":"Tfos dews II epidemiology report","volume":"15","author":"Stapleton","year":"2017","journal-title":"Ocul. Surf."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"100334","DOI":"10.1016\/j.xops.2023.100334","article-title":"A deep learning approach for Meibomian gland appearance evaluation","volume":"3","author":"Swiderska","year":"2023","journal-title":"Ophthalmol Sci"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"2006","DOI":"10.1167\/iovs.10-6997f","article-title":"The international workshop on Meibomian gland dysfunction: report of the diagnosis subcommittee","volume":"52","author":"Tomlinson","year":"2011","journal-title":"Invest. Ophthalmol. Vis. Sci."},{"key":"ref20","doi-asserted-by":"publisher","first-page":"2697","DOI":"10.3390\/electronics12122697","article-title":"An explainable artificial intelligence-based robustness optimization approach for age-related macular degeneration detection based on medical Iot systems","volume":"12","author":"Wang","year":"","journal-title":"Electronics"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"1517918","DOI":"10.3389\/frai.2025.1517918","article-title":"Balancing accuracy and user satisfaction: the role of prompt engineering in AI-driven healthcare solutions","volume":"8","author":"Wang","year":"","journal-title":"Front. Artif. Intell."},{"key":"ref22","first-page":"64","article-title":"Leveraging artificial intelligence and clinical laboratory evidence to advance Mobile health applications in ophthalmology: taking the ocular surface disease as a case study","volume-title":"iLABMED","author":"Wang","year":""},{"key":"ref23","doi-asserted-by":"publisher","first-page":"445","DOI":"10.26599\/BDMA.2023.9020024","article-title":"Ai-based advanced approaches and dry eye disease detection based on multi-source evidence: cases, applications, issues, and future directions","volume":"7","author":"Wang","year":"2024","journal-title":"Big Data Min. Anal."},{"key":"ref24","doi-asserted-by":"publisher","first-page":"012025","DOI":"10.1088\/1742-6596\/2650\/1\/012025","article-title":"Can explainable artificial intelligence optimize the data quality of machine learning model? Taking Meibomian gland dysfunction detections as a case study","volume":"2650","author":"Wang","year":"","journal-title":"J. Phys. Conf. Ser."},{"key":"ref25","doi-asserted-by":"publisher","first-page":"539","DOI":"10.1016\/j.jtos.2017.05.001","article-title":"Tfos dews ii diagnostic methodology report","volume":"15","author":"Wolffsohn","year":"2017","journal-title":"Ocul. Surf."},{"key":"ref26","doi-asserted-by":"publisher","first-page":"17345","DOI":"10.1038\/s41598-019-54013-4","article-title":"Diagnostic test efficacy of Meibomian gland morphology and function","volume":"9","author":"Xiao","year":"2019","journal-title":"Sci. Rep."},{"key":"ref27","first-page":"3","volume-title":"Unet++: A nested U-net architecture for medical image segmentation","author":"Zhou","year":"2018"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1642361\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,29]],"date-time":"2025-10-29T07:06:58Z","timestamp":1761721618000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1642361\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,29]]},"references-count":27,"alternative-id":["10.3389\/frai.2025.1642361"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1642361","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,29]]},"article-number":"1642361"}}