{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T15:15:17Z","timestamp":1780326917392,"version":"3.54.1"},"reference-count":38,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T00:00:00Z","timestamp":1778457600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Med."],"abstract":"<jats:p>\n                    Rosacea is a chronic inflammatory skin disorder characterized by persistent facial erythema. Its clinical assessment relies on the Clinician\u2019s Erythema Assessment (CEA), a subjective scale prone to inter-observer variability. To address the need for diagnostic consistency, this study developed a multimodal artificial intelligence framework for objective CEA grading using standardized VISIA\u00ae imaging. We analyzed a retrospective cohort of 1,001 patients. To establish a robust reference standard, three expert dermatologists conducted a multi-step collective audit to reach a unanimous consensus for each case. The framework integrated handcrafted image-derived tabular features with deep learning representations. During training, spatial data augmentations and Focal Loss were implemented to address dataset imbalance and mitigate overfitting. Our results demonstrated that the multimodal fusion model achieved statistically significant improvements over the strong image-only baseline (McNemar\u2019s\n                    <jats:inline-formula>\n                      <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"M1\">\n                        <mml:mi>p<\/mml:mi>\n                        <mml:mo>=<\/mml:mo>\n                        <mml:mn>0.031<\/mml:mn>\n                      <\/mml:math>\n                    <\/jats:inline-formula>\n                    ; DeLong\u2019s\n                    <jats:inline-formula>\n                      <mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\" id=\"M2\">\n                        <mml:mi>p<\/mml:mi>\n                        <mml:mo>=<\/mml:mo>\n                        <mml:mn>0.024<\/mml:mn>\n                      <\/mml:math>\n                    <\/jats:inline-formula>\n                    ), yielding a Macro-AUC of 0.902 (95% CI: 0.862\u20130.937). Furthermore, to address the ordinal nature of the disease severity, the fusion model achieved a Quadratic Weighted Kappa (QWK) of 0.800 and an Intraclass Correlation Coefficient (ICC) of 0.801 (95% CI: 0.720\u20130.860), indicating excellent alignment with expert consensus. Error analysis revealed that over 95% of misclassifications in intermediate grades (CEA3) were restricted to adjacent categories, reflecting strong clinical safety. Interpretability analysis via layer-wise relevance propagation confirmed the model\u2019s focus on clinically recognized erythema-prone regions.\n                  <\/jats:p>\n                  <jats:p>This study establishes a robust proof-of-concept tool that transforms rosacea assessment from subjective inspection into an objective digital measurement, offering significant translational potential for clinical trials and teledermatology.<\/jats:p>","DOI":"10.3389\/fmed.2026.1829629","type":"journal-article","created":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T05:29:46Z","timestamp":1778477386000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Objective assessment of rosacea erythema severity: a multimodal artificial intelligence framework integrating VISIA\u00ae imaging and image-derived tabular features"],"prefix":"10.3389","volume":"13","author":[{"given":"Jun","family":"Zhang","sequence":"first","affiliation":[{"name":"Department of Dermatology, Wuhan No. 1 Hospital","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peijian","family":"Hong","sequence":"additional","affiliation":[{"name":"Department of Dermatology, Wuhan No. 1 Hospital","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ning","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Dermatology, Wuhan No. 1 Hospital","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yang","family":"Liu","sequence":"additional","affiliation":[{"name":"Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yue","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Dermatology, Wuhan No. 1 Hospital","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"LiuQing","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Dermatology, Wuhan No. 1 Hospital","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,5,11]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"1722","DOI":"10.1016\/j.jaad.2018.08.049","article-title":"Global epidemiology and clinical spectrum of rosacea, highlighting skin of color: review and clinical practice experience","volume":"80","author":"Alexis","year":"2018","journal-title":"J Am Acad Dermatol"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"617","DOI":"10.1001\/jamadermatol.2022.0526","article-title":"Assessment of frequency of Rosacea subtypes in patients with Rosacea: a systematic review and Meta-analysis","volume":"158","author":"Barakji","year":"2022","journal-title":"JAMA Dermatol"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"1501","DOI":"10.1016\/j.jaad.2020.01.077","article-title":"Standard management options for rosacea: the 2019 update by the National Rosacea Society expert committee","volume":"82","author":"Thiboutot","year":"2020","journal-title":"J Am Acad Dermatol"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"1754","DOI":"10.1056\/NEJMcp1506630","article-title":"Rosacea","volume":"377","author":"Van Zuuren","year":"2017","journal-title":"N Engl J Med"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"8737","DOI":"10.1038\/s41467-024-52946-7","article-title":"Single-cell transcriptomics reveals aberrant skin-resident cell populations and identifies fibroblasts as a determinant in rosacea","volume":"15","author":"Chen","year":"2024","journal-title":"Nat Commun"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"3958","DOI":"10.1038\/s41467-023-39761-2","article-title":"Whole genome sequencing identifies genetic variants associated with neurogenic inflammation in rosacea","volume":"14","author":"Deng","year":"2023","journal-title":"Nat Commun"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"178","DOI":"10.1177\/12034754241229365","article-title":"Rosacea: pathogenesis and therapeutic correlates","volume":"28","author":"Geng","year":"2024","journal-title":"J Cutan Med Surg"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"2580","DOI":"10.1016\/j.jid.2022.03.026","article-title":"Aging-conferred SIRT7 decline inhibits Rosacea-like skin inflammation by modulating toll-like receptor 2\u2013NF-\u03baB signaling","volume":"142","author":"Li","year":"2022","journal-title":"J Invest Dermatol"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"232","DOI":"10.1186\/s12920-024-02008-0","article-title":"Construction of a molecular diagnostic system for neurogenic rosacea by combining transcriptome sequencing and machine learning","volume":"17","author":"Mao","year":"2024","journal-title":"BMC Med Genet"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"101","DOI":"10.1016\/j.jdermsci.2025.05.005","article-title":"17\u03b2-estradiol promotes LL37-induced rosacea-like skin inflammation via G protein-coupled estrogen receptor 30","volume":"119","author":"Tang","year":"2025","journal-title":"J Dermatol Sci"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"505","DOI":"10.1007\/s00403-025-03898-5","article-title":"Oxidative stress mediated by the NOX2\/ROS\/NF-\u03baB signaling axis is involved in rosacea","volume":"317","author":"Wang","year":"2025","journal-title":"Arch Dermatol Res"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"1367994","DOI":"10.3389\/fimmu.2024.1367994","article-title":"Signaling pathways and targeted therapy for rosacea","volume":"15","author":"Yang","year":"2024","journal-title":"Front Immunol"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"33","DOI":"10.1016\/j.jid.2023.05.028","article-title":"Integrated omics reveal the molecular characterization and pathogenic mechanism of Rosacea","volume":"144","author":"Zhang","year":"2024","journal-title":"J Invest Dermatol"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"e70172","DOI":"10.1111\/srt.70172","article-title":"A study on the difference in aging characteristics of sensitive and non-sensitive skin","volume":"31","author":"Guihua","year":"2025","journal-title":"Skin Res Technol"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"337","DOI":"10.1111\/ijd.16946","article-title":"Clinical characteristics of the well-defined upper eyelid vascular network pattern in patients with rosacea","volume":"63","author":"Peng","year":"2024","journal-title":"Int J Dermatol"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"7397","DOI":"10.1109\/TVCG.2024.3456213","article-title":"Expressive 3D facial animation generation based on local-to-global latent diffusion","volume":"30","author":"Song","year":"2024","journal-title":"IEEE Trans Vis Comput Graph"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"178","DOI":"10.1109\/TNSRE.2023.3346955","article-title":"B2-ViT net: broad vision transformer network with broad attention for seizure prediction","volume":"32","author":"Shi","year":"2024","journal-title":"IEEE Trans Neural Syst Rehabil Eng"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"e0130140","DOI":"10.1371\/journal.pone.0130140","article-title":"On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation","volume":"10","author":"Bach","year":"2015","journal-title":"PLoS One"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"760","DOI":"10.1016\/j.jaad.2014.05.044","article-title":"Reliability of clinician erythema assessment grading scale","volume":"71","author":"Tan","year":"2014","journal-title":"J Am Acad Dermatol"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"126372","DOI":"10.1016\/j.eswa.2024.126372","article-title":"Class distance weighted cross entropy loss for classification of disease severity","volume":"269","author":"Polat","year":"2025","journal-title":"Expert Syst Appl"},{"key":"ref21","first-page":"248","author":"Deng","year":"2009"},{"key":"ref22","volume-title":"EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks","author":"Tan","year":"2020"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"841","DOI":"10.1214\/08-AOAS169","article-title":"Random survival forests","volume":"2","author":"Ishwaran","year":"2008","journal-title":"Ann Appl Stat"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"273","DOI":"10.1007\/BF00994018","article-title":"Support-vector networks","volume":"20","author":"Cortes","year":"1995","journal-title":"Mach Learn"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"370","DOI":"10.2307\/2344614","article-title":"Generalized linear models","volume":"135","author":"Nelder","year":"1972","journal-title":"Journal of the Royal Statistical Society Series A (General)"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1109\/TIT.1967.1053964","article-title":"Nearest neighbor pattern classification","volume":"13","author":"Cover","year":"1967","journal-title":"IEEE Trans Inf Theory"},{"key":"ref27","volume-title":"Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"Chen","year":"2016"},{"key":"ref28","article-title":"Gray level co-occurrence matrices: generalisation and some new features","author":"V","year":"2012"},{"key":"ref29","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-31438-4_10","article-title":"From local binary patterns to pixel difference networks for efficient visual representation learning","author":"Su","year":"2023"},{"key":"ref30","article-title":"A simple framework for contrastive learning of visual representations","author":"Chen","year":"2020"},{"key":"ref31","first-page":"1722","author":"Alexis","year":"2018"},{"key":"ref32","doi-asserted-by":"crossref","DOI":"10.1016\/j.jeconom.2024.105809","article-title":"On LASSO for high dimensional predictive regression","author":"Mei","year":"2024"},{"key":"ref33","article-title":"PyTorch: an imperative style, high-performance deep learning library","author":"Paszke","year":"2019"},{"key":"ref34","article-title":"Scikit-learn: machine learning in Python","author":"Pedregosa","year":"2018"},{"key":"ref35","article-title":"cuDNN: efficient primitives for deep learning","author":"Chetlur","year":"2014"},{"key":"ref36","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-08277-1_17","article-title":"Scheduling techniques for liver segmentation: ReduceLRonPlateau vs OneCycleLR","author":"Al-Kababji","year":"2022"},{"key":"ref37","doi-asserted-by":"publisher","first-page":"740","DOI":"10.1111\/srt.13194","article-title":"Effectiveness of VISIA system in evaluating the severity of rosacea","volume":"28","author":"Pan","year":"2022","journal-title":"Skin Res Technol"},{"key":"ref38","doi-asserted-by":"publisher","first-page":"180","DOI":"10.1038\/s41746-023-00914-8","article-title":"Systematic review of deep learning image analyses for the diagnosis and monitoring of skin disease","volume":"6","author":"Choy","year":"2023","journal-title":"NPJ Digit Med"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1829629\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T14:40:49Z","timestamp":1780324849000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1829629\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,11]]},"references-count":38,"alternative-id":["10.3389\/fmed.2026.1829629"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2026.1829629","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,11]]},"article-number":"1829629"}}