{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T11:07:44Z","timestamp":1791544064503,"version":"4.3.4"},"reference-count":25,"publisher":"Wiley","issue":"4","license":[{"start":{"date-parts":[[2025,3,26]],"date-time":"2025-03-26T00:00:00Z","timestamp":1742947200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["J of Cosmetic Dermatology"],"published-print":{"date-parts":[[2025,4]]},"abstract":"<jats:title>ABSTRACT<\/jats:title><jats:sec><jats:title>Background<\/jats:title><jats:p>Due to high patient demand, increasing numbers of non\u2010dermatologists are performing skin assessments and carrying out laser interventions in medical spas, leading to inferior outcomes and higher complications. A machine learning tool that automatically analyzes patient skin has the potential to aid non\u2010dermatologists.<\/jats:p><\/jats:sec><jats:sec><jats:title>Aims<\/jats:title><jats:p>To develop a high\u2010performing machine learning model that predicts Fitzpatrick skin type, hyperpigmentation, redness, and wrinkle severity simultaneously.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We developed the SkinAnalysis dataset of 3662 images, labeled by a dermatologist across five skin scales. We trained and evaluated machine learning models across 15 different configurations, including three neural network architectures and two loss functions.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The best\u2010performing model was an EfficientNet\u2010V2M architecture with a custom cross entropy loss. This model's mean test set accuracy across all labels was 85.41\u2009\u00b1\u20099.86 and its mean test set AUROC was 0.8306\u2009\u00b1\u20090.09599. An interesting trend emerged in which machine learning model performance was higher at the extremes of the scales, suggesting greater clinical ambiguity in the middle of the scales.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>Machine learning models are capable of predicting multiple skin characteristics simultaneously from color photographs of the face. In the future, similar models could assist non\u2010dermatologists in patient skin evaluation to enhance treatment planning.<\/jats:p><\/jats:sec>","DOI":"10.1111\/jocd.70050","type":"journal-article","created":{"date-parts":[[2025,3,29]],"date-time":"2025-03-29T04:58:39Z","timestamp":1743224319000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":12,"title":["Artificial Intelligence Predicts Fitzpatrick Skin Type, Pigmentation, Redness, and Wrinkle Severity From Color Photographs of the Face"],"prefix":"10.1111","volume":"24","author":[{"given":"Rachel L.","family":"Draelos","sequence":"first","affiliation":[{"name":"Vismedica AI, LLC  Durham North Carolina USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chelsea E.","family":"Kesty","sequence":"additional","affiliation":[{"name":"St. Petersburg Skin and Laser  St. Petersburg Florida USA"},{"name":"Kesty AI  St. Petersburg Florida USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Katarina R.","family":"Kesty","sequence":"additional","affiliation":[{"name":"St. Petersburg Skin and Laser  St. Petersburg Florida USA"},{"name":"Kesty AI  St. Petersburg Florida USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2025,3,26]]},"reference":[{"key":"e_1_2_11_2_1","doi-asserted-by":"publisher","DOI":"10.1080\/14764170802524411"},{"key":"e_1_2_11_3_1","volume-title":"Consumer Survey on Cosmetic Dermatologic Procedures","author":"ASDS","year":"2021"},{"issue":"2","key":"e_1_2_11_4_1","first-page":"1","article-title":"Occupational Employment and Wages","author":"US Bureau of Labor Statistics","year":"2019","journal-title":"Occupational Employment Statistics"},{"key":"e_1_2_11_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/J.CLINDERMATOL.2021.11.007"},{"key":"e_1_2_11_6_1","doi-asserted-by":"publisher","DOI":"10.1097\/DSS.0000000000002344"},{"key":"e_1_2_11_7_1","doi-asserted-by":"publisher","DOI":"10.1097\/DSS.0000000000001829"},{"key":"e_1_2_11_8_1","doi-asserted-by":"publisher","DOI":"10.1186\/S12886\u2010024\u201003383\u2010Z"},{"key":"e_1_2_11_9_1","doi-asserted-by":"publisher","DOI":"10.1177\/12034754231174852"},{"key":"e_1_2_11_10_1","doi-asserted-by":"publisher","DOI":"10.1503\/CMAJ.109\u20103230"},{"key":"e_1_2_11_11_1","unstructured":"G.Team R.Anil S.Borgeaud et\u00a0al. \u201cGemini: A Family of Highly Capable Multimodal Models \u201d(2023) https:\/\/arxiv.org\/abs\/2312.11805v4."},{"key":"e_1_2_11_12_1","unstructured":"Z.Yang L.Li K.Lin et\u00a0al. \u201cThe Dawn of LMMs: Preliminary Explorations With GPT\u20104V(Ision) \u201d(2023) https:\/\/arxiv.org\/abs\/2309.17421v2."},{"key":"e_1_2_11_13_1","doi-asserted-by":"crossref","unstructured":"C. C.Chang S. T.Hsing Y. C.Chuang et\u00a0al. \u201cRobust Skin Type Classification Using Convolutional Neural Networks \u201dProceedings of the 13th IEEE Conference on Industrial Electronics and Applications ICIEA 2018. Published online June 26 2018:2011\u20132014 https:\/\/doi.org\/10.1109\/ICIEA.2018.8398040.","DOI":"10.1109\/ICIEA.2018.8398040"},{"key":"e_1_2_11_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/J.HELIYON.2023.E21176"},{"key":"e_1_2_11_15_1","doi-asserted-by":"crossref","unstructured":"M.Groh C.Harris L.Soenksen et\u00a0al. \u201cEvaluating Deep Neural Networks Trained on Clinical Images in Dermatology With the Fitzpatrick 17k Dataset \u201d2021 IEEE\/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) 2021:1820\u20131828 https:\/\/doi.org\/10.1109\/CVPRW53098.2021.00201.","DOI":"10.1109\/CVPRW53098.2021.00201"},{"key":"e_1_2_11_16_1","doi-asserted-by":"publisher","DOI":"10.1016\/J.CMPB.2024.108044"},{"key":"e_1_2_11_17_1","first-page":"1548","volume-title":"FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age for Bias Measurement and Mitigation","author":"Karkkainen K.","year":"2021"},{"key":"e_1_2_11_18_1","unstructured":"Openverse https:\/\/openverse.org\/."},{"key":"e_1_2_11_19_1","unstructured":"K.SimonyanandA.Zisserman \u201cVery Deep Convolutional Networks for Large\u2010Scale Image Recognition \u201d(2014) http:\/\/arxiv.org\/abs\/1409.1556."},{"key":"e_1_2_11_20_1","unstructured":"K.He X.Zhang S.Ren andJ.Sun \u201cDeep Residual Learning for Image Recognition \u201d(2015) http:\/\/arxiv.org\/abs\/1512.03385."},{"key":"e_1_2_11_21_1","unstructured":"M.TanandQ. V.Le \u201cEfficientNet: Rethinking Model Scaling for Convolutional Neural Networks \u201d36th International Conference on Machine Learning ICML 2019. 2019; 2019 June: 10691\u201310700 https:\/\/arxiv.org\/abs\/1905.11946v5."},{"key":"e_1_2_11_22_1","volume-title":"ImageNet: A Large\u2010Scale Hierarchical Image Database","author":"Deng J.","year":"2009"},{"key":"e_1_2_11_23_1","doi-asserted-by":"publisher","DOI":"10.1016\/J.MEDIA.2020.101857"},{"key":"e_1_2_11_24_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2017.10.009"},{"key":"e_1_2_11_25_1","unstructured":"L.PerezandJ.Wang \u201cThe Effectiveness of Data Augmentation in Image Classification Using Deep Learning \u201d(2017) https:\/\/arxiv.org\/abs\/1712.04621v1."},{"key":"e_1_2_11_26_1","doi-asserted-by":"crossref","unstructured":"J.Cheng Z.Wang andG.Pollastri \u201cA Neural Network Approach to Ordinal Regression \u201dProceedings of the International Joint Conference on Neural Networks. Published online April 8 2007:1279\u20131284 https:\/\/doi.org\/10.1109\/IJCNN.2008.4633963.","DOI":"10.1109\/IJCNN.2008.4633963"}],"container-title":["Journal of Cosmetic Dermatology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/jocd.70050","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,25]],"date-time":"2025-09-25T06:37:43Z","timestamp":1758782263000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/jocd.70050"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,26]]},"references-count":25,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2025,4]]}},"alternative-id":["10.1111\/jocd.70050"],"URL":"https:\/\/doi.org\/10.1111\/jocd.70050","archive":["Portico"],"relation":{},"ISSN":["1473-2130","1473-2165"],"issn-type":[{"value":"1473-2130","type":"print"},{"value":"1473-2165","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,3,26]]},"assertion":[{"value":"2024-12-19","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-02-03","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-03-26","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70050"}}