{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,4]],"date-time":"2026-07-04T18:20:46Z","timestamp":1783189246422,"version":"3.54.6"},"reference-count":25,"publisher":"SAGE Publications","license":[{"start":{"date-parts":[[2026,7,2]],"date-time":"2026-07-02T00:00:00Z","timestamp":1782950400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["Phlebology"],"abstract":"<jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>Generative artificial intelligence (AI) models capable of producing photorealistic medical images are increasingly proposed for patient education, clinical illustration, and trainee instruction. However, their ability to accurately represent anatomically distinct disease subtypes remains unclear. This study evaluated the diagnostic accuracy of a widely used generative AI model in producing images corresponding to the five anatomical lipedema types defined by the Schmeller classification.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>In this prospective audit, ChatGPT\u2019s image-generation interface was prompted to create 60 images for each lipedema type (Types I\u2013V),yielding 300 images. Prompts were standardized and limited to the subtype label without additional descriptors. Two clinicians independently classified each image into one of the five lipedema types or as indeterminate, blinded to the original prompt; disagreements were resolved by a third clinician. Diagnostic performance was assessed using a confusion matrix and per-type sensitivity, specificity, positive predictive value(PPV), negative predictive value (NPV),F1-score,and one-vs-rest receiver operating characteristic area under the curve (ROC AUC). Overall accuracy and Cohen\u2019s \u03ba statistics were also calculated.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>All 300 images were evaluable. The model generated anatomically consistent images for Types I,II, and III (sensitivity = 1.00 for each). Specificity was 1.00 for Types I and II but 0.50 for Type III because all images requested as Types IV and V were classified as Type III. Consequently, the model failed to generate any images consistent with Type IV(arm-predominant) or Type V(calf-isolated) lipedema (sensitivity = 0.00 for both). Overall accuracy was 0.600. Unweighted and quadratic-weighted Cohen\u2019s \u03ba values were 0.500 and 0.667, respectively. Micro- and macro-averaged ROC AUC were both 0.750.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The model reproduces severity gradients within lower-extremity lipedema but systematically collapses anatomically distinct subtypes into the dominant Type III phenotype, failing to depict arm-predominant and calf-isolated disease. Current generative AI systems may therefore encode lipedema as a single visual phenotype rather than a distributed anatomical entity, limiting their reliability for medical education and clinical communication.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1177\/02683555261467340","type":"journal-article","created":{"date-parts":[[2026,7,2]],"date-time":"2026-07-02T09:44:32Z","timestamp":1782985472000},"update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":0,"title":["Evaluation of generative artificial intelligence in producing anatomically distinct lipedema subtypes: A diagnostic accuracy study"],"prefix":"10.1177","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0508-8868","authenticated-orcid":false,"given":"Ilhan Celil","family":"\u00d6zbek","sequence":"first","affiliation":[{"name":"University of Health Sciences Derince Training and Research Hospital"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"B\u00fclent","family":"Alyanak","sequence":"additional","affiliation":[{"name":"G\u00f6lc\u00fck Necati \u00c7elik State Hospital"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Burak Tayyip","family":"Dede","sequence":"additional","affiliation":[{"name":"Prof. Dr. Cemil Ta\u015fc\u0131o\u011flu City Hospital"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0256-5833","authenticated-orcid":false,"given":"Mustafa H\u00fcseyin","family":"Temel","sequence":"additional","affiliation":[{"name":"University of Health Sciences Sultan Abdulhamid Han Training and Research Hospital"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mustafa Turgut","family":"Y\u0131ld\u0131zg\u00f6ren","sequence":"additional","affiliation":[{"name":"University of Necmettin Erbakan Faculty of Meram Medicine"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fatih","family":"Ba\u011fc\u0131er","sequence":"additional","affiliation":[{"name":"Physical Medicine and Rehabilitation"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"179","published-online":{"date-parts":[[2026,7,2]]},"reference":[{"key":"e_1_3_7_2_2","doi-asserted-by":"publisher","DOI":"10.7326\/0003-4819-34-5-1243"},{"key":"e_1_3_7_3_2","volume-title":"Endotext. South Dartmouth, MA: MDText","author":"Herbst KL","year":"2019","unstructured":"Herbst KL. Subcutaneous adipose tissue diseases: dercum disease, lipedema, familial multiple lipomatosis, and madelung disease. In: Feingold KR, Anawalt B, Boyce A (eds). Endotext. South Dartmouth, MA: MDText. com, 2019. PMID: 31895524."},{"key":"e_1_3_7_4_2","first-page":"268355526145157","article-title":"Prevalence of sarcopenia and its functional correlates in women with lower-extremity lipedema: a cross-sectional observational study","author":"Ozbek IC","year":"2026","unstructured":"Ozbek IC, Kuculmez O, Dundar Ahi E. Prevalence of sarcopenia and its functional correlates in women with lower-extremity lipedema: a cross-sectional observational study. Phlebology 2026: 2683555261451570. https:\/\/doi.org\/10.1177\/02683555261451570","journal-title":"Phlebology"},{"key":"e_1_3_7_5_2","doi-asserted-by":"publisher","DOI":"10.1097\/GOX.0000000000001043"},{"key":"e_1_3_7_6_2","doi-asserted-by":"publisher","DOI":"10.1111\/j.1758-8111.2012.00045.x"},{"key":"e_1_3_7_7_2","doi-asserted-by":"publisher","DOI":"10.66163\/jokohtu.1864613"},{"key":"e_1_3_7_8_2","doi-asserted-by":"publisher","DOI":"10.1046\/j.1439-0353.2004.04051.x"},{"key":"e_1_3_7_9_2","doi-asserted-by":"publisher","DOI":"10.1111\/ddg.13036"},{"key":"e_1_3_7_10_2","doi-asserted-by":"publisher","DOI":"10.12968\/jowc.2020.29.Sup11b.1"},{"key":"e_1_3_7_11_2","doi-asserted-by":"publisher","DOI":"10.1177\/0268355516639421"},{"key":"e_1_3_7_12_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-018-0300-7"},{"key":"e_1_3_7_13_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-018-0316-z"},{"key":"e_1_3_7_14_2","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMsr2214184"},{"key":"e_1_3_7_15_2","doi-asserted-by":"publisher","DOI":"10.1145\/3593013.3594095"},{"key":"e_1_3_7_16_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00283"},{"key":"e_1_3_7_17_2","doi-asserted-by":"publisher","DOI":"10.1001\/jamadermatol.2018.2348"},{"key":"e_1_3_7_18_2","doi-asserted-by":"publisher","DOI":"10.3390\/healthcare11060887"},{"issue":"1","key":"e_1_3_7_19_2","article-title":"Lipedema: friend and foe","volume":"33","author":"Torre YS","year":"2018","unstructured":"Torre YS, Wadeea R, Rosas V, et al. Lipedema: friend and foe. Horm Mol Biol Clin Invest 2018; 33(1). https:\/\/doi.org\/10.1515\/hmbci-2017-0076","journal-title":"Horm Mol Biol Clin Invest"},{"key":"e_1_3_7_20_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.knee.2025.04.008"},{"key":"e_1_3_7_21_2","doi-asserted-by":"publisher","DOI":"10.1038\/nature21056"},{"key":"e_1_3_7_22_2","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2016.17216"},{"key":"e_1_3_7_23_2","doi-asserted-by":"publisher","DOI":"10.1016\/S2589-7500(19)30123-2"},{"key":"e_1_3_7_24_2","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.h5527"},{"key":"e_1_3_7_25_2","doi-asserted-by":"publisher","DOI":"10.11613\/BM.2012.031"},{"key":"e_1_3_7_26_2","first-page":"2825","article-title":"Scikit-learn: machine learning in python","volume":"12","author":"Pedregosa F","year":"2011","unstructured":"Pedregosa F, Varoquaux G, Gramfort A, et al. Scikit-learn: machine learning in python. J Mach Learn Res 2011; 12: 2825\u20132830.","journal-title":"J Mach Learn Res"}],"container-title":["Phlebology: The Journal of Venous Disease"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/02683555261467340","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/02683555261467340","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/02683555261467340","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,4]],"date-time":"2026-07-04T17:52:04Z","timestamp":1783187524000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/02683555261467340"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,2]]},"references-count":25,"alternative-id":["10.1177\/02683555261467340"],"URL":"https:\/\/doi.org\/10.1177\/02683555261467340","relation":{},"ISSN":["0268-3555","1758-1125"],"issn-type":[{"value":"0268-3555","type":"print"},{"value":"1758-1125","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,2]]},"article-number":"02683555261467340"}}