{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T15:02:47Z","timestamp":1775142167050,"version":"3.50.1"},"reference-count":31,"publisher":"MDPI AG","issue":"7","license":[{"start":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T00:00:00Z","timestamp":1775088000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background: Melanoma remains a leading cause of cancer-related mortality, with early detection being the primary determinant of survival. The emergence of MLLMs offers a potential paradigm shift in accessible screening. However, the diagnostic reliability and safety of these general-purpose models in oncology remain insufficiently characterized. Methods: This study performed a head-to-head comparison of GPT-5, Gemini 3, and Grok 4 to evaluate their efficacy as first-level screening tools for cutaneous melanoma. A retrospective analysis was conducted using a balanced dataset of 100 clinical images (50 histopathologically confirmed benign, 50 malignant) from the ISIC archive. Results: Gemini 3 achieved the highest overall accuracy (71%) and specificity (94%), while Grok 4 demonstrated the highest sensitivity (52%). All models exhibited a critical deficit in sensitivity, missing approximately half of the malignant lesions. Statistical testing revealed no significant performance differences between the models (p &gt; 0.05). Notably, Gemini 3 exhibited severe overconfidence, maintaining a high CI (84.62%) even during false-negative predictions, whereas GPT-5 and Grok 4 showed better calibration with a significant drop in confidence upon incorrect diagnosis. Conclusions: While current MLLMs possess a foundational capacity for dermatological analysis, their unacceptably low sensitivity and potential for overconfident misdiagnosis render them unsafe as standalone screening tools. At present, MLLMs should only be utilized as complementary tools under strict clinical supervision.<\/jats:p>","DOI":"10.3390\/diagnostics16071077","type":"journal-article","created":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T14:09:54Z","timestamp":1775138994000},"page":"1077","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["From ABCD to AI: Assessing the Diagnostic Reliability of MLLMs in Cutaneous Melanoma Screening\u2014A Head-to-Head Comparison"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-1641-3671","authenticated-orcid":false,"given":"R\u0103zvan Ioan","family":"Andrei","sequence":"first","affiliation":[{"name":"Department of General Surgery, \u201cCarol Davila\u201d University of Medicine and Pharmacy, B-dul Eroii Sanitari 8, 050474 Bucharest, Romania"},{"name":"Department of Surgical Oncology, Institute of Oncology \u201cProf. Dr. Al. Trestioreanu\u201d, \u015eos. Fundeni 252, 022328 Bucharest, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Aniela Roxana","family":"Nodi\u021bi-Cuc","sequence":"additional","affiliation":[{"name":"Department of General Surgery, \u201cCarol Davila\u201d University of Medicine and Pharmacy, B-dul Eroii Sanitari 8, 050474 Bucharest, Romania"},{"name":"Department of Surgical Oncology, Institute of Oncology \u201cProf. Dr. Al. Trestioreanu\u201d, \u015eos. Fundeni 252, 022328 Bucharest, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Silviu Cristian","family":"Voinea","sequence":"additional","affiliation":[{"name":"Department of General Surgery, \u201cCarol Davila\u201d University of Medicine and Pharmacy, B-dul Eroii Sanitari 8, 050474 Bucharest, Romania"},{"name":"Department of Surgical Oncology, Institute of Oncology \u201cProf. Dr. Al. Trestioreanu\u201d, \u015eos. Fundeni 252, 022328 Bucharest, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1325-2031","authenticated-orcid":false,"given":"Cristian Ioan","family":"Bordea","sequence":"additional","affiliation":[{"name":"Department of General Surgery, \u201cCarol Davila\u201d University of Medicine and Pharmacy, B-dul Eroii Sanitari 8, 050474 Bucharest, Romania"},{"name":"Department of Surgical Oncology, Institute of Oncology \u201cProf. Dr. Al. Trestioreanu\u201d, \u015eos. Fundeni 252, 022328 Bucharest, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4320-9704","authenticated-orcid":false,"given":"Alexandru","family":"Blidaru","sequence":"additional","affiliation":[{"name":"Department of General Surgery, \u201cCarol Davila\u201d University of Medicine and Pharmacy, B-dul Eroii Sanitari 8, 050474 Bucharest, Romania"},{"name":"Department of Surgical Oncology, Institute of Oncology \u201cProf. Dr. Al. Trestioreanu\u201d, \u015eos. Fundeni 252, 022328 Bucharest, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,4,2]]},"reference":[{"key":"ref_1","first-page":"10","article-title":"Cancer statistics, 2025","volume":"75","author":"Siegel","year":"2025","journal-title":"CA Cancer J. Clin."},{"key":"ref_2","unstructured":"Ferlay, J., Ervik, M., Lam, F., Laversanne, M., Colombet, M., Mery, L., Pi\u00f1eros, M., Znaor, A., Soerjomataram, I., and Bray, F. (2024). Global Cancer Observatory: Cancer Today, International Agency for Research on Cancer."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"366","DOI":"10.1080\/15398285.2012.701177","article-title":"Cancer Facts & Figures 2012. American Cancer Society (ACS)","volume":"16","author":"Tarver","year":"2012","journal-title":"J. Consum. Health Internet"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"669","DOI":"10.1016\/S0190-9622(98)70195-1","article-title":"An estimate of the annual direct cost of treating cutaneous melanoma","volume":"38","author":"Tsao","year":"1998","journal-title":"J. Am. Acad. Dermatol."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"411","DOI":"10.1016\/j.ijwd.2021.05.005","article-title":"Skin cancer discovery during total body skin examinations","volume":"7","author":"Jiang","year":"2021","journal-title":"Int. J. Women\u2019s Dermatol."},{"key":"ref_6","first-page":"130","article-title":"Early detection of malignant melanoma: The role of physician examination and self-examination of the skin","volume":"35","author":"Friedman","year":"1985","journal-title":"CA Cancer J. Clin."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"2771","DOI":"10.1001\/jama.292.22.2771","article-title":"Early diagnosis of cutaneous melanoma: Revisiting the ABCD criteria","volume":"292","author":"Abbasi","year":"2004","journal-title":"JAMA"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"103","DOI":"10.1001\/archderm.134.1.103-a","article-title":"The \u2018ugly duckling\u2019 sign: Identification of the common characteristics of nevi in an individual as a basis for melanoma screening","volume":"134","author":"Grob","year":"1998","journal-title":"Arch. Dermatol."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"j2813","DOI":"10.1136\/bmj.j2813","article-title":"Pathologists\u2019 diagnosis of invasive melanoma and melanocytic proliferations: Observer accuracy and reproducibility study","volume":"357","author":"Elmore","year":"2017","journal-title":"BMJ"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"570","DOI":"10.1111\/j.1365-2133.2006.07396.x","article-title":"Micro-melanoma detection: A clinical study on 206 consecutive cases of pigmented skin lesions with a diameter \u2264 3 mm","volume":"155","author":"Bono","year":"2006","journal-title":"Br. J. Dermatol."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"115","DOI":"10.1038\/nature21056","article-title":"Dermatologist-level classification of skin cancer with deep neural networks","volume":"542","author":"Esteva","year":"2017","journal-title":"Nature"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"e55508","DOI":"10.2196\/55508","article-title":"Assessing the Utility of Multimodal Large Language Models (GPT-4 Vision and Large Language and Vision Assistant) in Identifying Melanoma Across Different Skin Tones","volume":"7","author":"Cirone","year":"2024","journal-title":"JMIR Dermatol."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"5649","DOI":"10.1038\/s41467-024-50043-3","article-title":"Pre-trained multimodal large language model enhances dermatological diagnosis using SkinGPT-4","volume":"15","author":"Zhou","year":"2024","journal-title":"Nat. Commun."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"e54704","DOI":"10.2196\/54704","article-title":"A Preliminary Checklist (METRICS) to Standardize the Design and Reporting of Studies on Generative Artificial Intelligence-Based Models in Health Care Education and Practice: Development Study Involving a Literature Review","volume":"13","author":"Sallam","year":"2024","journal-title":"Interact. J. Med. Res."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Karampinis, E., Zoumpourli, C.-M., Kontogianni, C., Arkoumanis, T., Koumaki, D., Mantzaris, D., Filippakis, K., Papadopoulou, M.-M., Theofili, M., and Enechukwu, N.A. (2026). Dermatology \u201cAI Babylon\u201d: Cross-Language Evaluation of AI-Crafted Dermatology Descriptions. Medicina, 62.","DOI":"10.3390\/medicina62010227"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"551","DOI":"10.1016\/j.det.2017.06.014","article-title":"Smartphone-Based Applications for Skin Monitoring and Melanoma Detection","volume":"35","author":"Chao","year":"2017","journal-title":"Dermatol. Clin."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"e175","DOI":"10.1111\/ajd.12599","article-title":"Efficacy of smartphone applications in high-risk pigmented lesions","volume":"59","author":"Ngoo","year":"2018","journal-title":"Australas. J. Dermatol."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"m127","DOI":"10.1136\/bmj.m127","article-title":"Algorithm based smartphone apps to assess risk of skin cancer in adults: Systematic review of diagnostic accuracy studies","volume":"368","author":"Freeman","year":"2020","journal-title":"BMJ"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1229","DOI":"10.1038\/s41591-020-0942-0","article-title":"Human\u2013computer collaboration for skin cancer recognition","volume":"26","author":"Tschandl","year":"2020","journal-title":"Nat. Med."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"e582","DOI":"10.1111\/jdv.16382","article-title":"Sensitivity and specificity of SkinVision are likely to have been overestimated","volume":"34","author":"Deeks","year":"2020","journal-title":"J. Eur. Acad. Dermatol. Venereol."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"648","DOI":"10.1111\/jdv.15935","article-title":"Accuracy of a smartphone application for triage of skin lesions based on machine learning algorithms","volume":"34","author":"Udrea","year":"2020","journal-title":"J. Eur. Acad. Dermatol. Venereol."},{"key":"ref_22","unstructured":"CIVIO (2026, March 13). Mole or Cancer? The Algorithm that Gets One in Three Melanomas Wrong and Erases Patients with Dark Skin. Available online: https:\/\/civio.es\/sanidad\/2025\/07\/03\/mole-or-cancer-the-algorithm-that-gets-one-in-three-melanomas-wrong-and-erases-patients-with-dark-skin\/."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Martin-Gonzalez, M., Azcarraga, C., Martin-Gil, A., Carpena-Torres, C., and Jaen, P. (2022). Efficacy of a Deep Learning Convolutional Neural Network System for Melanoma Diagnosis in a Hospital Population. Int. J. Environ. Res. Public Health, 19.","DOI":"10.3390\/ijerph19073892"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"e64","DOI":"10.1016\/S2589-7500(21)00252-1","article-title":"Characteristics of publicly available skin cancer image datasets: A systematic review","volume":"4","author":"Wen","year":"2022","journal-title":"Lancet Digit. Health"},{"key":"ref_25","first-page":"e94909","article-title":"Exploring the Diagnostic Capability of Artificial Intelligence in Dermatology for Darker Skin Tones: A Narrative Review","volume":"17","author":"Dowie","year":"2025","journal-title":"Cureus"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"S26","DOI":"10.1016\/j.jaad.2011.05.034","article-title":"Racial and ethnic variations in incidence and survival of cutaneous melanoma in the United States, 1999\u20132006","volume":"65","author":"Wu","year":"2011","journal-title":"J. Am. Acad. Dermatol."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Yang, R., Gao, S., and Jiang, Y. (2024). Digital divide as a determinant of health in the U.S. older adults: Prevalence, trends, and risk factors. BMC Geriatr., 24.","DOI":"10.1186\/s12877-024-05612-y"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"156","DOI":"10.1038\/s41746-024-01161-1","article-title":"Learnings from the first AI-enabled skin cancer device for primary care authorized by FDA","volume":"7","author":"Venkatesh","year":"2024","journal-title":"npj Digit. Med."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"701","DOI":"10.1007\/s10103-019-02894-2","article-title":"Elastic scattering spectroscopy for monitoring skin cancer transformation and therapy in the near infrared window","volume":"35","author":"Shurrab","year":"2020","journal-title":"Lasers Med. Sci."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"21501319251342106","DOI":"10.1177\/21501319251342106","article-title":"DERM-SUCCESS FDA Pivotal Study: A Multi-Reader Multi-Case Evaluation of Primary Care Physicians\u2019 Skin Cancer Detection Using AI-Enabled Elastic Scattering Spectroscopy","volume":"16","author":"Ferris","year":"2025","journal-title":"J. Prim. Care Community Health"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"e047709","DOI":"10.1136\/bmjopen-2020-047709","article-title":"Developing a reporting guideline for artificial intelligence-centred diagnostic test accuracy studies: The STARD-AI protocol","volume":"11","author":"Sounderajah","year":"2021","journal-title":"BMJ Open"}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/7\/1077\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T14:19:43Z","timestamp":1775139583000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/7\/1077"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,2]]},"references-count":31,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,4]]}},"alternative-id":["diagnostics16071077"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16071077","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,2]]}}}