Evaluation of multimodal large language models for psoriasis diagnosis, severity grading, and treatment recommendations from clinical photographs: ChatGPT shows superior performance compared to other large language models

Mehdi Boostani; Paolo Gisondi; Francesco Bellinato; Tara Kiss; Christos C Zouboulis; Drew Kuraitis; Noah Goldfarb; Nóra Nádudvari; Banu Farabi; Balazs Hodosi; Péter Holló; Norbert M Wikonkal; Kende Lőrincz; András Banvölgyi; Gyorgy Paragh; Norbert Kiss

doi:10.3389/fmed.2026.1791488

Evaluation of multimodal large language models for psoriasis diagnosis, severity grading, and treatment recommendations from clinical photographs: ChatGPT shows superior performance compared to other large language models

Front Med (Lausanne). 2026 May 13:13:1791488. doi: 10.3389/fmed.2026.1791488. eCollection 2026.

Authors

Mehdi Boostani¹, Paolo Gisondi², Francesco Bellinato², Tara Kiss³, Christos C Zouboulis⁴, Drew Kuraitis¹, Noah Goldfarb^{5

6}, Nóra Nádudvari³, Banu Farabi⁷, Balazs Hodosi⁸, Péter Holló³, Norbert M Wikonkal^{3

9}, Kende Lőrincz³, András Banvölgyi³, Gyorgy Paragh¹, Norbert Kiss³

Affiliations

¹ Department of Dermatology, Roswell Park Comprehensive Cancer Center, Buffalo, NY, United States.
² Department of Medicine, Section of Dermatology, University of Verona, Verona, Italy.
³ Department of Dermatology, Venereology and Dermatooncology, Semmelweis University, Budapest, Hungary.
⁴ Brandenburg Medical School Theodor Fontane and Faculty of Health Sciences Brandenburg, Neuruppin, Germany.
⁵ Departments of Medicine and Dermatology, University of Minnesota, Minneapolis, MN, United States.
⁶ Departments of Medicine and Dermatology, Minneapolis VA Health Care System, Minneapolis, MN, United States.
⁷ Dermatology Department, Icahn School of Medicine, Mount Sinai Hospital, New York, NY, United States.
⁸ Kings College Hospital London, Gulf Medical University, Dubai, United Arab Emirates.
⁹ Central Hospital of Northern Pest-Military Hospital, Budapest, Hungary.

Abstract

Introduction: Psoriasis is common, but diagnosis and early severity assessment can be delayed because of variable presentation and overlap with mimicking dermatoses. Multimodal large language models (LLMs) may assist image-based triage.

Objective: To evaluate web-based multimodal LLMs for psoriasis identification, Physician Global Assessment (PGA) scoring, and treatment recommendation quality from clinical photographs.

Methods: We retrospectively analyzed 303 standardized photographs from 160 patients (Semmelweis University, May 2022-January 2025), including 163 psoriasis lesions and 140 mimickers. Reference diagnosis and PGA were assigned by two dermatologists with third-expert adjudication, treatment outputs were rated for appropriateness. ChatGPT-5, ChatGPT-4o, Gemini 2.5 Flash, and Claude Sonnet 4.5 received identical prompts for diagnosis and, when psoriasis was predicted, PGA and treatment; sessions were periodically reset.

Results: Diagnostic accuracy was highest for ChatGPT-5 (93.1%) and ChatGPT-4o (90.1%), followed by Claude (83.6%) and Gemini (61.5%). Among correctly identified psoriasis cases, PGA accuracy was 93.3% (ChatGPT-5), 92.9% (ChatGPT-4o), 82.7% (Gemini), and 78.1% (Claude). Appropriate treatment recommendations were most frequent for ChatGPT-5 (83.7%) and ChatGPT-4o (82.1%), then Claude (75.0%) and Gemini (53.2%); test-retest agreement favored the OpenAI models.

Conclusion: Under standardized conditions, general-purpose multimodal LLMs, especially ChatGPT-5 and ChatGPT-4o, showed strong performance for psoriasis recognition and reasonable support for PGA scoring and treatment suggestions, supporting potential use by primary care physicians when dermatology specialist access is limited.

Keywords: AI; ChatGPT; Claude; Gemini; artificial intelligence; diagnosis; large language model; psoriasis.