Makroskopik ve Dermoskopik Görüntülere Dayalı Benign Deri Lezyonlarının Tanısında ChatGPT-5’in Bir Tanısal Araç Olarak Kullanımı

Year 2026, Volume: 7 Issue: 1, 171 - 178, 31.01.2026

Esranur Ünal , Deniz Duman Günsay , Muhammed Burak Yücel

https://doi.org/10.47482/acmr.1789948

Abstract

Arka Plan: Tıbbi tanı alanında ChatGPT’nin potansiyeli dikkat çekici olmakla birlikte, konu karmaşık olup farklı tıbbi disiplinlerde çok sayıda çalışmada incelenmiştir. Bu bağlamda, bu çalışmanın amacı, makroskopik ve dermoskopik görüntüler kullanılarak benign deri lezyonlarının tanısında ChatGPT-5’in tanısal doğruluğunu değerlendirmektir.

Yöntemler: Çalışma kapsamında, yüz yüze muayene sırasında dermatolog tarafından 40 hastanın her birine ait makroskopik ve dermoskopik görüntüler kaydedildi. Bu görüntüler, temel klinik bilgilerle birlikte ChatGPT-5’e yüklendi. Değerlendirme süreci iki aşamalı olarak planlandı. İlk aşamada yalnızca makroskopik görüntüler modele sunuldu. İkinci aşamada ise değerlendirme, makroskopik görüntülere ek olarak dermoskopik görüntülerin de dâhil edilmesiyle gerçekleştirildi. Modelden ön tanı koyması ve tanının hatalı olması durumunda üç ayırıcı tanı önermesi istendi. Modelin doğruluk düzeyi, elde edilen tanıların histopatolojik sonuçlarla karşılaştırılması yoluyla belirlendi.

Bulgular: ChatGPT-5 ile yapılan değerlendirmede, yalnızca makroskopik görüntülere dayalı tanısal doğruluk %32,5 iken, makroskopik ve dermoskopik görüntülerin birlikte kullanılmasıyla doğruluk %27,5’e düştü (p = 0,450). Üç ayırıcı tanının dikkate alındığı durumda, makroskopik görüntülerle doğru tanı oranı %48,1 iken, dermoskopik görüntülerin eklenmesiyle bu oran %29,6’ya düştü (p < 0,001).

Sonuç: ChatGPT-5 benign deri lezyonları için sınırlı düzeyde tanısal doğruluk göstermiş, dermoskopik görüntüler eklendiğinde performansı düşmüştür. Bu sonuçlar, ChatGPT-5’in bağımsız bir tanı aracı olarak değil, destekleyici bir yardımcı olarak değerlendirilmesi gerektiğini göstermektedir.

Keywords

Yapay zeka , ChatGPT-5 , benign deri lezyonları , tanısal doğruluk

Ethical Statement

Bu çalışma, anonimleştirilmiş hasta verileri kullanılarak retrospektif olarak gerçekleştirildi ve bu nedenle etik kuruldan onay alınması gerekmedi. Lezyon görüntülerinin kullanımı için hastadan yazılı bilgilendirilmiş onam alındı.

Supporting Institution

yok

Thanks

yok

Evaluating ChatGPT-5 for the Diagnosis of Benign Skin Lesions: Insights from Macroscopic and Dermoscopic Imaging

Abstract

Background: While the potential of ChatGPT in the domain of medical diagnosis is noteworthy, the subject is intricate and has been examined in numerous studies across various medical disciplines. In this context, the objective of this study is to utilize ChatGPT-5 to evaluate its diagnostic accuracy for benign skin lesions using macroscopic and dermoscopic images.
Methods: During the in-person examination, the dermatologist documented macroscopic and dermoscopic images of each of the 40 patients. These images, along with basic clinical information, were uploaded to ChatGPT-5. The evaluation process was meticulously structured into two distinct phases. In the initial phase, the presentation was limited to macroscopic images alone. In the subsequent phase, the presentation expanded to encompass both macroscopic and dermoscopic images. The model was tasked with making a preliminary diagnosis and, in the event of an inaccuracy, was expected to provide three differential diagnoses. The model's accuracy was assessed by comparing its diagnoses with the histopathological results.
Results: In the evaluation conducted with ChatGPT-5, the diagnostic accuracy based solely on macroscopic images was 32.5%, whereas the accuracy for combined macroscopic and dermoscopic images decreased to 27.5% (p = 0.450). When three differential diagnoses were considered, the correct diagnosis was achieved in 48.1% of cases using macroscopic images, whereas this rate declined to 29.6% with the inclusion of dermoscopic images (p < 0.001).
Conclusion: ChatGPT-5 demonstrated modest diagnostic accuracy for benign skin lesions, with performance declined when dermoscopic images were included. These results suggest that ChatGPT-5 should be considered a supportive aid rather than a standalone diagnostic tool.

Keywords

Artificial intelligence , ChatGPT-5 , benıgn skin lesion , diagnostic accuracy

Ethical Statement

All ethical approvals for the study were obtained from the Ufuk University Faculty of Medicine Ethics Committee with the file dated 03.11.2025 and decision number 25.11.03.08/10. All participants were informed about the study protocol and provided with written consent forms.

Supporting Institution

None.

Thanks

There are no individuals to acknowledge.

References

• Collin H, Keogh K, Basto M, Loeb S, Roberts MJ. ChatGPT can help guide and empower patients after prostate cancer diagnosis. Prostate Cancer Prostatic Dis. 2025;28:513–5.
• De A, Sarda A, Gupta S, Das S. Use of Artificial Intelligence in Dermatology. Indian J Dermatol. 2020;65:352–7.
• Khamaysi Z, Awwad M, Jiryis B, Bathish N, Shapiro J. The Role of ChatGPT in Dermatology Diagnostics. Diagnostics (Basel). 2025;15:1529.
• Zheng Y, Sun X, Feng B, Kang K, Yang Y, Zhao A, et al. Rare and complex diseases in focus: ChatGPT's role in improving diagnosis and treatment. Front Artif Intell. 2024;7:1338433.
• Sonthalia S, Yumeen S, Kaliyadan F. Dermoscopy overview and extradiagnostic applications. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2025–.
• Wang Z, Wang C, Peng L, Lin K, Xue Y, Chen X, et al. Radiomic and deep learning analysis of dermoscopic images for skin lesion pattern decoding. Sci Rep. 2024;14:19781.
• Pankhurst A, Kumar N. Identifying the validity of ChatGPT in the diagnosis of orthopaedic emergencies. Br J Surg. 2024;111:znae163.699.
• Belgam Syed SY, Lipoff JB, Chatterjee K. Acrochordon. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2025–.
• Gabashvili IS. ChatGPT in dermatology: a comprehensive systematic review. medRxiv. 2023;2023-06.
• Nazir A, Wang Z. A comprehensive survey of ChatGPT: advancements, applications, prospects, and challenges. Meta Radiol. 2023;1:100022.
• Ray PP. ChatGPT: a comprehensive review on background, applications, key challenges, bias, ethics, limitations and future scope. Internet Things Cyber-Phys Syst. 2023;3:121–54.
• Ali R, Cui H. Evaluating ChatGPT's diagnostic accuracy in skin diseases based on images. Am J Transl Res. 2025;17:5553–61.
• Liu X, Duan C, Kim MK, Zhang L, Jee E, Maharjan B, et al. Claude 3 Opus and ChatGPT with GPT-4 in dermoscopic image analysis for melanoma diagnosis: comparative performance analysis. JMIR Med Inform. 2024;12:e59273.
• Lakdawala N, Channa L, Gronbeck C, Lakdawala N, Weston G, Sloan B, et al. Assessing the accuracy and comprehensiveness of ChatGPT in offering clinical guidance for atopic dermatitis and acne vulgaris. JMIR Dermatol. 2023;6:e50409.
• Alanezi F. Factors influencing patients’ engagement with ChatGPT for accessing health-related information. Crit Public Health. 2024;34:1–20.
• Sattler SS, Chetla N, Chen M, Hage TR, Chang J, Guo WY, et al. Evaluating the diagnostic accuracy of ChatGPT-4 Omni and ChatGPT-4 Turbo in identifying melanoma: comparative study. JMIR Dermatol. 2025;8:e67551.
• Chetla N, Chen M, Chang J, Smith A, Hage TR, Patel R, et al. Assessing the diagnostic accuracy of ChatGPT-4 in identifying diverse skin lesions against squamous and basal cell carcinoma. JMIR Dermatol. 2025;8:e67299.
• Pinto-Coelho L. How artificial intelligence is shaping medical imaging technology: a survey of innovations and applications. Bioengineering (Basel). 2023;10:1435.
• Hameed EK, Al-Ameri LT. Artificial intelligence: the gateway to better medical diagnosis. Al-Kindy Coll Med J. 2024;20:1–3.
• Adelaja O, Alkattan H. Operating artificial intelligence to assist physicians diagnose medical images: a narrative review. Mesopotam J Artif Intell Healthc. 2023;2023:45–51.
• Gonçalves FL, Souza HV da P, Almeida FM de, Pinto MS. Use of artificial intelligence (AI) tools in image diagnosis. Res Soc Dev. 2024;13:e64131147312.
• Sahay R, Singh A, Aggarwal M. Role of artificial intelligence in diagnostic medicine. Int J Res Rev Appl Sci Humanit Technol. 2024;2024:49–53.
• Rundle CW, Szeto MD, Presley CL, Shahwan KT, Carr DR. Analysis of ChatGPT generated differential diagnoses in response to physical exam findings for benign and malignant cutaneous neoplasms. J Am Acad Dermatol. 2024;90:615–6.
• Scheinkman R, Tordjman L, Dender L, Green D, Korsch M, Nouri K. ChatGPT-4o proves effective in accurately diagnosing dermoscopic images of benign and malignant skin conditions. Skin J Cutan Med. 2025;9:2634–7.