Prompt Engineering in AI-Generated Clinical Ophthalmic Images of Hordeolum: From Minimal to Comprehensive Prompts
Date Published : 25 January 2026
Contributors
Nuraini
Ophthalmology/Faculty of Medicine, Universitas Muhammadiyah Prof. Dr. Hamka, Tangerang, 13460, Indonesia
Author
Kuncoro Hadi
Management/Faculty of Economic and Business, Universitas Al-Azhar Indonesia, Jakarta, 12110, Indonesia
Author
Muhammad Rafi Ar-Rantisi
Biomedical Engineering/School of Electrical Engineering and Informatics, Institut Teknologi Bandung, Bandung, 40132, Indonesia
Author
Muhammad Syamil Haniyyah
Information System/Faculty of Science and Technology, Universitas Islam Negeri Syarif Hidayatullah Jakarta, South Tangerang, 15412, Indonesia
Author
Keywords
artificial intelligence
clinical image
hordeolum
ophthalmology
prompt engineering
Proceeding
Track
General Track
Abstract
High-quality clinical images are crucial in ophthalmology education for recognizing clinical diagnostic sign and decision-making. However, access to real patient photographs is restricted by consent and data protection. Generative artificial intelligence (AI) offers a potential alternative, yet the impact of prompt complexity on AI-generated clinical images remains variable. Hordeolum, an eyelid infection with well-defined clinical features, was chosen as the model condition. Objectives. To evaluate the influence of prompt complexity on the quality and anatomical accuracy of AI-generated clinical images of hordeolum using ChatGPT-5o. Methods. This descriptive study used ChatGPT-5o to generate three standardized text prompts of varying complexity, minimal (Level 1, L1), intermediate (Level 2, L2) and comprehensive (Level 3, L3). Each prompt was used to create three images (n = 9) through an integrated text-to-image model. Prompts were analyzed for five key components (anatomy, pathology, morphology, terminology and technical elements) and the resulting images were visually compared across levels. Prompt complexity directly influenced image richness. L1 generated simple nodules, L2 incorporated pathological details and local tissue reactions, while L3 added precise anatomical landmarks, surface texture and slit-lamp-like lighting. Some images showed inconsistencies, such as misplaced hyperemia or unessential detail. Conclusion: ChatGPT-5o can generate structured and tiered prompts to produce clinical images of hordeolum, offering a reproducible and ethically safer alternative for educational use.
References
1. Gurwin J, Revere KE, Niepold S, Bassett B, Mitchell R, Davidson S, et al. A Randomized Controlled Study of Art Observation Training to Improve Medical Student Ophthalmology Skills. Ophthalmology [Internet]. 2018 Jan 1;125(1):8–14. Available from: https://doi.org/10.1016/j.ophtha.2017.06.031
2. Ferjaoui W. Image in Medical Education: Why, When and How to Use It? A Critical Review of the Literature. Biomed J Sci Tech Res [Internet]. 2020 Mar 6 [cited 2025 Sep 19];26(3). Available from: https://biomedres.us/fulltexts/BJSTR.MS.ID.004343.php
3. Teven CM, Goldman JJ, Rebecca AM. Ethical Considerations in Clinical Photography: An Updated Perspective. Plast Reconstr Surg [Internet]. 2021 Feb;147(2):366e–8e. Available from: https://journals.lww.com/plasreconsurg/fulltext/2021/02000/ethical_considerations_in_clinical_photography__an.76.aspx
4. Padmapriya ST, Parthasarathy S. Ethical Data Collection for Medical Image Analysis: a Structured Approach. Asian Bioeth Rev [Internet]. 2024 Jan 10 [cited 2025 Sep 19];16(1):95–108. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC10088772/
5. Wongvibulsin S, Feterik K. Recommendations for Better Adoption of Medical Photography as a Clinical Tool. Interact J Med Res [Internet]. 2022 Jul 18 [cited 2025 Nov 6];11(2):e36102. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC9345030/
6. Dehouche N, Dehouche K. What’s in a text-to-image prompt? The potential of stable diffusion in visual arts education. Heliyon [Internet]. 2023 Jun [cited 2025 Sep 19];9(6):e16757. Available from: https://www.sciencedirect.com/science/article/pii/S2405844023039646
7. Safadi MF, Zayegh O, Hawoot Z. Advancing Innovation in Medical Presentations: A Guide for Medical Educators to Use Images Generated With Artificial Intelligence. Cureus [Internet]. 2024 Dec 2 [cited 2025 Sep 13];16(12):e74978. Available from: https://www.cureus.com/articles/323853-advancing-innovation-in-medical-presentations-a-guide-for-medical-educators-to-use-images-generated-with-artificial-intelligence#!/
8. Temsah M, Alhuzaimi A, Almansour M, Aljamaan F, Alhasan K, Batarfi MA, et al. Art or Artifact: Evaluating the Accuracy, Appeal, and Educational Value of AI-Generated Imagery in DALL·E 3 for Illustrating Congenital Heart Diseases. J Med Syst [Internet]. 2024 [cited 2025 Sep 29];48(1). Available from: https://www.researchgate.net/publication/377725918_Art_or_Artifact_Evaluating_the_Accuracy_Appeal_and_Educational_Value_of_AI-Generated_Imagery_in_DALLE_3_for_Illustrating_Congenital_Heart_Diseases
9. Haider SA, Prabha S, Gomez-Cabello CA, Borna S, Pressman SM, Genovese A, et al. A Validity Analysis of Text-to-Image Generative Artificial Intelligence Models for Craniofacial Anatomy Illustration. J Clin Med [Internet]. 2025 Mar 21 [cited 2025 Sep 13];14(7):2136. Available from: https://www.mdpi.com/2077-0383/14/7/2136
10. Taloni A, Taloni M, Coco G, Muttuvelu D V., Busin M, Giannaccare G. AI deepfake: GPT-4o can produce near-authentic fundus images. Eye [Internet]. 2025 Sep 19 [cited 2025 Sep 28];39(13):2612–4. Available from: https://doi.org/10.1038/s41433-025-03937-5
11. Oshika T. Artificial Intelligence Applications in Ophthalmology. JMA J [Internet]. 2025 [cited 2025 Sep 28];8(1):66–75. Available from: https://www.jmaj.jp/detail.php?id=10.31662%2Fjmaj.2024-0139
12. Olawade DB, Weerasinghe K, Mathugamage MDDE, Odetayo A, Aderinto N, Teke J, et al. Enhancing Ophthalmic Diagnosis and Treatment with Artificial Intelligence. Medicina (B Aires) [Internet]. 2025 Feb 28 [cited 2025 Sep 19];61(3):433. Available from: https://doi.org/10.3390/medicina61030433
13. Heston T, Khun C. Prompt Engineering in Medical Education. International Medical Education [Internet]. 2023 Aug 31 [cited 2025 Sep 15];2(3):198–205. Available from: https://www.mdpi.com/2813-141X/2/3/19
14. Lee D, Palmer E. Prompt engineering in higher education: a systematic review to help inform curricula. International Journal of Educational Technology in Higher Education [Internet]. 2025;22(1):7. Available from: https://doi.org/10.1186/s41239-025-00503-7
15. Sweeney AR, Baharestani S, Hossain K, Yen MT, Marcet MM, Silkiss RZ, et al. Stye [Internet]. https://eyewiki.org/Stye. 2025 [cited 2025 Sep 28]. Available from: https://eyewiki.org/Stye
16. Willman D, Guier C, Patel BC. Hordeolum (Stye). In: https://www.ncbi.nlm.nih.gov/books/NBK459349/ [Internet]. Treasure Island (Fl): StatPearls Publishing; 2024 [cited 2025 Sep 28]. Available from: https://www.ncbi.nlm.nih.gov/books/NBK459349/
17. KONSIL KEDOKTERAN INDONESIA. STANDAR KOMPETENSI DOKTER INDONESIA [Internet]. Edisi Kedua. Jakarta: Konsil Kedokteran Indonesia; 2012 [cited 2025 Sep 28]. 37 p. Available from: https://lamptkes.org/secure/document/unduhan/Final_SPPDI,_21_Maret_2013_1555037477.pdf
18. Savastano MC, Rizzo C, Fossataro C, Bacherini D, Giansanti F, Savastano A, et al. Artificial intelligence in ophthalmology: Progress, challenges, and ethical implications. Prog Retin Eye Res [Internet]. 2025 Jul [cited 2025 Sep 19];107:101374. Available from: https://www.sciencedirect.com/science/article/pii/S1350946225000473
19. M’gadzah SAT, O’Malley A. Enhancing Diagnostic Accuracy of Ophthalmological Conditions With Complex Prompts in GPT-4: A Comparative Analysis of Global and LMIC-Specific Pathologies (Preprint). JMIR Form Res [Internet]. 2024 Jun 30 [cited 2025 Sep 29];9:e64986. Available from: https://formative.jmir.org/2025/1/e64986
20. Zhou J, Zhang J, Wan R, Cui X, Liu Q, Guo H, et al. Integrating AI into clinical education: evaluating general practice trainees’ proficiency in distinguishing AI-generated hallucinations and impacting factors. BMC Med Educ [Internet]. 2025 Mar 19 [cited 2025 Sep 13];25(1):406. Available from: https://bmcmededuc.biomedcentral.com/articles/10.1186/s12909-025-06916-2
21. Vaira LA, Lechien JR, Abbate V, Gabriele G, Frosolini A, Vito AD, et al. Enhancing AI Chatbot Responses in Health Care: The SMART Prompt Structure in Head and Neck Surgery. OTO Open [Internet]. 2025 [cited 2025 Sep 29];9(1). Available from: https://aao-hnsfjournals.onlinelibrary.wiley.com/doi/full/10.1002/oto2.70075
22. Ekin S. Prompt Engineering For ChatGPT: A Quick Guide To Techniques, Tips, And Best Practices. 2023 May 4 [cited 2025 Sep 29];1. Available from: https://www.techrxiv.org/users/690417/articles/681648-prompt-engineering-for-chatgpt-a-quick-guide-to-techniques-tips-and-best-practices
23. Luo MJ, Bi S, Pang J, Liu L, Tsui CK, Lai Y, et al. A large language model digital patient system enhances ophthalmology history taking skills. NPJ Digit Med [Internet]. 2025 Aug 4 [cited 2025 Sep 13];8(1):502. Available from: https://www.nature.com/articles/s41746-025-01841-6
24. Davis P, Napole A, Reddy C, Palepu SK, Wu AQ, Mannam SS, et al. Systematic evaluation of AI-based text-to-image models for generating medical illustrations in neurosurgery: a multi-stage comparative study. Clin Neurol Neurosurg [Internet]. 2025 Oct [cited 2025 Sep 13];257:109039. Available from: https://www.sciencedirect.com/science/article/pii/S0303846725003221
25. Meskó B. Prompt Engineering as an Important Emerging Skill for Medical Professionals: Tutorial. J Med Internet Res [Internet]. 2023 Oct 4 [cited 2025 Sep 15];25:e50638. Available from: https://www.jmir.org/2023/1/e50638
26. Noel GPJC. Evaluating AI‐powered text‐to‐image generators for anatomical illustration: A comparative study. Anat Sci Educ [Internet]. 2024 Jul 11 [cited 2025 Sep 16];17(5):979–83. Available from: https://anatomypubs.onlinelibrary.wiley.com/doi/10.1002/ase.2336
27. Moin KA, Nasir AA, Petroff DJ, Loveless BA, Moshirfar OA, Hoopes PC, et al. Assessment of Generative Artificial Intelligence (AI) Models in Creating Medical Illustrations for Various Corneal Transplant Procedures. Cureus [Internet]. 2024 Aug 26 [cited 2025 Sep 13];16(8):e67833. Available from: https://www.cureus.com/articles/285243-assessment-of-generative-artificial-intelligence-ai-models-in-creating-medical-illustrations-for-various-corneal-transplant-procedures#!/
28. Petroff DJ, Nasir AA, Moin KA, Loveless BA, Moshirfar OA, Hoopes PC, et al. Evaluating the Accuracy of Artificial Intelligence (AI)-Generated Illustrations for Laser-Assisted In Situ Keratomileusis (LASIK), Photorefractive Keratectomy (PRK), and Small Incision Lenticule Extraction (SMILE). Cureus [Internet]. 2024 Aug 25 [cited 2025 Sep 13];e67747. Available from: https://www.cureus.com/articles/285461-evaluating-the-accuracy-of-artificial-intelligence-ai-generated-illustrations-for-laser-assisted-in-situ-keratomileusis-lasik-photorefractive-keratectomy-prk-and-small-incision-lenticule-extraction-smile
29. Lee KH, Lee RW, Kwon YE. Validation of a Deep Learning Chest X-Ray Interpretation Model: Integrating Large-Scale AI and Large Language Models for Comparative Analysis With ChatGPT. Diagnostics [Internet]. 2023 [cited 2025 Sep 29];14(1):90. Available from: https://www.mdpi.com/2075-4418/14/1/90
Downloads
How to Cite
Nuraini, N., Hadi, K. ., Ar-Rantisi, M. R. ., & Haniyyah, M. S. . (2026). Prompt Engineering in AI-Generated Clinical Ophthalmic Images of Hordeolum: From Minimal to Comprehensive Prompts. InSight 2025 - International Conference on Healthcare Safety and Transformative Learning in Health Education, 1(1). https://conferences.uhamka.ac.id/InSight/paper/view/21