Evaluating GPT-4.1 Mini and GPT-5 in Postmastectomy Lymphedema Patient Education: A Comparative Study of Usefulness and Reliability

Benil Nesli Ata; Ece Cinar; Sibel Eyigör

doi:10.19161/etd.1891901

TR EN

Postmastektomi Sonrası Lenfödemli Hastaların Eğitiminde GPT-4.1 Mini ve GPT-5 Yanıtlarının Kullanışlılık ve Güvenilirlik Açısından Analizi

Abstract

Amaç: Bu çalışmanın amacı, postmastektomiye bağlı lenfödemli hastaların eğitiminde GPT-4.1 mini ve GPT-5 büyük dil modellerinin güvenilirlik ve kullanışlılık açısından karşılaştırılmasıdır. Gereç ve Yöntem: Güvenilir bilimsel ve klinik kaynaklardan yararlanılarak hasta odaklı yirmi soru oluşturulmuştur. Her iki model tarafından üretilen yanıtlar, güvenilirlik ve kullanışlılık düzeyleri açısından, 7 puanlı Likert ölçeği kullanılarak üç kör fiziyatrist tarafından bağımsız olarak değerlendirilmiştir. Bulgular: GPT-5, primer ve sekonder lenfödem arasındaki ayrımı daha net ortaya koyması nedeniyle bir soruda anlamlı derecede daha yüksek kullanışlılık puanı elde etmiştir (p = 0,038). Diğer sorulara ait karşılaştırmalarda istatistiksel olarak anlamlı fark saptanmamakla birlikte, GPT-5’in kompleks dekonjestif tedaviyi standart bakım yaklaşımı olarak vurgulaması gibi nedenlerle daha kapsamlı ve klinik uygulamalarla daha uyumlu yanıtlar sunduğu gözlenmiştir. Sonuç: Her iki büyük dil modeli de lenfödem hasta eğitimi için güvenilir ve kullanışlı yanıtlar üretmiştir. GPT-5 sınırlı bir üstünlük göstermiş olmakla birlikte, GPT-4.1 mini’nin temel hasta eğitimi gereksinimlerini karşılayabilecek potansiyele sahip olduğu görülmektedir. Gelecekte yapılacak çalışmaların, yapay zekâ destekli rehabilitasyon uygulamalarının etkinliğini artırmak amacıyla hasta kullanım deneyimi ve gerçek zamanlı klinik kaynakları entegre etmesi önerilmektedir.

Keywords

Evaluating GPT-4.1 Mini and GPT-5 in Postmastectomy Lymphedema Patient Education: A Comparative Study of Usefulness and Reliability

Abstract

Aim: To compare the reliability and usefulness of GPT-4.1 mini and GPT-5 in postmastectomy lymphedema patient education. Materials and Methods: Twenty patient-centered questions were developed from reputable sources. Responses from both models were independently rated by three blinded physiatrists using 7-point Likert scales for reliability and usefulness. Results: GPT-5 demonstrated higher usefulness and reliability scores compared to GPT-4.1 mini (p = 0.005 and p = 0.021, respectively). Conclusions: Both models produced reliable and useful responses for lymphedema education. GPT-5 demonstrated higher usefulness and reliability scores, although these differences appear to reflect incremental improvements rather than substantial differences in clinical applicability. GPT-4.1 mini may already support basic patient education tasks. Future research should integrate patient usability data and real-time clinical resources to optimize AI-assisted rehabilitation.

Keywords

References

1. Mintz Y, Brodie R. Introduction to artificial intelligence in medicine. Minim Invasive Ther Allied Technol 2019;28:73-81.
2. Thirunavukarasu AJ, Ting DSJ, Elangovan K, et al. Large language models in medicine. Nat Med 2023;29:1930-40.
3. Sarraju A, Bruemmer D, Van Iterson E, et al. Appropriateness of cardiovascular disease prevention recommendations obtained from a popular online chat-based artificial intelligence model. JAMA 2023;329:842-4.
4. Xue VW, Lei P, Cho WC. The potential impact of ChatGPT in clinical and translational medicine. Clin Transl Med 2023;13:e1216.
5. van Dis EAM, Bollen J, Zuidema W, et al. ChatGPT: five priorities for research. Nature 2023;614:224-6.
6. OpenAI [homepage on the Internet]. Introducing GPT-4.1. San Francisco (CA): OpenAI; 2025 Apr [cited 2026 Jan 19]. Available from: https://openai.com/index/introducing-gpt-4-1
7. OpenAI [homepage on the Internet]. Introducing GPT-5. San Francisco (CA): OpenAI; 2025 Aug [cited 2026 Jan 19]. Available from: https://openai.com/index/introducing-gpt-5
8. Sallam M. ChatGPT utility in healthcare education, research, and practice: systematic review on the promising perspectives and valid concerns. Healthcare (Basel) 2023;11:887.

9. Kung TH, Cheatham M, Medenilla A, et al. Performance of ChatGPT on USMLE: potential for AI-assisted medical education using large language models. PLoS Digit Health 2023;2:e0000198.
10. Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med 2019;25:44-56.
11. Donahue PMC, MacKenzie A, Filipovic A, et al. Advances in the prevention and treatment of breast cancer-related lymphedema. Breast Cancer Res Treat 2023;200:1-14.
12. Basha MA, Aboelnour NH, Alsharidah AS, et al. Effect of exercise mode on physical function and quality of life in breast cancer-related lymphedema: a randomized trial. Support Care Cancer 2022;30:2101-10.
13. Groeneveld IF, de Boer AG, Frings-Dresen MH. Physical exercise and return to work: cancer survivors' experiences. J Cancer Surviv 2013;7:237-46.
14. Executive Committee of the International Society of Lymphology. The diagnosis and treatment of peripheral lymphedema: 2020 consensus document of the International Society of Lymphology. Lymphology 2020;53:3-19.
15. Perdomo M, Davies C, Levenhagen K, et al. Patient education for breast cancer-related lymphedema: a systematic review. J Cancer Surviv 2023;17:384-98.
16. Borman P, Yaman A, Yasrebi S, et al. The importance of awareness and education in patients with breast cancer-related lymphedema. J Cancer Educ 2017;32:629-33.
17. Peng S, Wang D, Liang Y, Xiao W, Zhang Y, Liu L. AI-ChatGPT/GPT-4: An Booster for the Development of Physical Medicine and Rehabilitation in the New Era! Ann Biomed Eng 2024;52:462-6.
18. Sina EM, Campbell DJ, Duffy A, et al. Evaluating ChatGPT as a patient education tool for COVID-19-induced olfactory dysfunction. OTO Open 2024;8:e70011.
19. Lechien JR, Carroll TL, Huston MN, et al. ChatGPT-4 accuracy for patient education in laryngopharyngeal reflux. Eur Arch Otorhinolaryngol 2024;281:2547-52.
20. American Cancer Society [homepage on the Internet]. Lymphedema after breast cancer [cited 2026 Jan 19]. Available from: https://www.cancer.org/
21. National Cancer Institute [homepage on the Internet]. Lymphedema (PDQ®)–Patient Version [cited 2026 Jan 19]. Available from: https://www.cancer.gov/
22. BreastCancer.org [homepage on the Internet]. Lymphedema after breast cancer [cited 2026 Jan 19]. Available from: https://www.breastcancer.org/
23. Lymphatic Education & Research Network [homepage on the Internet]. [cited 2026 Jan 19]. Available from: https://lymphaticnetwork.org/
24. Uz C, Umay E. "Dr ChatGPT": is it a reliable and useful source for common rheumatic diseases? Int J Rheum Dis 2023;26:1343-9.
25. Reuters. OpenAI says ChatGPT’s weekly users have grown to 200 million [serial on the Internet]. 2024 Aug 29 [cited 2026 Jan 19]. Available from: https://www.reuters.com/technology/artificial-intelligence/openai-says-chatgpts-weekly-users-have-grown-200-million-2024-08-29/
26. Rajpurkar P, Chen E, Banerjee O, et al. AI in health and medicine. Nat Med 2022;28:31-8.
27. World Health Organization. Ethics and governance of artificial intelligence for health. Geneva: World Health Organization; 2021.
28. Küçükakkaş O, İnce B. Can YouTube be used as an educational tool in lymphedema rehabilitation? Arch Physiother 2022;12:5.
29. Kandemir D, Temiz Z, Aydın Sayılan A. YouTube as a source of information on breast cancer-related lymphedema exercises: a content, quality, and reliability analysis. Health Care Women Int 2025;46:1206-20.
30. Millenson ML, Baldwin JL, Zipperer L, Singh H. Beyond Dr. Google: the evidence on consumer-facing digital tools for diagnosis. Diagnosis (Berl) 2018;5:95-105.
31. Temel MH, Erden Y, Bağcıer F. Information quality and readability: ChatGPT's responses to the most common questions about spinal cord injury. World Neurosurg 2024;181:e1138-44.
32. Scheschenja M, Viniol S, Bastian MB, Wessendorf J, König AM, Mahnken AH. Feasibility of GPT-3 and GPT-4 for in-depth patient education prior to interventional radiological procedures: a comparative analysis. Cardiovasc Intervent Radiol 2024;47:245-50.
33. Singhal K, Azizi S, Tu T, et al. Large language models encode clinical knowledge. Nature 2023;620:172-80.
34. Eaton LH, Narkthong N, Hulett JM. Psychosocial issues associated with breast cancer-related lymphedema: a literature review. Curr Breast Cancer Rep 2020;12:216-24.
35. Jørgensen MG, Toyserkani NM, Hansen FG, Bygum A, Sørensen JA. The impact of lymphedema on health-related quality of life up to 10 years after breast cancer treatment. NPJ Breast Cancer 2021;7:70.
36. Dominick SA, Natarajan L, Pierce JP, Madanat H, Madlensky L. The psychosocial impact of lymphedema-related distress among breast cancer survivors in the WHEL Study. Psychooncology 2014;23:1049-56.
37. Ambrosi E, Mezzalira E, Canzan F, Leardini C, Vita G, Marini G, Longhini J. Effectiveness of digital health interventions for chronic conditions management in European primary care settings: systematic review and meta-analysis. Int J Med Inform 2025;196:105820.
38. Johnson E, Corrick S, Isley S, et al. Mind-body internet and mobile-based interventions for depression and anxiety in adults with chronic physical conditions: a systematic review of RCTs. PLoS Digit Health 2024;3:e0000435.
39. Ayers JW, Poliak A, Dredze M, et al. Comparing physician and artificial intelligence chatbot responses to patient questions posted to a public social media forum. JAMA Intern Med 2023;183:589-96.

Details

Primary Language

English

Subjects

Clinical Oncology

Journal Section

Research Article

Authors

Benil Nesli Ata ^*
0000-0003-0900-0069
Türkiye

Ece Cinar
0000-0002-9710-1582
Türkiye

Sibel Eyigör
0000-0002-9781-2712
Türkiye

Publication Date

September 3, 2026

Submission Date

February 18, 2026

Acceptance Date

June 15, 2026

Published in Issue

Year 2026 Volume: 65 Number: 3

DOI

https://doi.org/10.19161/etd.1891901

IZ

https://izlik.org/JA75RZ88BJ

Cite

RIS / Bibtex

APA

Ata, B. N., Cinar, E., & Eyigör, S. (2026). Evaluating GPT-4.1 Mini and GPT-5 in Postmastectomy Lymphedema Patient Education: A Comparative Study of Usefulness and Reliability. Ege Tıp Dergisi, 65(3), 519-526. https://doi.org/10.19161/etd.1891901

AMA

1.Ata BN, Cinar E, Eyigör S. Evaluating GPT-4.1 Mini and GPT-5 in Postmastectomy Lymphedema Patient Education: A Comparative Study of Usefulness and Reliability. EJM. 2026;65(3):519-526. doi:10.19161/etd.1891901

Chicago

Ata, Benil Nesli, Ece Cinar, and Sibel Eyigör. 2026. “Evaluating GPT-4.1 Mini and GPT-5 in Postmastectomy Lymphedema Patient Education: A Comparative Study of Usefulness and Reliability”. Ege Tıp Dergisi 65 (3): 519-26. https://doi.org/10.19161/etd.1891901.

EndNote

Ata BN, Cinar E, Eyigör S (September 1, 2026) Evaluating GPT-4.1 Mini and GPT-5 in Postmastectomy Lymphedema Patient Education: A Comparative Study of Usefulness and Reliability. Ege Tıp Dergisi 65 3 519–526.

IEEE

[1]B. N. Ata, E. Cinar, and S. Eyigör, “Evaluating GPT-4.1 Mini and GPT-5 in Postmastectomy Lymphedema Patient Education: A Comparative Study of Usefulness and Reliability”, EJM, vol. 65, no. 3, pp. 519–526, Sept. 2026, doi: 10.19161/etd.1891901.

ISNAD

Ata, Benil Nesli - Cinar, Ece - Eyigör, Sibel. “Evaluating GPT-4.1 Mini and GPT-5 in Postmastectomy Lymphedema Patient Education: A Comparative Study of Usefulness and Reliability”. Ege Tıp Dergisi 65/3 (September 1, 2026): 519-526. https://doi.org/10.19161/etd.1891901.

JAMA

1.Ata BN, Cinar E, Eyigör S. Evaluating GPT-4.1 Mini and GPT-5 in Postmastectomy Lymphedema Patient Education: A Comparative Study of Usefulness and Reliability. EJM. 2026;65:519–526.

MLA

Ata, Benil Nesli, et al. “Evaluating GPT-4.1 Mini and GPT-5 in Postmastectomy Lymphedema Patient Education: A Comparative Study of Usefulness and Reliability”. Ege Tıp Dergisi, vol. 65, no. 3, Sept. 2026, pp. 519-26, doi:10.19161/etd.1891901.

Vancouver

1.Benil Nesli Ata, Ece Cinar, Sibel Eyigör. Evaluating GPT-4.1 Mini and GPT-5 in Postmastectomy Lymphedema Patient Education: A Comparative Study of Usefulness and Reliability. EJM. 2026 Sep. 1;65(3):519-26. doi:10.19161/etd.1891901