Journals / Endoüroloji Bülteni / 2026 / Cilt: 18 Sayı: 2

Occupational Health and Safety in Urological Surgeries: A Comparison of the Performance of Artificial Intelligence-Based Large Language Models (ChatGPT, Gemini, Perplexity, DeepSeek)

Ürolojik Ameliyatlarda İş Sağlığı ve Güvenliği: Yapay Zeka Tabanlı Büyük Dil Modellerinin (ChatGPT, Gemini, Perplexity, DeepSeek) Performanslarının Karşılaştırılması

Şeyma Demir(VAN YUZUNCU YIL UNIVERSITY)0009-0002-1097-7627
Murat Demir(VAN YUZUNCU YIL UNIVERSITY)0000-0001-5029-8800
Pages
73–79
Publication Date
2026-05-29
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

Therefore, AI tools should not be used as standalone decision-makers in clinical practice but should be supported by expert oversight and guideline-based validation.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Objective: This study aims to comparatively evaluate the knowledge level, guideline compliance, scope, and consistency of responses provided by artificial intelligence-based large language models (ChatGPT, Gemini, Perplexity, DeepSeek) to questions related to occupational health and safety (OHS) in urological surgeries.
📊
Key Findings & Data:Materials and Methods: Based on the literature and international guidelines (EAU, WHO, OSHA, etc.), 10 standardized questions covering biological, chemical, physical, ergonomic, environmental, and psychosocial risks were developed.
💡
Conclusion & Impact:Therefore, AI tools should not be used as standalone decision-makers in clinical practice but should be supported by expert oversight and guideline-based validation.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:büyük dil modeli · ChatGPT · DeepSeek

Abstract

Objective: This study aims to comparatively evaluate the knowledge level, guideline compliance, scope, and consistency of responses provided by artificial intelligence-based large language models (ChatGPT, Gemini, Perplexity, DeepSeek) to questions related to occupational health and safety (OHS) in urological surgeries. Materials and Methods: Based on the literature and international guidelines (EAU, WHO, OSHA, etc.), 10 standardized questions covering biological, chemical, physical, ergonomic, environmental, and psychosocial risks were developed. These questions were presented to four different models on the same day and in the same format. The responses were blindly evaluated by three independent experts using DISCERN and Global Quality Score (GQS) criteria. Statistical analyses were performed using ANOVA/Kruskal–Wallis tests and Fleiss’ kappa. Results: ChatGPT and Gemini achieved the highest performance in both DISCERN and GQS scores. Perplexity demonstrated moderate performance, while DeepSeek showed the lowest quality and consistency. ChatGPT scored significantly higher than Perplexity and DeepSeek (p<0.01), and Gemini was also significantly superior to Perplexity and DeepSeek (p<0.05). No significant difference was found between ChatGPT and Gemini. Conclusion: AI-based large language models can provide supportive information on OHS in urological surgeries; however, there are notable quality differences among models. While ChatGPT and Gemini deliver more reliable and comprehensive content, the limitations of Perplexity and especially DeepSeek are evident. Therefore, AI tools should not be used as standalone decision-makers in clinical practice but should be supported by expert oversight and guideline-based validation.

Özet

Amaç: Bu çalışma, yapay zeka (YZ) tabanlı büyük dil modellerinin (ChatGPT, Gemini, Perplexity, DeepSeek) ürolojik ameliyatlarda iş sağlığı ve güvenliği (İSG) ile ilgili sorulara verdikleri yanıtların bilgi düzeyi, rehberlerle uyumu, kapsamı ve tutarlılığını karşılaştırmalı olarak değerlendirmeyi amaçlamaktadır. Gereç ve Yöntemler: Literatür ve uluslararası kılavuzlar (EAU, WHO, OSHA vb.) temel alınarak biyolojik, kimyasal, fiziksel, ergonomik, çevresel ve psikososyal riskleri kapsayan 10 standart soru hazırlanmıştır. Bu sorular dört farklı modele aynı gün ve aynı formatta yöneltilmiş, yanıtlar üç bağımsız uzman tarafından DISCERN ve GQS kriterleriyle kör olarak değerlendirilmiştir. Bulgular: ChatGPT ve Gemini, hem DISCERN hem de GQS skorlarında en yüksek performansı sergilemiştir. Perplexity orta düzeyde, DeepSeek ise en düşük kalite ve tutarlılığı göstermiştir. ChatGPT, Perplexity ve DeepSeek’ten anlamlı derecede daha yüksek skor alırken (p<0,01), Gemini de Perplexity ve DeepSeek’ten anlamlı şekilde üstündür (p<0,05). ChatGPT ile Gemini arasında anlamlı fark bulunmamıştır. Sonuç: YZ tabanlı büyük dil modelleri, ürolojik ameliyatlarda İSG konusunda bilgi desteği sağlayabilmektedir; ancak modeller arasında önemli kalite farklılıkları mevcuttur. ChatGPT ve Gemini daha güvenilir ve kapsamlı içerikler sunarken, Perplexity ve özellikle DeepSeek’in sınırlılıkları belirgindir. Bu nedenle YZ araçları, klinik karar desteğinde tek başına kullanılmamalı, uzman denetimi ve kılavuz temelli doğrulamalarla desteklenmelidir.

🔖

Cite This Article

Demir, Ş., Demir, M. (2026). Occupational Health and Safety in Urological Surgeries: A Comparison of the Performance of Artificial Intelligence-Based Large Language Models (ChatGPT, Gemini, Perplexity, DeepSeek). Endoüroloji Bülteni, Cilt: 18 Sayı: 2, 73–79. https://doi.org/10.54233/endourolbull-1792399

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (11)

  1. Landford WN, Ngaage LM, Lee E, et al. Occupational exposures in the operating room: Are surgeons well-equipped? PLoS One. 2021;16(7):e0253785. https://doi.org/10.1371/journal.pone.0253785
  2. Dixon K, Dasgupta P, Vasdev N. A systematic review of the harmful effects of surgical smoke inhalation on operating room personnel. Health Sci Rev. 2023;6:100077. https://doi.org/10.1016/j.hsr.2022.100077
  3. Memon AG, Naeem Z, Zaman A, Zahid F. Occupational health related concerns among surgeons. Int J Health Sci (Qassim). 2016;10(2):279-91.
  4. Meng X, Yan X, Zhang K, et al. The application of large language models in medicine: A scoping review. iScience. 2024;27(5):109713. https://doi.org/10.1016/j.isci.2024.109713
  5. Hager P, Jungmann F, Holland R, et al. Evaluation and mitigation of the limitations of large language models in clinical decision-making. Nat Med. 2024;30(9):2613-22. https://doi.org/10.1038/s41591-024-03215-7
  6. Rossettini G, Bargeri S, Cook C, et al. Accuracy of ChatGPT-3.5, ChatGPT-4o, Copilot, Gemini, Claude, and Perplexity in advising on lumbosacral radicular pain against clinical practice guidelines: Cross-sectional study. Front Digit Health. 2025;7:1574287. https://doi.org/10.3389/fdgth.2025.1574287
  7. Charnock D, Shepperd S, Needham G, Gann R. DISCERN: an instrument for judging the quality of written consumer health information on treatment choices. J Epidemiol Community Health. 1999;53(2):105-11. https://doi.org/10.1136/jech.53.2.105
  8. Bromage S, Batis C, Bhupathiraju SN, et al. Development and validation of a novel food-based global diet quality score (GDQS). J Nutr. 2021;151(Suppl 1):75S-92S. https://doi.org/10.1093/jn/nxab244
  9. van den Berg-Dijkmeijer ML, Frings-Dresen MH, Sluiter JK. Risks and health effects in operating room personnel. Work. 2011;39(3):331-44. https://doi.org/10.3233/WOR-2011-1181
  10. Johnson D, Goodman R, Patrinely J, et al. Assessing the accuracy and reliability of AI-generated medical responses: An evaluation of the ChatGPT model. Res Sq. 2023;rs.3.rs-2566942. https://doi.org/10.21203/rs.3.rs-2566942/v1
  11. Wang C, Liu S, Yang H, et al. Ethical considerations of using ChatGPT in health care. J Med Internet Res. 2023;25:e48009. https://doi.org/10.2196/48009