Top

Graefe's Archive for Clinical and Experimental Ophthalmology

Open Access 04-04-2024 | Glaucoma | Glaucoma

Large language models as assistance for glaucoma surgical cases: a ChatGPT vs. Google Gemini comparison

Authors: Matteo Mario Carlà, Gloria Gambini, Antonio Baldascino, Francesco Boselli, Federico Giannuzzi, Fabio Margollicci, Stanislao Rizzo

Published in: Graefe's Archive for Clinical and Experimental Ophthalmology

Abstract

Purpose

The aim of this study was to define the capability of ChatGPT-4 and Google Gemini in analyzing detailed glaucoma case descriptions and suggesting an accurate surgical plan.

Methods

Retrospective analysis of 60 medical records of surgical glaucoma was divided into “ordinary” (n = 40) and “challenging” (n = 20) scenarios. Case descriptions were entered into ChatGPT and Bard’s interfaces with the question “What kind of surgery would you perform?” and repeated three times to analyze the answers’ consistency. After collecting the answers, we assessed the level of agreement with the unified opinion of three glaucoma surgeons. Moreover, we graded the quality of the responses with scores from 1 (poor quality) to 5 (excellent quality), according to the Global Quality Score (GQS) and compared the results.

Results

ChatGPT surgical choice was consistent with those of glaucoma specialists in 35/60 cases (58%), compared to 19/60 (32%) of Gemini (p = 0.0001). Gemini was not able to complete the task in 16 cases (27%). Trabeculectomy was the most frequent choice for both chatbots (53% and 50% for ChatGPT and Gemini, respectively). In “challenging” cases, ChatGPT agreed with specialists in 9/20 choices (45%), outperforming Google Gemini performances (4/20, 20%). Overall, GQS scores were 3.5 ± 1.2 and 2.1 ± 1.5 for ChatGPT and Gemini (p = 0.002). This difference was even more marked if focusing only on “challenging” cases (1.5 ± 1.4 vs. 3.0 ± 1.5, p = 0.001).

Conclusion

ChatGPT-4 showed a good analysis performance for glaucoma surgical cases, either ordinary or challenging. On the other side, Google Gemini showed strong limitations in this setting, presenting high rates of unprecise or missed answers.

Ozdemir S (2023) Quick start guide to large language models: strategies and best practices for using ChatGPT and other LLMs. Addison-Wesley Professional

Thirunavukarasu AJ, Ting DSJ, Elangovan K, Gutierrez L, Tan TF, Ting DSW (2023) Large language models in medicine. Nat Med 29:1930–1940CrossRefPubMed

Singhal K, Azizi S, Tu T, Mahdavi SS, Wei J, Chung HW, Scales N, Tanwani A, Cole-Lewis H, Pfohl S, Payne P, Seneviratne M, Gamble P, Kelly C, Babiker A, Scharli N, Chowdhery A, Mansfield P, Demner-Fushman D, Aguera YAB, Webster D, Corrado GS, Matias Y, Chou K, Gottweis J, Tomasev N, Liu Y, Rajkomar A, Barral J, Semturs C, Karthikesalingam A, Natarajan V (2023) Large language models encode clinical knowledge. Nature 620:172–180. https://doi.org/10.1038/s41586-023-06291-2CrossRefPubMedPubMedCentral

Nath S, Marie A, Ellershaw S, Korot E, Keane PA (2022) New meaning for NLP: the trials and tribulations of natural language processing with GPT-3 in ophthalmology. Br J Ophthalmol 106:889–892. https://doi.org/10.1136/bjophthalmol-2022-321141CrossRefPubMed

Kung TH, Cheatham M, Medenilla A, Sillos C, De Leon L, Elepano C, Madriaga M, Aggabao R, Diaz-Candido G, Maningo J, Tseng V (2023) Performance of ChatGPT on USMLE: potential for AI-assisted medical education using large language models. PLOS Digit Health 2:e0000198. https://doi.org/10.1371/journal.pdig.0000198CrossRefPubMedPubMedCentral

Antaki F, Touma S, Milad D, El-Khoury J, Duval R (2023) Evaluating the performance of ChatGPT in ophthalmology: an analysis of its successes and shortcomings. Ophthalmol Sci 3:100324. https://doi.org/10.1016/j.xops.2023.100324CrossRefPubMedPubMedCentral

Siad S (2023) The promise and perils of Google’s Bard for scientific research. AI 1:1–5

Koga S, Martin NB, Dickson DW (2023) Evaluating the performance of large language models: ChatGPT and Google Bard in generating differential diagnoses in clinicopathological conferences of neurodegenerative disorders. Brain Pathol 8:e13207

Gan RK, Ogbodo JC, Wee YZ, Gan AZ, González PA (2024) Performance of Google Bard and ChatGPT in mass casualty incidents triage. Am J Emerg Med 75:72–78CrossRefPubMed

10.

Ali R, Tang OY, Connolly ID, Fridley JS, Shin JH, Sullivan PLZ, Cielo D, Oyelese AA, Doberstein CE, Telfeian AE (2022) Performance of ChatGPT, GPT-4, and Google Bard on a neurosurgery oral boards preparation question bank. Neurosurgery: https://doi.org/10.1227/neu.0000000000002551

11.

Team G (2024) Bard becomes Gemini: try Ultra 1.0 and a new mobile app today. Google, Inc. https://blog.google/products/gemini/bard-gemini-advanced-app/

12.

Yousefi S, Pasquale LR, Boland MV, Johnson CA (2022) Machine-identified patterns of visual field loss and an association with rapid progression in the ocular hypertension treatment study. Ophthalmology 129:1402–1411. https://doi.org/10.1016/j.ophtha.2022.07.001CrossRefPubMed

13.

Medeiros FA, Jammal AA, Thompson AC (2019) From machine to machine: an OCT-trained deep learning algorithm for objective quantification of glaucomatous damage in fundus photographs. Ophthalmology 126:513–521. https://doi.org/10.1016/j.ophtha.2018.12.033CrossRefPubMed

14.

Yousefi S (2023) Clinical applications of artificial intelligence in glaucoma. J Ophthalmic Vis Res 18:97–112. https://doi.org/10.18502/jovr.v18i1.12730CrossRefPubMedPubMedCentral

15.

European Glaucoma Prevention Study G, Miglior S, Pfeiffer N, Torri V, Zeyen T, Cunha-Vaz J, Adamsons I (2007) Predictive factors for open-angle glaucoma among patients with ocular hypertension in the European Glaucoma Prevention Study. Ophthalmology 114:3–9. https://doi.org/10.1016/j.ophtha.2006.05.075CrossRef

16.

Le A, Mukesh BN, McCarty CA, Taylor HR (2003) Risk factors associated with the incidence of open-angle glaucoma: the visual impairment project. Invest Ophthalmol Vis Sci 44:3783–3789. https://doi.org/10.1167/iovs.03-0077CrossRefPubMed

17.

Jonas JB, Aung T, Bourne RR, Bron AM, Ritch R, Panda-Jonas S (2017) Glaucoma Lancet 390:2183–2193. https://doi.org/10.1016/S0140-6736(17)31469-1CrossRefPubMed

18.

Bovee CE, Pasquale LR (2017) Evolving surgical interventions in the treatment of glaucoma. Semin Ophthalmol 32:91–95. https://doi.org/10.1080/08820538.2016.1228393CrossRefPubMed

19.

Sunaric Megevand G, Bron AM (2021) Personalising surgical treatments for glaucoma patients. Prog Retin Eye Res 81:100879. https://doi.org/10.1016/j.preteyeres.2020.100879CrossRefPubMed

20.

Brown T, Mann B, Ryder N, Subbiah M, Kaplan JD, Dhariwal P, Neelakantan A, Shyam P, Sastry G, Askell A (2020) Language models are few-shot learners. Adv Neural Inf Process Syst 33:1877–1901

21.

OpenAI (2023) GPT-4 is OpenAI’s most advanced system, producing safer and more useful responses. OpenAI San Francisco, CA, USA. https://chat.openai.com/

22.

Singh SK, Kumar S, Mehra PS (2023) Chat GPT & Google Bard AI: a review 2023 International Conference on IoT, Communication and Automation Technology (ICICAT). IEEE 1:1–6

23.

Thoppilan R, De Freitas D, Hall J, Shazeer N, Kulshreshtha A, Cheng H-T, Jin A, Bos T, Baker L, Du Y (2022) Lamda: language models for dialog applications. arXiv:2201.08239

24.

Pichai S, Hassabis D (2023) Introducing Gemini: our largest and most capable AI model. Google Retrieved December 8 2023. https://blog.google/intl/en-africa/company-news/technology/introducing-gemini-our-largest-and-most-capable-ai-model/

25.

Bernard A, Langille M, Hughes S, Rose C, Leddin D, Van Zanten SV (2007) A systematic review of patient inflammatory bowel disease information resources on the World Wide Web. Am J Gastroenterol 102:2070–2077CrossRefPubMed

26.

Dave T, Athaluri SA, Singh S (2023) ChatGPT in medicine: an overview of its applications, advantages, limitations, future prospects, and ethical considerations. Front Artif Intell 6:1169595. https://doi.org/10.3389/frai.2023.1169595CrossRefPubMedPubMedCentral

27.

Pryss R, Kraft R, Baumeister H, Winkler J, Probst T, Reichert M, Langguth B, Spiliopoulou M, Schlee W (2019) Using Chatbots to support medical and psychological treatment procedures: challenges, opportunities, technologies, reference architecture. Digital Phenotyping and Mobile Sensing: New Developments in Psychoinformatics 1:249–260CrossRef

28.

Zagabathuni Y (2022) Applications, scope, and challenges for AI in healthcare. Int J 10:195–199

29.

Ren LY (2019) Product: Isabel Pro–the DDX generator. The Journal of the Canadian Health Libraries Association= Journal de l'Association des Bibliothèques de la Santé du Canada 40: 63–69

30.

Delsoz M, Raja H, Madadi Y, Tang AA, Wirostko BM, Kahook MY, Yousefi S (2023) The use of ChatGPT to assist in diagnosing glaucoma based on clinical case reports. Ophthalmol Ther 12:3121–3132. https://doi.org/10.1007/s40123-023-00805-xCrossRefPubMedPubMedCentral

31.

Kianian R, Sun D, Crowell EL, Tsui E (2023) The use of large language models to generate education materials about uveitis. Ophthalmol Retina 8(2):195–201. https://doi.org/10.1016/j.oret.2023.09.008CrossRefPubMed

32.

Alser M, Waisberg E (2023) Concerns with the usage of ChatGPT in academia and medicine: a viewpoint. Am J Med Open 9(100036):1–2

33.

Marks J, Harding A, Harper R, Williams E, Haque S, Spencer A, Fenerty C (2012) Agreement between specially trained and accredited optometrists and glaucoma specialist consultant ophthalmologists in their management of glaucoma patients. Eye 26:853–861CrossRefPubMedPubMedCentral

34.

Fisher S, Rosella LC (2022) Priorities for successful use of artificial intelligence by public health organizations: a literature review. BMC Public Health 22:2146CrossRefPubMedPubMedCentral

35.

Carlà MM, Gambini G, Baldascino A, Giannuzzi F, Boselli F, Crincoli E, D’Onofrio NC, Rizzo S (2024) Exploring AI-chatbots’ capability to suggest surgical planning in ophthalmology: ChatGPT versus Google Gemini analysis of retinal detachment cases. British Journal of Ophthalmology: bjo-2023–325143 https://doi.org/10.1136/bjo-2023-325143

36.

Radford A, Kim JW, Hallacy C, Ramesh A, Goh G, Agarwal S, Sastry G, Askell A, Mishkin P, Clark J (2021) Learning transferable visual models from natural language supervision. International conference on machine learning. arXiv:2103.00020

Title: Large language models as assistance for glaucoma surgical cases: a ChatGPT vs. Google Gemini comparison
Authors: Matteo Mario Carlà
Gloria Gambini
Antonio Baldascino
Francesco Boselli
Federico Giannuzzi
Fabio Margollicci
Stanislao Rizzo
Publication date: 04-04-2024
Publisher: Springer Berlin Heidelberg
Keywords: Glaucoma
Trabeculectomy
Artificial Intelligence
Cataract
Published in: Graefe's Archive for Clinical and Experimental Ophthalmology
Print ISSN: 0721-832X
Electronic ISSN: 1435-702X
DOI: https://doi.org/10.1007/s00417-024-06470-5

Keynote webinar | Spotlight on medication adherence

Springer Medicine

Large language models as assistance for glaucoma surgical cases: a ChatGPT vs. Google Gemini comparison

Abstract

Purpose

Methods

Results

Conclusion

Keynote webinar | Spotlight on medication adherence

Springer Medicine

Abstract

Purpose

Methods

Results

Conclusion

Please log in to get access to this content