On this page you will find a search on “Chat GPT and Ophthalmology” in Consensus.

This is best seen on a computer, not a phone.

Important!!

  1. The link to Consensus may not work well if you do not have an account. If they do not work for you use the DOI’.

  2. To use the DOI as a link, copy and paste it into your browser. This will take you to several pages, where you will often find the full text.

  3. The links will open on this page when clicked. Use your browser's “back” arrow to return to this page.

Assessing the medical reasoning skills of GPT-4 in complex ophthalmology cases

GPT-4 improves its performance in complex ophthalmology cases using zero-shot prompting strategies, but does not surpass senior residents in diagnostic accuracy and next-step accuracy.

Authors: Daniel Milad, F. Antaki, Jason Milad, Andrew Farah, Thomas Khairy, David Mikhail, Charles-Édouard Giguère, Samir Touma, Allison Bernstein, Andrei-Alexandru Szigiato, Taylor Nayman, G. Mullie, Renaud Duval

Year: 2024

Citations: 10

Study Type: non-rct observational study

Journal: British Journal of Ophthalmology

SJR Quartile: 1.0

DOI: 10.1136/bjo-2023-325053

Consensus Link: https://consensus.app/papers/assessing-the-medical-reasoning-skills-of-gpt4-in-complex-milad-antaki/a8a46b4765055e82bfb66b03d8e7a85d/

---

Unveiling the clinical incapabilities: a benchmarking study of GPT-4V(ision) for ophthalmic multimodal image analysis

GPT-4V is not yet suitable for clinical decision-making in ophthalmology, with weaker performance in lesion identification, diagnosis, and decision support.

Authors: Pusheng Xu, Xiaolan Chen, Ziwei Zhao, Danli Shi

Year: 2024

Citations: 3

Study Type: non-rct observational study

Journal: British Journal of Ophthalmology

SJR Quartile: 1.0

DOI: 10.1136/bjo-2023-325054

Consensus Link: https://consensus.app/papers/unveiling-the-clinical-incapabilities-a-benchmarking-xu-chen/2b3165991a71545d8fa17cef6dde5908/

---

Accuracy of an Artificial Intelligence Chatbot's Interpretation of Clinical Ophthalmic Images.

The ChatGPT-4 chatbot accurately responded to approximately two-thirds of multiple-choice questions pertaining to ophthalmic cases based on imaging interpretation, with better performance on nonimage-based questions.

Authors: Andrew Mihalache, Ryan S. Huang, Marko M. Popovic, Nikhil S Patil, Bhadra U. Pandya, Reut Shor, Austin Pereira, Jason Kwok, Peng Yan, David T. Wong, P. Kertes, Rajeev H. Muni

Year: 2024

Citations: 19

Study Type: non-rct observational study

Journal: JAMA ophthalmology

SJR Quartile: 1.0

DOI: 10.1001/jamaophthalmol.2024.0017

Consensus Link: https://consensus.app/papers/accuracy-of-an-artificial-intelligence-chatbots-mihalache-huang/d5bbeb54c4a35a9796649466349419c9/

---

Application, investigation and prediction of ChatGpt/GPT-4 for clinical cases in medical field

ChatGPT and GPT-4 large language models show professional and powerful abilities in analyzing medical cases, outperforming professional clinicians in clinical diagnosis and rehabilitation.

Authors: Xingyu Zhao

Year: 2024

Citations: 0

Study Type: N/A

Journal: Applied and Computational Engineering

SJR Quartile: N/A

DOI: 10.54254/2755-2721/54/20241225

Consensus Link: https://consensus.app/papers/application-investigation-and-prediction-of-chatgptgpt4-zhao/c571e429c326560f93296bc9949793db/

---

Exploring Diagnostic Precision and Triage Proficiency: A Comparative Study of GPT-4 and Bard in Addressing Common Ophthalmic Complaints

Chatbots like GPT-4 and Bard are better at ophthalmic triage than diagnosis, but they are not a replacement for professional ophthalmic evaluation or advice.

Authors: Roya Zandi, Joseph D. Fahey, Michael Drakopoulos, John M. Bryan, Siyuan Dong, Paul J Bryar, Ann E. Bidwell, R. C. Bowen, Jeremy A. Lavine, R. Mirza

Year: 2024

Citations: 14

Study Type: non-rct observational study

Journal: Bioengineering

SJR Quartile: 3.0

DOI: 10.3390/bioengineering11020120

Consensus Link: https://consensus.app/papers/exploring-diagnostic-precision-and-triage-proficiency-a-zandi-fahey/198747c9998a51bb9a39c126382ac6db/

---

Fine-tuning Large Language Model (LLM) Artificial Intelligence Chatbots in Ophthalmology and LLM-based evaluation using GPT-4

GPT-4-based evaluation significantly aligns with human clinician rankings, allowing for efficient and automated evaluation of LLM chatbot responses to healthcare-related queries.

Authors: Ting Fang Tan, Kabilan Elangovan, Liyuan Jin, Yao Jie, Li Yong, Joshua Lim, Stanley Poh, Wei Yan Ng, D. Lim, Yuhe Ke, Nan Liu, D. Ting

Year: 2024

Citations: 4

Study Type: N/A

Journal: ArXiv

SJR Quartile: N/A

DOI: 10.48550/arXiv.2402.10083

Consensus Link: https://consensus.app/papers/finetuning-large-language-model-llm-artificial-tan-elangovan/9e5bb601ae735c7a8c5ae00001d6e5b9/

---

Performance of ChatGPT on Ophthalmology-Related Questions Across Various Examination Levels: Observational Study

ChatGPT-3.5 and 4.0 show potential in ophthalmology-related questions, but future models with higher accuracy are needed for mainstream medical education.

Authors: Firas Haddad, Joanna S Saade

Year: 2024

Citations: 9

Study Type: non-rct observational study

Journal: JMIR Medical Education

SJR Quartile: 1.0

DOI: 10.2196/50842

Consensus Link: https://consensus.app/papers/performance-of-chatgpt-on-ophthalmologyrelated-haddad-saade/8618d2a4fb5958fc994148dff2fe0a2b/

---

Evaluating ChatGPT-4’s Accuracy in Identifying Final Diagnoses Within Differential Diagnoses Compared With Those of Physicians: Experimental Study for Diagnostic Cases

ChatGPT-4 shows fair to good agreement in identifying the final diagnosis from differential-diagnosis lists, comparable to physicians for case report series.

Authors: Takanobu Hirosawa, Yukinori Harada, Kazuya Mizuta, Tetsu Sakamoto, K. Tokumasu, Taro Shimizu

Year: 2024

Citations: 2

Study Type: non-rct observational study

Journal: JMIR Formative Research

SJR Quartile: 2.0

DOI: 10.2196/59267

Consensus Link: https://consensus.app/papers/evaluating-chatgpt4-%E2%80%99-s-accuracy-in-identifying-final-hirosawa-harada/21849f1ec03f593d858f95bb577acb7a/

---

Testing the Ability and Limitations of ChatGPT to Generate Differential Diagnoses from Transcribed Radiologic Findings.

ChatGPT is most accurate when using the most up-to-date model (GPT-4) and prompted for a single diagnosis, but repeatability is an issue for both models.

Authors: Shawn H Sun, K. Huynh, Gillean Cortes, Robert Hill, Julia Tran, Leslie Yeh, Amanda L Ngo, R. Houshyar, Vahid Yaghmai, Mark Tran

Year: 2024

Citations: 1

Study Type: non-rct observational study

Journal: Radiology

SJR Quartile: 1.0

DOI: 10.1148/radiol.232346

Consensus Link: https://consensus.app/papers/testing-the-ability-and-limitations-of-chatgpt-to-generate-sun-huynh/05775922b2305606a3a25a29fc144ade/

---

The Comparison of ChatGPT 3.5, Microsoft Bing, and Google Gemini for Diagnosing Cases of Neuro-Ophthalmology

ChatGPT 3.5 performs better than Microsoft Bing and Google Gemini in diagnosing neuro-ophthalmological cases, with potential benefits for medical education and ocular diagnostics.

Authors: Ruchi Shukla, A. Mishra, Nilakshi Banerjee, Archana Verma

Year: 2024

Citations: 5

Study Type: non-rct observational study

Journal: Cureus

SJR Quartile: N/A

DOI: 10.7759/cureus.58232

Consensus Link: https://consensus.app/papers/the-comparison-of-chatgpt-35-microsoft-bing-and-google-shukla-mishra/ea74df5fc9c25077b7d60c0ea11f37c1/

---

Performance of ChatGPT in Diagnosis of Corneal Eye Diseases

ChatGPT-4.0 has significantly improved accuracy in diagnosing various corneal conditions compared to ChatGPT-3.5, offering potential for clinical integration.

Authors: Mohmmad Delsoz, Yeganeh Madadi, Hina Raja, Wuqaas M. Munir, Brendan Tamm, Shiva Mehravaran, Mohammad Soleimani, Ali R. Djalilian, Siamak Yousefi

Year: 2024

Citations: 7

Study Type: non-rct observational study

Journal: Cornea

SJR Quartile: 1.0

DOI: 10.1097/ICO.0000000000003492

Consensus Link: https://consensus.app/papers/performance-of-chatgpt-in-diagnosis-of-corneal-eye-delsoz-madadi/365a204484925683a3393d63ebf84ac7/

---

The Diagnostic Ability of GPT-3.5 and GPT-4.0 in Surgery: Comparative Analysis

ChatGPT, particularly GPT-4.0, shows higher diagnostic accuracy than GPT-3.5 for colon cancer, but still has limitations in recognizing patient symptoms and laboratory data.

Authors: Jiayu Liu, Xiuting Liang, Dandong Fang, Jiqi Zheng, Chengliang Yin, Hui Xie, Yanteng Li, Xiaochun Sun, Yue Tong, Hebin Che, Ping Hu, Fan Yang, Bingxian Wang, Yuanyuan Chen, Gang Cheng, Jianning Zhang

Year: 2024

Citations: 0

Study Type: non-rct observational study

Journal: Journal of Medical Internet Research

SJR Quartile: 1.0

DOI: 10.2196/54985

Consensus Link: https://consensus.app/papers/the-diagnostic-ability-of-gpt35-and-gpt40-in-surgery-liu-liang/a17a14f0d4df5a42935158a2d01fa826/

---

ChatGPT-4 Consistency in Interpreting Laryngeal Clinical Images of Common Lesions and Disorders.

ChatGPT-4 is more efficient in primary diagnosis than image analysis, selecting the most appropriate additional examinations and treatments.

Authors: A. Maniaci, C. Chiesa-Estomba, Jerome R. Lechien

Year: 2024

Citations: 0

Study Type: non-rct observational study

Journal: Otolaryngology--head and neck surgery : official journal of American Academy of Otolaryngology-Head and Neck Surgery

SJR Quartile: 1.0

DOI: 10.1002/ohn.897

Consensus Link: https://consensus.app/papers/chatgpt4-consistency-in-interpreting-laryngeal-clinical-maniaci-chiesa-estomba/76db16e1ffda5fe797405ad90f92ed70/

---

Evaluation of ChatGPT as a diagnostic tool for medical learners and clinicians

ChatGPT is not accurate as a diagnostic tool, but offers utility as an educational tool by ruling out specific differential diagnoses and providing reasonable next steps for medical learners.

Authors: Ali Hadi, Edward Tran, Branavan Nagarajan, A. Kirpalani

Year: 2024

Citations: 2

Study Type: non-rct observational study

Journal: PLOS ONE

SJR Quartile: 1.0

DOI: 10.1371/journal.pone.0307383

Consensus Link: https://consensus.app/papers/evaluation-of-chatgpt-as-a-diagnostic-tool-for-medical-hadi-tran/4a7825ec0c715451ba01fa2b91bc3dd3/

---

A Quantitative Assessment of ChatGPT as a Neurosurgical Triaging Tool.

ChatGPT 4 accurately diagnoses and triages neurosurgical scenarios at the level of a senior neurosurgical resident, with improvements between GPT 3.5 and 4.

Authors: Max Ward, Prashin Unadkat, Daniel Toscano, A. Kashanian, Daniel G Lynch, Alexander C Horn, Randy S. D’Amico, Mark A. Mittler, Griffin R Baum

Year: 2024

Citations: 8

Study Type: non-rct observational study

Journal: Neurosurgery

SJR Quartile: 1.0

DOI: 10.1227/neu.0000000000002867

Consensus Link: https://consensus.app/papers/a-quantitative-assessment-of-chatgpt-as-a-neurosurgical-ward-unadkat/03e6add53e125d85b1c09278f93bdf08/

---

Assessment of a Large Language Model's Responses to Questions and Cases About Glaucoma and Retina Management.

Large language models (LLMs) chatbots show promising diagnostic accuracy and completeness compared to fellowship-trained ophthalmologists in glaucoma and retina management.

Authors: Andy S Huang, Kyle Hirabayashi, Laura Barna, Deep Parikh, Louis R. Pasquale

Year: 2024

Citations: 15

Study Type: non-rct observational study

Journal: JAMA ophthalmology

SJR Quartile: 1.0

DOI: 10.1001/jamaophthalmol.2023.6917

Consensus Link: https://consensus.app/papers/assessment-of-a-large-language-models-responses-to-huang-hirabayashi/1770e8def4565ff48b29cae8e88ee698/

---

The theranostic performance of Chat-GPT against urological trauma

Chat-GPT shows potential in diagnosing and treating urological trauma, with accuracy ranging from 3 to 5 out of 5 based on clinical guidelines.

Authors: Jin Li, Xianyanling Yi, Zeyu Han, Dazhou Liao, Tianyi Zhang, Liangren Liu, Jianzhong Ai

Year: 2024

Citations: 0

Study Type: non-rct observational study

Journal: International Journal of Surgery (London, England)

SJR Quartile: 1.0

DOI: 10.1097/js9.0000000000001410

Consensus Link: https://consensus.app/papers/the-theranostic-performance-of-chatgpt-against-li-yi/d9c11ffe73c85086a6b97a901646debe/

---

Performance of GPT-4V(ision) in Ophthalmology: Use of Images in Clinical Questions

Adding images to GPT-4V improves its diagnostic accuracy in ophthalmology clinical questions, indicating the importance of integrating multimodal data in medical diagnostic tools.

Authors: K. Tomita, T. Nishida, Y. Kitaguchi, M. Miyake, K. Kitazawa

Year: 2024

Citations: 0

Study Type: non-rct observational study

Journal: N/A

SJR Quartile: N/A

DOI: 10.1101/2024.01.26.24301802

Consensus Link: https://consensus.app/papers/performance-of-gpt4vision-in-ophthalmology-use-of-images-tomita-nishida/56dbc1b12fe2594fa4e25cd4f7fe2389/

---

Accuracy Evaluation of GPT-Assisted Differential Diagnosis in Emergency Department

ChatGPT-3.5 and GPT-4 accurately predict body system diagnoses in emergency departments, but their performance in granular category levels is inconsistent.

Authors: Fatemeh Shah-Mohammadi, Joseph Finkelstein

Year: 2024

Citations: 0

Study Type: non-rct observational study

Journal: Diagnostics

SJR Quartile: 2.0

DOI: 10.3390/diagnostics14161779

Consensus Link: https://consensus.app/papers/accuracy-evaluation-of-gptassisted-differential-shah-mohammadi-finkelstein/612368bc0fa25b43934b53865f862be3/

---

Evaluating accuracy and reproducibility of ChatGPT responses to patient-based questions in Ophthalmology: An observational study

ChatGPT provides moderate accuracy and reproducibility in answering patients' ophthalmology questions, but should not replace medical advice.

Authors: A. Alqudah, Abdelwahab J. Aleshawi, Mohammed Baker, Zaina Alnajjar, Ibrahim Ayasrah, Yaqoot Ta'ani, Mohammad Al Salkhadi, Shaima'a Aljawarneh

Year: 2024

Citations: 0

Study Type: non-rct observational study

Journal: Medicine

SJR Quartile: 3.0

DOI: 10.1097/MD.0000000000039120

Consensus Link: https://consensus.app/papers/evaluating-accuracy-and-reproducibility-of-chatgpt-alqudah-aleshawi/9b5bd4856d0b5883a6051884959a3421/

---

Evaluation of ChatGPT's Usefulness and Accuracy in Diagnostic Surgical Pathology.

ChatGPT provided useful responses in 62.2% of pathology-related diagnostic cases, but its number of errors and variability suggest it is not yet adequate for everyday diagnostic practice.

Authors: V. Guastafierro, D. N. Corbitt, A. Bressan, B. Fernandes, O. Mintemur, F. Magnoli, S. Ronchi, S. La Rosa, S. Uccella, S. L. Renne

Year: 2024

Citations: 0

Study Type: N/A

Journal: N/A

SJR Quartile: N/A

DOI: 10.1101/2024.03.12.24304153

Consensus Link: https://consensus.app/papers/evaluation-of-chatgpts-usefulness-and-accuracy-in-guastafierro-corbitt/ca59df56af065145a4b57b070e50e1c2/

---

Performance of ChatGPT and GPT-4 on Polish National Specialty Exam (NSE) in Ophthalmology

GPT-4 shows significant improvement over GPT-3.5 on the Polish National Specialty Exam in ophthalmology, but both models need further development for specific medical domains like surgery and pediatrics.

Authors: Marcin Ciekalski, Maciej Laskowski, Agnieszka Koperczak, Maria Śmierciak, Sebastian Sirek

Year: 2024

Citations: 0

Study Type: N/A

Journal: Postępy Higieny i Medycyny Doświadczalnej

SJR Quartile: 4.0

DOI: 10.2478/ahem-2024-0006

Consensus Link: https://consensus.app/papers/performance-of-chatgpt-and-gpt4-on-polish-national-ciekalski-laskowski/c51c8e6a8a325c86a82f5ecbe2cf119f/

---

Evaluating Chatbot responses to patient questions in the field of glaucoma

ChatGPT shows promise in glaucoma patient education, with potential for self-correction and improved accuracy after further research and validation.

Authors: Darren Ngiap Hao Tan, Yih-Chung Tham, Victor Koh, Seng Chee Loon, Maria Cecilia Aquino, Katherine Lun, Ching-Yu Cheng, K. Ngiam, M. Tan

Year: 2024

Citations: 0

Study Type: non-rct observational study

Journal: Frontiers in Medicine

SJR Quartile: 1.0

DOI: 10.3389/fmed.2024.1359073

Consensus Link: https://consensus.app/papers/evaluating-chatbot-responses-to-patient-questions-in-the-tan-tham/72db5449328f53cf9e20df450bdf0152/

---

Evaluation of ChatGPT-Generated Differential Diagnosis for Common Diseases With Atypical Presentation: Descriptive Research

ChatGPT-4 shows potential in diagnosing typical and mildly atypical presentations of common diseases, but its performance declines with greater atypicality.

Authors: K. Shikino, Taro Shimizu, Yuki Otsuka, Masaki Tago, Hiromizu Takahashi, T. Watari, Y. Sasaki, Gemmei Iizuka, Hiroki Tamura, Koichi Nakashima, Kotaro Kunitomo, Morika Suzuki, Sayaka Aoyama, Shintaro Kosaka, T. Kawahigashi, Tomohiro Matsumoto, Fumina Orihara, Toru Morikawa, Toshinori Nishizawa, Yoji Hoshina, Yu Yamamoto, Yuichiro Matsuo, Yuto Unoki, Hirofumi Kimura, Midori Tokushima, Satoshi Watanuki, Takuma Saito, Fumio Otsuka, Y. Tokuda

Year: 2024

Citations: 1

Study Type: non-rct observational study

Journal: JMIR Medical Education

SJR Quartile: 1.0

DOI: 10.2196/58758

Consensus Link: https://consensus.app/papers/evaluation-of-chatgptgenerated-differential-diagnosis-shikino-shimizu/e019bcbdde1b59eeb58fa461ee531221/

---

Performance of GPT-4 with Vision on Text- and Image-based ACR Diagnostic Radiology In-Training Examination Questions.

GPT-4V demonstrated competence in text-based radiology questions but showed deficits in interpreting radiologic images.

Authors: Nolan Hayden, Spencer Gilbert, Laila M Poisson, Brent Griffith, C. Klochko

Year: 2024

Citations: 1

Study Type: non-rct observational study

Journal: Radiology

SJR Quartile: 1.0

DOI: 10.1148/radiol.240153

Consensus Link: https://consensus.app/papers/performance-of-gpt4-with-vision-on-text-and-imagebased-acr-hayden-gilbert/3b8ab2db76f1599c9f9f505bdf822e94/

---

ChatGPT to generate clinical vignettes for teaching and multiple-choice questions for assessment: A randomized controlled experiment.

ChatGPT generates clinical vignettes with comparable quality to human-written ones, and some multiple-choice questions with acceptable psychometric characteristics for medical education.

Authors: Özlem Coşkun, Yavuz Selim Kıyak, I. Budakoğlu

Year: 2024

Citations: 4

Study Type: rct

Journal: Medical teacher

SJR Quartile: 1.0

DOI: 10.1080/0142159X.2024.2327477

Consensus Link: https://consensus.app/papers/chatgpt-to-generate-clinical-vignettes-for-teaching-and-co%C5%9Fkun-k%C4%B1yak/f0387153fa1a53d7b692bc79958bdec4/

---

Assessing Generative Pretrained Transformers (GPT) in Clinical Decision-Making: Comparative Analysis of GPT-3.5 and GPT-4

ChatGPT-3.5 and GPT-4 show promising potential in assisting physicians with medical issues, but must complement human expertise and continue research for safe and effective implementation.

Authors: Adi Lahat, Kassem Sharif, Narmin Zoabi, Yonatan Shneor Patt, Yousra Sharif, Lior Fisher, U. Shani, M. Arow, Roni Levin, Eyal Klang

Year: 2024

Citations: 4

Study Type: non-rct observational study

Journal: Journal of Medical Internet Research

SJR Quartile: 1.0

DOI: 10.2196/54571

Consensus Link: https://consensus.app/papers/assessing-generative-pretrained-transformers-gpt-in-lahat-sharif/86e873903be2579994a64854d3ef0046/

---

Chatbots Vs. Human Experts: Evaluating Diagnostic Performance of Chatbots in Uveitis and the Perspectives on AI Adoption in Ophthalmology.

ChatGPT showed promising diagnostic capabilities in uveitis cases and ophthalmologists showed enthusiasm for integrating AI into clinical practice.

Authors: William Rojas-Carabali, Alok Sen, Aniruddha Agarwal, G. Tan, Carol Y Cheung, A. Rousselot, Rajdeep Agrawal, Renee Liu, Carlos Cifuentes-González, Tobias Elze, John H. Kempen, Lucia Sobrin, Q. Nguyen, Alejandra de-la-Torre, Bernett Lee, Vishali Gupta, Rupesh Agrawal

Year: 2023

Citations: 20

Study Type: non-rct observational study

Journal: Ocular immunology and inflammation

SJR Quartile: 2.0

DOI: 10.1080/09273948.2023.2266730

Consensus Link: https://consensus.app/papers/chatbots-vs-human-experts-evaluating-diagnostic-rojas-carabali-sen/8933f819834657cd8fa835f5a9717745/

---

The Utility of ChatGPT in Diabetic Retinopathy Risk Assessment: A Comparative Study with Clinical Diagnosis

ChatGPT shows potential as a preliminary diabetic retinopathy screening tool, with fair agreement with clinical diagnosis, but further optimization is needed for clinical use.

Authors: Keerthana Raghu, T. S, Chitralekha S Devishamani, S. M, R. Rajalakshmi, R. Raman

Year: 2023

Citations: 2

Study Type: non-rct observational study

Journal: Clinical Ophthalmology (Auckland, N.Z.)

SJR Quartile: 1.0

DOI: 10.2147/OPTH.S435052

Consensus Link: https://consensus.app/papers/the-utility-of-chatgpt-in-diabetic-retinopathy-risk-raghu-s/854255c781895b4f8975804b59d270cd/

---

Evaluating the Artificial Intelligence Performance Growth in Ophthalmic Knowledge

ChatGPT-4.0 significantly outperforms GPT-3.5 in addressing ophthalmic case challenges, particularly in neuro-ophthalmology, with improved accuracy even in image-related questions.

Authors: Cheng Jiao, Neel R Edupuganti, Parth A. Patel, Tommy Bui, Veeral Sheth

Year: 2023

Citations: 11

Study Type: N/A

Journal: Cureus

SJR Quartile: N/A

DOI: 10.7759/cureus.45700

Consensus Link: https://consensus.app/papers/evaluating-the-artificial-intelligence-performance-jiao-edupuganti/57228859c043532ba7aa5e4e9decfdad/

---

GPT-4 Multimodal Analysis on Ophthalmology Clinical Cases Including Text and Images

GPT-4V shows potential in accurately diagnosing ocular pathologies by integrating visual and textual data, but its current stage is not suitable for clinical application.

Authors: MD Vera Sorin, MD Noa Kapelushnik, MD Idan Hecht, MD Ofira Zloto, PhD Benjamin S. Glicksberg, MD Hila Bufman, MD Yiftach Barash, MD Girish N. Nadkarni, MD Eyal Klang

Year: 2023

Citations: 10

Study Type: non-rct observational study

Journal: N/A

SJR Quartile: N/A

DOI: 10.1101/2023.11.24.23298953

Consensus Link: https://consensus.app/papers/gpt4-multimodal-analysis-on-ophthalmology-clinical-cases-sorin-kapelushnik/09689e86872356c4bba7e06de49d4173/

---

Artificial Intelligence in Ophthalmology: A Comparative Analysis of GPT-3.5, GPT-4, and Human Expertise in Answering StatPearls Questions

GPT-4 significantly outperforms GPT-3.5 and human professionals on StatPearls ophthalmology questions, highlighting the potential of advanced conversational AI systems in medicine.

Authors: M. Moshirfar, Amal W Altaf, Isabella M. Stoakes, Jared J Tuttle, P. Hoopes

Year: 2023

Citations: 56

Study Type: non-rct observational study

Journal: Cureus

SJR Quartile: N/A

DOI: 10.7759/cureus.40822

Consensus Link: https://consensus.app/papers/artificial-intelligence-in-ophthalmology-a-comparative-moshirfar-altaf/f0e5f8436f295da9b2eebfd6b6684d11/

---

Evaluating the Performance of ChatGPT in Ophthalmology

ChatGPT shows promising accuracy in ophthalmology, with the best results in general medicine and the worst in neuro-ophthalmology and ophthalmic pathology.

Authors: F. Antaki, Samir Touma, Daniel Milad, J. El-Khoury, R. Duval

Year: 2023

Citations: 239

Study Type: N/A

Journal: Ophthalmology Science

SJR Quartile: 1.0

DOI: 10.1101/2023.01.22.23284882

Consensus Link: https://consensus.app/papers/evaluating-the-performance-of-chatgpt-in-ophthalmology-antaki-touma/58b143e499635607817166b7b223e1c7/

---

ChatGPT Assisting Diagnosis of Neuro-ophthalmology Diseases Based on Case Reports

ChatGPT v3.5 and ChatGPT Plus v4.0 accurately diagnose neuro-ophthalmic diseases, with potential for use in clinical care settings.

Authors: Yeganeh Madadi, M. Delsoz, Priscilla A. Lao, Joseph W. Fong, TJ Hollingsworth, M. Kahook, Siamak Yousefi

Year: 2023

Citations: 15

Study Type: non-rct observational study

Journal: medRxiv

SJR Quartile: N/A

DOI: 10.1101/2023.09.13.23295508

Consensus Link: https://consensus.app/papers/chatgpt-assisting-diagnosis-of-neuroophthalmology-madadi-delsoz/cb51fc83611c57bda938cf4e9521023b/

---

Evaluation of a digital ophthalmologist app built by GPT4-Vision

GPT4-Vision chatbot shows potential for interpreting ocular multimodal images, but currently lacks reliability for clinical decision-making and patient consultation in ophthalmology.

Authors: MD Pusheng Xu, MD Xiaolan Chen, Ziwei Zhao, MD PhD Yingfeng Zheng, MD PhD Guangming Jin, MD PhD Danli Shi, MD PhD Mingguang He

Year: 2023

Citations: 7

Study Type: non-rct observational study

Journal: N/A

SJR Quartile: N/A

DOI: 10.1101/2023.11.27.23299056

Consensus Link: https://consensus.app/papers/evaluation-of-a-digital-ophthalmologist-app-built-by-xu-chen/795ed4254f205782977f1408f57c0aa6/

---

Utility of ChatGPT in Clinical Practice

ChatGPT is a valuable tool in clinical practice, improving clinical decision-making and patient care, but its use must be carefully considered for potential risks and challenges.

Authors: Jialin Liu, Changyu Wang, Siru Liu

Year: 2023

Citations: 172

Study Type: N/A

Journal: Journal of Medical Internet Research

SJR Quartile: 1.0

DOI: 10.2196/48568

Consensus Link: https://consensus.app/papers/utility-of-chatgpt-in-clinical-practice-liu-wang/90c7cc7683d95bc2b85daefb23dcb8ce/

---

Abstract 16022: Evaluating the Utility of OpenAI’s GPT-4 as a Diagnostic and Management Aid in Medicine

GPT-4, OpenAI's most advanced model, can consider the correct diagnosis in differentials and provide appropriate management decisions for common problems, but only correctly identified the diagnosis in one out of five cases.

Authors: Alan Ge, Vidish Pandya, K. Ferrick, Andrew Krumerman

Year: 2023

Citations: 0

Study Type: N/A

Journal: Circulation

SJR Quartile: 1.0

DOI: 10.1161/circ.148.suppl_1.16022

Consensus Link: https://consensus.app/papers/abstract-16022-evaluating-the-utility-of-openai-%E2%80%99-s-gpt4-as-a-ge-pandya/4fd15d5a5852547ca4eeea25c35013c5/

---

Exploring the role of ChatGPT in patient care (diagnosis and treatment) and medical research: A systematic review

ChatGPT can aid in patient care and research, but has limitations due to concerns about accuracy, authorship, and bias.

Authors: R. Garg, V. L. Urs, Akshya Anand Agrawal, Sarvesh Kumar Chaudhary, V. Paliwal, Sujita Kumar Kar

Year: 2023

Citations: 59

Study Type: systematic review

Journal: Health Promotion Perspectives

SJR Quartile: 1.0

DOI: 10.1101/2023.06.13.23291311

Consensus Link: https://consensus.app/papers/exploring-the-role-of-chatgpt-in-patient-care-diagnosis-and-garg-urs/123058a55fa8515dad5652527109dfea/

---

Chat GPT vs Clinician: challenging the diagnostic capabilities of A.I. in dermatology.

Chat GPT has diagnostic capabilities in dermatology, but in its current form does not significantly improve diagnostic yield in primary or secondary care.

Authors: Sophie Stoneham, A. Livesey, H. Cooper, C. Mitchell

Year: 2023

Citations: 8

Study Type: non-rct observational study

Journal: Clinical and experimental dermatology

SJR Quartile: 2.0

DOI: 10.1093/ced/llad402

Consensus Link: https://consensus.app/papers/chat-gpt-vs-clinician-challenging-the-diagnostic-stoneham-livesey/db8b522884325c088c43baa87711e1a9/

---

Diagnostic Accuracy of Differential-Diagnosis Lists Generated by Generative Pretrained Transformer 3 Chatbot for Clinical Vignettes with Common Chief Complaints: A Pilot Study

ChatGPT-3 chatbot has high diagnostic accuracy for differential diagnoses in clinical cases with common chief complaints, but the order of these lists could be improved in the future.

Authors: Takanobu Hirosawa, Y. Harada, M. Yokose, Tetsu Sakamoto, Ren Kawamura, Taro Shimizu

Year: 2023

Citations: 176

Study Type: non-rct observational study

Journal: International Journal of Environmental Research and Public Health

SJR Quartile: 2.0

DOI: 10.3390/ijerph20043378

Consensus Link: https://consensus.app/papers/diagnostic-accuracy-of-differentialdiagnosis-lists-hirosawa-harada/f507217fd42a56a8ac6c164b9903e6ee/

---

Assessing the Utility of ChatGPT Throughout the Entire Clinical Workflow: Development and Usability Study

ChatGPT achieves impressive accuracy in clinical decision-making, particularly in final diagnosis tasks, but limitations include model hallucinations and unclear training data set composition.

Authors: Arya Rao, Michael Pang, John Kim, M. Kamineni, Winston Lie, Anoop K Prasad, A. Landman, Keith Dreyer, M. Succi

Year: 2023

Citations: 96

Study Type: non-rct observational study

Journal: Journal of Medical Internet Research

SJR Quartile: 1.0

DOI: 10.2196/48659

Consensus Link: https://consensus.app/papers/assessing-the-utility-of-chatgpt-throughout-the-entire-rao-pang/53f9dcb9f2b85ba48ba262608e1253fc/

---

Uncovering Language Disparity of ChatGPT on Retinal Vascular Disease Classification: Cross-Sectional Study

ChatGPT can provide diagnostic performance for retinal vascular diseases in non-English clinical environments, but still has performance gaps and language disparities compared to professionals.

Authors: Xiaocong Liu, Jiageng Wu, An Shao, Wenyue Shen, Panpan Ye, Yao Wang, Juan Ye, Kai Jin, Jie Yang

Year: 2023

Citations: 13

Study Type: non-rct observational study

Journal: Journal of Medical Internet Research

SJR Quartile: 1.0

DOI: 10.2196/51926

Consensus Link: https://consensus.app/papers/uncovering-language-disparity-of-chatgpt-on-retinal-liu-wu/3d07d59f3a52538b802bca67d8b5c8bf/

---

Are Different Versions of ChatGPT’s Ability Comparable to the Clinical Diagnosis Presented in Case Reports? A Descriptive Study

ChatGPT shows potential for clinical decision-making, but integrating it with electronic health record systems is necessary for enhanced accuracy.

Authors: Jingfang Chen, Linlin Liu, Shujin Ruan, Mengjun Li, Chengliang Yin

Year: 2023

Citations: 1

Study Type: non-rct observational study

Journal: Journal of Multidisciplinary Healthcare

SJR Quartile: 1.0

DOI: 10.2147/JMDH.S441790

Consensus Link: https://consensus.app/papers/are-different-versions-of-chatgpt-%E2%80%99-s-ability-comparable-to-chen-liu/aafc3e7426865e0f8bbfedff3674e20e/

---

Performance of an Artificial Intelligence Chatbot in Ophthalmic Knowledge Assessment.

ChatGPT correctly answered approximately half of ophthalmic board certification practice questions in the OphthoQuestions free trial, with better performance in general medicine and less in retina and vitreous.

Authors: Andrew Mihalache, M. Popovic, Rajeev H. Muni

Year: 2023

Citations: 151

Study Type: non-rct observational study

Journal: JAMA ophthalmology

SJR Quartile: 1.0

DOI: 10.1001/jamaophthalmol.2023.1144

Consensus Link: https://consensus.app/papers/performance-of-an-artificial-intelligence-chatbot-in-mihalache-popovic/bdd2b20e68ba50768096aec56f9d88be/

---

Assessment of ChatGPT in the preclinical management of ophthalmological emergencies - an analysis of ten fictional case vignettes

ChatGPT shows 87.2% triage accuracy and appropriate preclinical measures recommendations for ophthalmological emergencies, but its safety and efficacy in preclinical management require regular reassessment.

Authors: D. Knebel, S. Priglinger, N. Scherer, J. Siedlecki, B. Schworm

Year: 2023

Citations: 7

Study Type: non-rct observational study

Journal: N/A

SJR Quartile: N/A

DOI: 10.1101/2023.04.16.23288645

Consensus Link: https://consensus.app/papers/assessment-of-chatgpt-in-the-preclinical-management-of-knebel-priglinger/f0c93278a4e3533194431631b5c531a2/

---

ChatGPT-Generated Differential Diagnosis Lists for Complex Case–Derived Clinical Vignettes: Diagnostic Accuracy Evaluation

ChatGPT-4 shows potential diagnostic accuracy in generating differential diagnosis lists for complex clinical vignettes from general internal medicine department case reports, with a rate of correct diagnoses exceeding 80%.

Authors: Takanobu Hirosawa, Ren Kawamura, Y. Harada, Kazuya Mizuta, K. Tokumasu, Yuki Kaji, Tomoharu Suzuki, Taro Shimizu

Year: 2023

Citations: 42

Study Type: N/A

Journal: JMIR Medical Informatics

SJR Quartile: 2.0

DOI: 10.2196/48808

Consensus Link: https://consensus.app/papers/chatgptgenerated-differential-diagnosis-lists-for-hirosawa-kawamura/262865b4112f521a9b56d39c84b5a3ef/

---

Assessment of ChatGPT in the Prehospital Management of Ophthalmological Emergencies – An Analysis of 10 Fictional Case Vignettes

ChatGPT shows 93.6% triage accuracy and most appropriate prehospital measures for ophthalmological emergencies, but its safety and efficacy in prehospital management of these emergencies need further assessment.

Authors: D. Knebel, Siegfried Priglinger, Nicolas Scherer, Julian Klaas, J. Siedlecki, B. Schworm

Year: 2023

Citations: 10

Study Type: non-rct observational study

Journal: Klinische Monatsblätter für Augenheilkunde

SJR Quartile: 3.0

DOI: 10.1055/a-2149-0447

Consensus Link: https://consensus.app/papers/assessment-of-chatgpt-in-the-prehospital-management-of-knebel-priglinger/c39965fb23fa5b7280f63572f3d99b38/

---

FROM TEXT TO DIAGNOSE: CHATGPT'S EFFICACY IN MEDICAL DECISION-MAKING.

ChatGPT shows strong potential as a diagnostic tool in clinical medicine, outperforming physicians in accuracy and highlighting the need for supplementary data and complexity of medical diagnosis.

Authors: Yaroslav Mykhalko, Pavlo Kish, Yelyzaveta І Rubtsova, Oleksandr Kutsyn, V. Koval

Year: 2023

Citations: 5

Study Type: non-rct experimental

Journal: Wiadomosci lekarskie

SJR Quartile: 4.0

DOI: 10.36740/WLek202311101

Consensus Link: https://consensus.app/papers/from-text-to-diagnose-chatgpts-efficacy-in-medical-mykhalko-kish/35bf46e8ad635fada788aed64658f256/

---

Using ChatGPT in Medical Research: Current Status and Future Directions

ChatGPT has the potential to revolutionize medical research, but accuracy, originality, academic integrity, and ethical issues need improvement before widespread implementation in clinical research and medical practice.

Authors: Suebsarn Ruksakulpiwat, Ayanesh Kumar, Anuoluwapo Ajibade

Year: 2023

Citations: 61

Study Type: systematic review

Journal: Journal of Multidisciplinary Healthcare

SJR Quartile: 1.0

DOI: 10.2147/JMDH.S413470

Consensus Link: https://consensus.app/papers/using-chatgpt-in-medical-research-current-status-and-ruksakulpiwat-kumar/35e0eae6e1175f26935e9b3930b7e69c/

---

Limitations of GPT‐4 in analyzing real‐life medical notes related to cognitive impairment

GPT-4 can help clinicians make 70% of correct diagnoses of cognitive impairment, but has limitations in correlating medical histories with neuroimaging findings and identifying subtle clinical features.

Authors: Y. Shea, Nok-Yee Charlotte Ma

Year: 2023

Citations: 0

Study Type: N/A

Journal: Psychogeriatrics

SJR Quartile: 2.0

DOI: 10.1111/psyg.13002

Consensus Link: https://consensus.app/papers/limitations-of-gpt%E2%80%904-in-analyzing-real%E2%80%90life-medical-notes-shea-ma/a6fae0b66ae85638b3417f4d66e90965/

---

Capabilities of GPT-4 in ophthalmology: an analysis of model entropy and progress towards human-level medical question answering

GPT-4, an LLM trained on non-ophthalmology-specific data, performs significantly better than its predecessor on simulated ophthalmology board-style exams, with potential for superiority to historical human performance.

Authors: F. Antaki, Daniel Milad, Mark A Chia, Charles-Édouard Giguère, Samir Touma, J. El-Khoury, P. Keane, Renaud Duval

Year: 2023

Citations: 24

Study Type: N/A

Journal: British Journal of Ophthalmology

SJR Quartile: 1.0

DOI: 10.1136/bjo-2023-324438

Consensus Link: https://consensus.app/papers/capabilities-of-gpt4-in-ophthalmology-an-analysis-of-model-antaki-milad/676044ab69995fb79dda9f7011712448/

---

Assessing the Utility of ChatGPT Throughout the Entire Clinical Workflow

ChatGPT achieves impressive accuracy in clinical decision making, with particular strengths emerging as it has more clinical information at its disposal.

Authors: Arya Rao, Michael Pang, John Kim, M. Kamineni, Winston Lie, Anoop K Prasad, A. Landman, K. Dreyer, M. Succi

Year: 2023

Citations: 78

Study Type: non-rct observational study

Journal: medRxiv : the preprint server for health sciences

SJR Quartile: N/A

DOI: 10.1101/2023.02.21.23285886

Consensus Link: https://consensus.app/papers/assessing-the-utility-of-chatgpt-throughout-the-entire-rao-pang/9885b9bab06056d0baa2617425f73def/

---

ChatGPT and Ophthalmology: Exploring Its Potential with Discharge Summaries and Operative Notes

ChatGPT shows promising performance in constructing ophthalmic discharge summaries and operative notes, with potential for positive healthcare impact when trained and verified by human experts.

Authors: Swati Singh, A. Djalilian, M. Ali

Year: 2023

Citations: 79

Study Type: non-rct observational study

Journal: Seminars in Ophthalmology

SJR Quartile: 2.0

DOI: 10.1080/08820538.2023.2209166

Consensus Link: https://consensus.app/papers/chatgpt-and-ophthalmology-exploring-its-potential-with-singh-djalilian/8d2bee80f98e503e9e13176fab67e4c6/

---

Use of GPT-4 to Diagnose Complex Clinical Cases

GPT-4 accurately diagnoses complex medical cases, outperforming 99.98% of simulated human readers, but further improvements and validation are needed before clinical implementation.

Authors: A. V. Eriksen, Sören Möller, J. Ryg

Year: 2023

Citations: 60

Study Type: N/A

Journal: NEJM AI

SJR Quartile: N/A

DOI: 10.1056/aip2300031

Consensus Link: https://consensus.app/papers/use-of-gpt4-to-diagnose-complex-clinical-cases-eriksen-m%C3%B6ller/66faec164f845b46b18652fdd2ce7e7a/

---

neuroGPT-X: Towards an Accountable Expert Opinion Tool for Vestibular Schwannoma

A context-enriched GPT model can provide accurate, relevant, and often superior responses to complex neurosurgical problems, outperforming experienced neurosurgeons in complex neurosurgical cases.

Authors: Edward Guo, Mansi Gupta, Sarthak Sinha, Karl Rössler, Marcos Tatagiba, R. Akagami, Ossama Al-Mefty, Taku Sugiyama, P. Stieg, Gwynedd E Pickett, Madeleine de Lotbiniere-Bassett, Rahul Singh, S. Lama, Garnette R. Sutherland

Year: 2023

Citations: 9

Study Type: non-rct observational study

Journal: N/A

SJR Quartile: N/A

DOI: 10.1101/2023.02.25.23286117

Consensus Link: https://consensus.app/papers/neurogptx-towards-an-accountable-expert-opinion-tool-for-guo-gupta/30fbf28eeb745e0abcfd02f9a780233f/

---

Evaluating ChatGPT on Orbital and Oculofacial Disorders: Accuracy and Readability Insights

ChatGPT 4.0 shows potential in providing accurate information on orbital and oculofacial diseases, with potential for educational and clinical use in ophthalmology.

Authors: M. Balas, Ana Janic, Patrick Daigle, N. Nijhawan, Ahsen Hussain, Harmeet Gill, Gabriela L Lahaie, Michel J Belliveau, Sean A Crawford, Parnian Arjmand, Edsel B. Ing

Year: 2023

Citations: 3

Study Type: non-rct observational study

Journal: Ophthalmic Plastic and Reconstructive Surgery

SJR Quartile: 2.0

DOI: 10.1097/IOP.0000000000002552

Consensus Link: https://consensus.app/papers/evaluating-chatgpt-on-orbital-and-oculofacial-disorders-balas-janic/5b041911bdb2515686538fa4616a31b8/

---

Potential of ChatGPT and GPT-4 for Data Mining of Free-Text CT Reports on Lung Cancer.

GPT-4 outperforms ChatGPT in extracting oncologic phenotypes from free-text lung cancer CT reports, demonstrating better oncologic reasoning and fewer confabulations.

Authors: M. Fink, A. Bischoff, C. Fink, Martin Moll, Jonas Kroschke, Luca Dulz, C. Heussel, H. Kauczor, T. Weber

Year: 2023

Citations: 78

Study Type: non-rct observational study

Journal: Radiology

SJR Quartile: 1.0

DOI: 10.1148/radiol.231362

Consensus Link: https://consensus.app/papers/potential-of-chatgpt-and-gpt4-for-data-mining-of-freetext-ct-fink-bischoff/efd9db5046095dd88a54c2b7599ed1f9/

---

Performance of ChatGPT in Diagnosis of Corneal Eye Diseases

ChatGPT-4.0 has significantly improved accuracy in diagnosing various corneal conditions compared to ChatGPT-3.5, making it a promising AI tool for potential clinical integration.

Authors: M. Delsoz, Yeganeh Madadi, Wuqaas M. Munir, Brendan Tamm, S. Mehravaran, Mohammad Soleimani, Ali R. Djalilian, Siamak Yousefi

Year: 2023

Citations: 15

Study Type: non-rct observational study

Journal: medRxiv

SJR Quartile: N/A

DOI: 10.1101/2023.08.25.23294635

Consensus Link: https://consensus.app/papers/performance-of-chatgpt-in-diagnosis-of-corneal-eye-delsoz-madadi/e41e19a8c5815f9aa782b4391629d199/

---

Evaluating Chatgpt As an Adjunct for Analyzing Challenging Case

ChatGPT is a comprehensive and accurate tool for diagnosing and treating challenging cases, particularly aiding inexperienced doctors in remote areas and underdeveloped medical conditions.

Authors: Xiaoni Liu, Ying Song, Hai Lin, Yiying Xu, Chao Chen, Changjian Yan, Xiaoliang Yuan, Fang Bao, Hongmei Jing, Chaoling Wu, Weilong Zhang

Year: 2023

Citations: 1

Study Type: N/A

Journal: Blood

SJR Quartile: 1.0

DOI: 10.1182/blood-2023-181518

Consensus Link: https://consensus.app/papers/evaluating-chatgpt-as-an-adjunct-for-analyzing-liu-song/35b5705218675bda999339f8443fe782/

---

Eyes on AI: ChatGPT's Transformative Potential Impact on Ophthalmology

ChatGPT has potential in ophthalmology, but accuracy, ethical integrity, and data privacy must be ensured for its successful integration into clinical, educational, and research settings.

Authors: Jason Dossantos, Jella A. An, R. Javan

Year: 2023

Citations: 10

Study Type: N/A

Journal: Cureus

SJR Quartile: N/A

DOI: 10.7759/cureus.40765

Consensus Link: https://consensus.app/papers/eyes-on-ai-chatgpts-transformative-potential-impact-on-dossantos-an/947b90fb71bc5c3aad2d5799d9ff5626/

---