ChatGPT-4: An assessment of an upgraded artificial intelligence chatbot in the United States Medical Licensing Examination

Summary: This study investigates the ability of generative artificial intelligence models to answer ophthalmology board-style questions. The results show that ChatGPT-4.0 and Bing Chat perform similarly to human respondents, but they struggle with image interpretation and may experience hallucinations and nonlogical reasoning.

AMERICAN JOURNAL OF OPHTHALMOLOGY (2023)

添加到收藏夹

Article Dermatology

Performance of ChatGPT on Specialty Certificate Examination in Dermatology multiple-choice questions

Lauren Passby et al.

Summary: ChatGPT is a language model trained by OpenAI that can perform language-based tasks and answer multiple-choice questions. In the testing, ChatGPT-4 achieved a score of 90% in sample dermatology questions, meeting the passing grade. The use of advanced artificial intelligence in medicine has significant educational and clinical implications.

CLINICAL AND EXPERIMENTAL DERMATOLOGY (2023)

添加到收藏夹

News Item Multidisciplinary Sciences

GPT-4 is here: what scientists think

Katharine Sanderson

Summary: Researchers are excited about AI, but many are frustrated by the lack of transparency in its engineering.

NATURE (2023)

添加到收藏夹

Article Ophthalmology

Performance of an Artificial Intelligence Chatbot in Ophthalmic Knowledge Assessment

Andrew Mihalache et al.

Summary: ChatGPT, an AI chatbot, had a mediocre performance in answering practice questions for ophthalmology board certification, correctly answering only about half of the questions.

JAMA OPHTHALMOLOGY (2023)

添加到收藏夹

Article Clinical Neurology

Evaluating the limits of AI in medical specialisation: ChatGPT's performance on the UK Neurology Specialty Certificate Examination

Panagiotis Giannos

Summary: The performance of ChatGPT-4 model in higher specialty training for neurology and neuroscience has shown improvement compared to its predecessors, demonstrating the potential of artificial intelligence models in specialized medical education and practice. However, ongoing development and collaboration between AI developers and medical experts are necessary to ensure the relevance and reliability of the models in the rapidly evolving field of medicine.

BMJ NEUROLOGY OPEN (2023)

添加到收藏夹

Article Multidisciplinary Sciences

AI bot ChatGPT writes smart essays — should professors worry?

Chris Stokel-Walker

NATURE (2022)

添加到收藏夹

Article Medicine, Research & Experimental

Insights from teaching artificial intelligence to medical students in Canada

Ricky Hu et al.

Summary: This article describes an AI training curriculum developed and delivered to Canadian medical undergraduates, and provides recommendations for future training programs.

COMMUNICATIONS MEDICINE (2022)

添加到收藏夹

Letter Mathematical & Computational Biology