The Role of ChatGPT, Generative Language Models, and Artificial Intelligence in Medical Education: A Conversation With ChatGPT and a Call for Papers.

Eysenbach G. · 2023

Read The Role of ChatGPT, Generative Language Models, and Artificial Intelligence in Medical Education: A Conversation With ChatGPT and a Call for Papers. by Eysenbach G. (2023) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc10028514.txt.

Introduction

Eysenbach reflects on the role of ChatGPT and similar AI tools in medical education in a 2023 paper. ChatGPT launched in late 2022 and reached a huge user base within two months. Teachers faced a tool that could write fluent medical text on demand, for good and for ill.

The paper takes the form of a conversation with ChatGPT itself about its classroom role. It weighs uses in learning and writing against risks to honesty and skill. It ends with a call for papers on the theme. The piece captures an early moment of change in medical training.

Catalog record

CREATOREysenbach G.
YEAR2023
COLLECTIONPAPERS
FORMATTXT
FILE SIZE66 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC10028514

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

The Role of ChatGPT, Generative Language Models, and Artificial Intelligence in Medical Education: A Conversation With ChatGPT and a Call for Papers By Gunther Eysenbach, Gunther Eysenbach Published in: JMIR Medical Education DOI: 10.2196/46885 ABSTRACT Abstract ChatGPT is a generative language model tool launched by OpenAI on November 30, 2022, enabling the public to converse with a machine on a broad range of topics. In January 2023, ChatGPT reached over 100 million users, making it the fastest-growing consumer application to date. This interview with ChatGPT is part 2 of a larger interview with ChatGPT. It provides a snapshot of the current capabilities of ChatGPT and illustrates the vast potential for medical education, research, and practice but also hints at current problems and limitations. In this conversation with Gunther Eysenbach, the founder and publisher of JMIR Publications, ChatGPT generated some ideas on how to use chatbots in medical education. It also illustrated its capabilities to generate a virtual patient simulation and quizzes for medical students; critiqued a simulated doctor-patient communication and attempts to summarize a research article (which turned out to be fabricated); commented on methods to detect machine-generated text to ensure academic integrity; generated a curriculum for health professionals to learn about artificial intelligence (AI); and helped to draft a call for papers for a new theme issue to be launched in JMIR Medical Education on ChatGPT. The conversation also highlighted the importance of proper “prompting.” Although the language generator does make occasional mistakes, it admits these when challenged. The well-known disturbing tendency of large language models to hallucinate became evident when ChatGPT fabricated references. The interview provides a glimpse into the capabilities and limitations of ChatGPT and the future of AI-supported medical education. Due to the impact of this new technology on medical education, JMIR Medical Education is launching a call for papers for a new e-collection and theme issue. The initial draft of the call for papers was entirely machine generated by ChatGPT, but will be edited by the human guest editors of the theme issue. Keywords: artificial intelligence, AI, ChatGPT, generative language model, medical education, interview, future of education INTRODUCTION "Any sufficiently advanced technology is indistinguishable from magic.Arthur C Clarke" Chatbots or conversational agents have been available in medicine and for consumers for several years now [1], but it was not until the release of ChatGPT, a generative language model launched by OpenAI on November 30, 2022, that the public discovered the power of a chatbot trained on approximately 300 billion words with the capability to converse on a broad range of topics. ChatGPT reached 57 million users within the first month and 100 million users by January 2023 [2], making it the fastest-growing consumer application of all time. A recent study [3] published in JMIR Medical Education on February 8, 2023, evaluated the potential of ChatGPT as a medical education tool. The study found that ChatGPT reaches the equivalent of a passing score for a third-year medical student [3]. Given the significance of ChatGPT and other generative language models for the future of medical education, we are now launching a theme issue (and ongoing e-collection of articles) on this topic [4]. This interview with ChatGPT (February 13, 2023 release) conducted by Gunther Eysenbach (GE), the founder and publisher of JMIR Publications, is part 2 of a larger interview with ChatGPT (part 1 will be published in the Journal of Medical Internet Research). The interview is completely unedited, meaning the responses of ChatGPT have not been copyedited or altered in any way, in order to provide a snapshot of the current capabilities of ChatGPT and to illustrate its potential for medical education, research, and practice. THE INTERVIEW GE: What is the role of generative language models in medical education? ChatGPT: Generative language models have several potential roles in medical education. Here are a few examples: - Generating realistic patient case scenarios: Language models can generate realistic patient case scenarios that medical students can use to practice their diagnostic and problem-solving skills. These scenarios can include patient history, symptoms, and laboratory results, allowing students to practice applying their knowledge to realistic situations. - Providing personalized learning experiences: Language models can analyze the performance of individual students and generate personalized learning materials that address their specific areas of weakness. For example, if a student struggles with a particular medical concept, the language model can generate additional resources or exercises to help them better understand it. - Enhancing medical textbooks: Language models can help enhance medical textbooks by generating additional explanations, examples, and visual aids. This can help make complex medical concepts more accessible to students and improve their overall understanding of the material. - Generating summaries of medical research: Language models can be trained to summarize medical research articles, making it easier for medical students to quickly understand the key findings and implications of a study. This can save students time and help them stay up-to-date with the latest research in their field. Overall, generative language models have the potential to play an important role in medical education by providing personalized learning experiences, generating realistic patient case scenarios, enhancing textbooks, and summarizing medical research. However, it's important to note that language models are not a substitute for hands-on clinical experience or the guidance of experienced medical professionals. GE: [To illustrate this,] please simulate a patient with undiagnosed diabetes. What symptoms would they present with? Simulate what they would say to the doctor. ChatGPT: Sure, I'd be happy to simulate a patient with undiagnosed diabetes and provide an example of what they might say to their doctor. Simulated patient: The patient is a 45-year-old woman who presents to her primary care physician with complaints of increased thirst, frequent urination, and fatigue. She reports feeling thirsty all the time, even after drinking large amounts of water. She also notes that she has been urinating more frequently than usual, including getting up multiple times during the night to use the bathroom. She reports feeling unusually tired and lethargic, even though she has been getting enough sleep. What the patient might say to their doctor: “I've been feeling really thirsty lately, like I can't get enough water. And I've been going to the bathroom all the time, like every hour. Even at night, I have to get up and go. I'm also just really tired all the time, even though I've been getting enough sleep. Do you think it could be something serious?” GE: What are the lab results? ChatGPT: As an AI language model, I don't have access to a specific patient's lab results. However, in a patient with undiagnosed diabetes, some common lab results that might be abnormal include: - Fasting plasma glucose (FPG) level: A normal FPG level is typically between 70-99 mg/dL. In a patient with undiagnosed diabetes, the FPG level may be elevated above 126 mg/dL. - Hemoglobin A1c (HbA1c) level: HbA1c is a measure of a person's average blood glucose levels over the past 2-3 months. A normal HbA1c level is typically less than 5.7%. In a patient with undiagnosed diabetes, the HbA1c level may be higher than 6.5%. - Oral glucose tolerance test (OGTT): This test involves drinking a glucose solution and measuring blood glucose levels

Read the complete text