Evaluating ChatGPT's Role in Nuclear Medicine: Reliability, Challenges, and Ethical Considerations

Recent advancements in large language models (LLMs) like OpenAI's ChatGPT have sparked both excitement and concerns within various academic communities, including nuclear medicine. While ChatGPT's ability to generate grammatically fluent and convincing content has raised questions about its potential in academic writing and research, its reliability in the specialized field of nuclear medicine remains uncertain. This editorial explores ChatGPT's performance in simulating nuclear medicine content, including its ability to answer multiple-choice questions from nuclear medicine examinations. Preliminary testing revealed that ChatGPT answered only ۳۴% of questions correctly, raising concerns about its accuracy and potential for confabulation. Despite its promising capabilities in generating abstracts and summarizing content, its limitations in producing reliable, expert-level medical knowledge are evident. The article calls for caution in using AI tools like ChatGPT for clinical and academic purposes, emphasizing the need for rigorous validation and ethical considerations as these technologies continue to evolve. We argue that while ChatGPT holds promise as an educational aid, it currently does not meet the standards required for high-stakes medical examinations. Consequently, the role of AI in medical training and practice must be carefully considered to ensure patient safety and the integrity of the medical profession.

کلیدواژه ها:

ChatGPT ، nuclear medicine ، artificial intelligence ، medical ethics ، academic integrity ، medical education ، AI reliability ، medical examinations ، AI in healthcare