Exploring the potential of Chat-GPT as a supportive tool for sialendoscopy clinical decision making and patient information support

Are artificial intelligence large language models a reliable tool for difficult differential diagnosis? An a posteriori analysis of a peculiar case of necrotizing otitis externa

“…18 Moreover, LLM represents an additional, user-friendlier tool that is readily available to any clinician and has revolutionized the perception of AI by the general public and specialists alike, with the latter recognizing the role of AI in providing simple and understandable information. 7 What is still not clear, however, and what we aimed to understand with this preliminary evaluation, is whether LLM can provide distinctive and useful insight into difficult clinical cases, by decontextualizing information and removing the physician's biases from the clinical evaluation. 19 Even more so, we are still uncertain about LLM's role in decision-making in the context of niche applications and with negligible internet presence, such as NOE.…”

“…However, it appears that the capabilities of AI in making a differential diagnosis do not exceed the physicians' expertise, and it has been shown that having an AI‐driven list of differential diagnoses does not meaningfully affect a physician's accuracy in making a diagnosis 18 . Moreover, LLM represents an additional, user‐friendlier tool that is readily available to any clinician and has revolutionized the perception of AI by the general public and specialists alike, with the latter recognizing the role of AI in providing simple and understandable information 7 …”

“…Despite its apparent ease of use and unlimited capabilities, this tool application has been only minimally explored, particularly in more niche settings such as otolaryngology. 7 The aim of this case report was therefore to employ this complex case of NOE to evaluate and present the capabilities of the aforementioned LLM in analyzing clinical data in an extremely niche setting. We submitted the case a posteriori to ChatGPT-3 and ChatGPT-4 in the form of objective-only data (clinical history and imaging reports) and prompted the AI to analyze the data and explore potential differential diagnoses, comparing its answers with the real-life clinical scenario.…”

“…Chat Generative Pre‐Trained Transformer (ChatGPT), an example of a large language model (LLM), is an easily accessible and user‐friendly AI tool that has gained recent attention due to its ability to textually interact with near‐human capability, due to the tool's extensive training on a wide range of texts available on the internet. Despite its apparent ease of use and unlimited capabilities, this tool application has been only minimally explored, particularly in more niche settings such as otolaryngology 7 …”

Validation of the QAMAI tool to assess the quality of health information provided by AI

“…These tools have primarily been designed for manual, human-centric assessments and are not compatible with AI-generated outputs. To date, no validated tool exists to accurately assess the health information provided by ChatGPT, and the few clinical studies published on this topic have used non-validated instruments [9,10,17,[24][25][26][27].…”