For medical personnel, a novel artificial intelligence-based chatbot may be useful.
ChatGPT, a new artificial intelligence (AI)-based chatbot now being used in medical science, seems to be valuable for medical personnel in a general manner, as elucidated from a recent study published in "Diabetology & Metabolic Syndrome". In recent years, AI and deep learning have advanced. OpenAI Inc. recently released ChatGPT, a novel conversational chatbot that is quick and user-friendly owing to its dialogue style.
This study explored ChatGPT's current status and the degree to which it was accurate with respect to medical research and healthcare. Search in Google, Scopus, and PubMed databases were carried out with the aid of keywords "ChatGPT" AND "medical research, healthcare, and scientific writing." In English language, 29 hits were discovered in the PubMed database and 9 results in the Scopus database. Additionally, interactions with ChatGPT were done multiple times to assess the precision of answers to different medical inquiries.
Utilizing literature searches and ChatGPT interactions with medical questions, it was deduced that this version creates replies quickly. However, it recounts information from existing internet literature in a usual way. However, as the firm noted on the ChatGPT main page, mistakes were noted in the medical questions and answers. Additionally, the narrated data was only valid through September 2021.
Positive characteristics encompass the acknowledgement of its limitations in the medical area and the fact that it is meant to learn from prior responses. As a narrative AI chatbot for medical staff, ChatGPT's current version may be beneficial in a limited way, however, researchers are urged to double-check all claims made while bearing in mind its limitations.
Diabetology & Metabolic Syndrome
ChatGPT: Is this version good for healthcare and research?
Raju Vaishya et al.
Comments (0)