ChatGPT’s ability to classify virtual reality studies in cardiology

Nakaya, Yuichiro; Higaki, Akinori; Yamaguchi, Osamu

doi:10.1093/ehjdh/ztad026

Cited by 9 publications

(5 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Section: Discussionsupporting

confidence: 71%

“…Our work extends the ndings of (Nakaya et al, 2023), and shows that even in tasks that require textual interpretation, an AI such as ChatGPT can provide a reliable classi cation of relevant articles across numerous screening criteria. Unlike (Nakaya et al, 2023) however, we show that, for more complex screening criteria, relying on a single agent can provide inconsistent results. Multiple repeat calls to the AI, supported by the random context string described in Section 2.3, can provide a greater level of reliability by reducing the chances of omitting relevant articles, albeit at the expense of returning a greater number of false positives to be screened at a subsequent stage.…”

Section: Discussionsupporting

confidence: 71%

See 1 more Smart Citation

Human-AI Collaboration to Identify Literature for Evidence Synthesis

Spillias

Tuohy

Andreotta

et al. 2023

Preprint

View full text Add to dashboard Cite

Systematic approaches to evidence synthesis can improve the rigour, transparency, and replicability of a traditional literature review. However, these systematic approaches are time and resource intensive. We evaluate the ability of OpenAI’s ChatGPT to undertake two initial stages of evidence syntheses (searching peer-reviewed literature and screening for relevance) and develop a novel collaborative framework to leverage the best of both human and AI intelligence. Using a scoping review of community-based fisheries management as a case study, we find that with substantial prompting, the AI can provide critical insight into the construction and content of a search string. Thereafter, we evaluate five strategies for synthesising AI output to screen articles based on predefined inclusion criteria. We find low omission rates (< 1%) of relevant literature by the AI are achievable, which is comparable to that of human screeners. These findings show that generalised AI tools can assist reviewers with evidence synthesis to accelerate the implementation and improve the reliability of a review.

show abstract

Section: Discussionsupporting

confidence: 71%

Section: Discussionsupporting

confidence: 71%

Human-AI Collaboration to Identify Literature for Evidence Synthesis

Spillias

Tuohy

Andreotta

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The development of AI language models has substantially advanced AI and has the potential to disrupt current clinical practices in all surgical and clinical specialties of medicine. These technologies' ethical and social implications must also be ( Nakaya et al, 2023) 2. Medical imaging Different case of error correction in medical radiation science of higher education.…”

Section: Potential Ethical Issues Associated With Chatgpt In Medicinementioning

confidence: 99%

Overview of Chatbots with special emphasis on artificial intelligence-enabled ChatGPT in medical science

Chakraborty,

Pal,

Bhattacharya

et al. 2023

Front. Artif. Intell.

View full text Add to dashboard Cite

The release of ChatGPT has initiated new thinking about AI-based Chatbot and its application and has drawn huge public attention worldwide. Researchers and doctors have started thinking about the promise and application of AI-related large language models in medicine during the past few months. Here, the comprehensive review highlighted the overview of Chatbot and ChatGPT and their current role in medicine. Firstly, the general idea of Chatbots, their evolution, architecture, and medical use are discussed. Secondly, ChatGPT is discussed with special emphasis of its application in medicine, architecture and training methods, medical diagnosis and treatment, research ethical issues, and a comparison of ChatGPT with other NLP models are illustrated. The article also discussed the limitations and prospects of ChatGPT. In the future, these large language models and ChatGPT will have immense promise in healthcare. However, more research is needed in this direction.

show abstract

“… 3 Chat Generative Pre-trained Transformer (ChatGPT) (Open AI, USA) can be used as a tool for conducting bibliometric analysis. 4 …”

Section: Introductionmentioning

confidence: 99%