CoChat

Artificial intelligence in orthopaedics: can Chat Generative Pre-trained Transformer (ChatGPT) pass Section 1 of the Fellowship of the Royal College of Surgeons (Trauma & Orthopaedics) examination?

Cuthbert, Rory, Simpson, Ashley I · 2023 · Postgraduate Medical Journal
Cited by 56

The DOI matches its Crossref record; 7 of the 7 references it cites were found in a registry or an index.

Abstract

Purpose: Chat Generative Pre-trained Transformer (ChatGPT) is a large language artificial intelligence (AI) model which generates contextually relevant text in response to questioning. After ChatGPT successfully passed the United States Medical Licensing Examinations, proponents have argued it should play an increasing role in medical service provision and education. AI in healthcare remains in its infancy, and the reliability of AI systems must be scrutinized. This study assessed whether ChatGPT could pass Section 1 of the Fellowship of the Royal College of Surgeons (FRCS) examination in Trauma and Orthopaedic Surgery. Methods: The UK and Ireland In-Training Examination (UKITE) was used as a surrogate for the FRCS. Papers 1 and 2 of UKITE 2022 were directly inputted into ChatGPT. All questions were in a single-best-answer format without wording alterations. Imaging was trialled to ensure ChatGPT utilized this information. Results: ChatGPT scored 35.8%: 30% lower than the FRCS pass rate and 8.2% lower than the mean score achieved by human candidates of all training levels. Subspecialty analysis demonstrated ChatGPT scored highest in basic science (53.3%) and lowest in trauma (0%). In 87 questions answered incorrectly, ChatGPT only stated it did not know the answer once and gave incorrect explanatory answers for the remaining questions. Conclusion: ChatGPT is currently unable to exert the higher-order judgement and multilogical thinking required to pass the FRCS examination. Further, the current model fails to recognize its own limitations. ChatGPT’s deficiencies should be publicized equally as much as its successes to ensure clinicians remain aware of its fallibility. Key messages: What is already known on this topic: Following ChatGPT’s much-publicized success in passing the United States Medical Licensing Examinations, clinicians and medical students are using the model increasingly frequently for medical service provision and education. However ChatGPT remains in its infancy, and the model’s reliability and accuracy remain unproven. What this study adds: This study demonstrates ChatGPT is currently unable to exert the higher-order judgement and multilogical thinking required to pass the Fellowship of the Royal College of Surgeons (FRCS) (Trauma & Orthopaedics) examination. Further, the current model fails to recognize its own limitations when offering both direct and explanatory answers. How this study might affect research, practice, or policy: This study highlights the need for medical students and clinicians to exert caution when employing ChatGPT as a revision tool or applying it in clinical practice, and for patients to be aware of its fallibilities when using it as a health resource. Future research questions include:

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1093/postmj/qgad053

Cites

7 of 7 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Cited by

See all 56

Metadata

Authors
Cuthbert, R., & Simpson, A. I.
Venue
Postgraduate Medical Journal
Published
2023-07-06
Volume
99
Issue
1176
Pages
1110-1114
ISSN
  • 0032-5473 (ISSN-L)
  • 1469-0756
OpenAlex
W4383371705
Cited by
56
References
7
FWCI
1.94
Type
journal article
Language
en
Open access
closed

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.