CoChat

Application of ChatGPT in Cosmetic Plastic Surgery: Ally or Antagonist?

Open access
Gupta, Rohun, Pande, Parna, Herzog, Isabel, Weisberger, Joseph, Chao, John, Chaiyasate, Kongkrit, Lee, Edward S · 2023 · Aesthetic Surgery Journal
Cited by 86

The DOI matches its Crossref record.

Abstract

From the development of life-saving devices to the introduction of electronic health records, technology and artificial intelligence (AI) have had a profound impact on medicine. More recently, OpenAI, a research laboratory based in San Francisco, CA, released a large language model, ChatGPT, to the general public on November 30, 2022. Unlike other large language models, ChatGPT has the ability to process and respond to commands in a humanistic fashion by acknowledging shortcomings and learning from previous mistakes.1 Plastic and reconstructive surgery is a field of innovation, relying on research to improve patient-centered outcomes. Cosmetic surgery, in particular, is an ever-evolving field, with over US$16.7 billion spent on procedures in the United States alone in 2020 according to statistics reported by The Aesthetic Society.2 Systematic reviews are important to the field of plastic surgery, as they serve to deliver a comprehensive depiction of available evidence on a specific topic in the field; furthermore, they help to identify knowledge gaps, address methodology concerns, and answer overarching research questions. Due to the novelty of ChatGPT and the importance of evidence-based research in cosmetic surgery, we wanted to determine whether or not ChatGPT could be utilized to aid in the development of unpublished systematic review topics. In order to incorporate both the broad and the specific aspects of cosmetic plastic surgery, ChatGPT was given 3 commands: “Give me 10 novel systematic review ideas that have not been published in cosmetic plastic surgery” (Table 1; Figure 1). “Give me 5 novel systematic review ideas that have not been published related to rhinoplasty” (Table 2). “Give me 5 novel systematic review ideas that have not been published related to blepharoplasty” (Table 3). Command given to ChatGPT and response provided by ChatGPT. Ten ChatGPT-generated Systematic Review Topics Within General Cosmetic Surgery, Corresponding Number of Nonsystematic and Systematic Reviews Already Published on Topic, and Novelty Status Five ChatGPT-generated Systematic Review Topics Involving Rhinoplasty, Corresponding Number of Nonsystematic and Systematic Reviews Already Published on Topic, and Novelty Status Five ChatGPT-generated Systematic Review Topics Involving Blepharoplasty, Corresponding Number of Nonsystematic and Systematic Reviews Already Published on Topic, and Novelty Status In total, the AI model would devise a total of 20 “unpublished” systematic review ideas within cosmetic surgery. In order to assess the accuracy of ChatGPT, we conducted a literature search in PubMed (National Institutes of Health; Bethesda, MD), CINAHL (EBSCO Information Services; Ipswich, MA), EMBASE (Elsevier; Amsterdam, the Netherlands), and Cochrane (Cochrane Library; London, UK) to account for general literature and the number of systematic reviews that had been published on each topic. Overall, we found that ChatGPT had a 65% accuracy in conceiving innovative systematic review ideas (Table 4). When stratified by general and specific topics within cosmetic surgery, we found that ChatGPT was 50% accurate for general cosmetic surgery and 80% accurate for specific topics (rhinoplasty and blepharoplasty). There were an average of 274.7 nonsystematic reviews and an average of 7.7 systematic reviews that were published for the group of topics that were found to have systematic reviews published. On the other hand, there were an average of 57.8 nonsystematic review articles that were published for the group of ideas that did not have any systematic reviews published. Number of Novel ChatGPT-generated Topics Within Cosmetic Surgery for Their Respective General Category and 2 Subcategories Our findings are not without limitations. Although ChatGPT has been trained on several major datasets, its last data upload was in 2021.3 As a result, there were some systematic reviews that had been published during 2022 and 2023 that were reported to be “novel” by ChatGPT. It is entirely possible that ChatGPT may not be able to present users with the most reliable and up-to-date information. Future updates or installations of ChatGPT may ameliorate this problem. Additionally, our sample size was small, consisting of only 20 “novel” systematic review ideas. However, we did find that the system responds more accurately with specific topics, which was demonstrated by an increase in accuracy from 50% to 80% when assessing general and specific topics within cosmetic plastic surgery. We believe this may be due to ChatGPT responding more accurately when specific questions or commands are given to the software. Further testing of the machine-learning capabilities of the ChatGPT software, in addition to giving more specific instructions, may aid in increasing the accuracy rates of ChatGPT. Although not perfect, ChatGPT has demonstrated itself to be a platform that can be utilized by clinicians to improve the quality of care that is provided to p

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/asj/sjad042

Cites

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

Cited by

Metadata

Authors
Gupta, R., Pande, P., Herzog, I., Weisberger, J., Chao, J., Chaiyasate, K., & Lee, E. S.
Venue
Aesthetic Surgery Journal
Published
2023-02-25
Volume
43
Issue
7
Pages
NP587-NP590
ISSN
  • 1090-820X (ISSN-L)
  • 1527-330X
OpenAlex
W4321850160
Cited by
86
References
0
FWCI
2.98
Type
journal article
Language
en
Open access
bronze

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.