Page 37 - Read Online

Brenac et al. Art Int Surg 2024;4:296-315 https://dx.doi.org/10.20517/ais.2024.49 Page 304

Table 2. Description of primary research studies analyzed

Section Author Title Year Journal Main aim Main finding
[28]
AI-enhanced Wang et al. Artificial intelligence in plastic 2024 Plastic and (1) Evaluate the readability of 10 online sources of (1) ChatGPT generally improved readability, but only one
readability of patient surgery: ChatGPT as a tool to Reconstructive breast reconstruction information against those website showed statistically significant improvement
education address disparities in health literacy Surgery modified by ChatGPT (2) Readability scores still exceeded the 5th-grade level,
(2) Assess ChatGPT’s ability to improve the indicating a need for more simplification of patient
readability of patient-facing medical information information
[19]
Baldwin et al. An artificial intelligence language 2024 Burns (1) Assess the effectiveness of ChatGPT in (1) After ChatGPT modifications, 18% of webpages met
model improves readability of burn improving the readability of burn first aid an 11-year-old literacy level, compared to only 4% prior
first aid information information to ChatGPT modifications
(2) Evaluate the changes in the readability of (2) ChatGPT significantly improved readability scores
online patient education materials after ChatGPT across all readability formulas (P < 0.001)
modification
[17]
Browne et al. ChatGPT-4 can help hand surgeons 2024 Journal of Hand (1) Evaluate ChatGPT’s ability to improve the (1) ChatGPT significantly improved the readability of
communicate better with patients Surgery Global readability of hand surgery information provided hand surgery information (P < 0.001) across all
Online by the American Society of Surgery of Hand and readability tests
the British Society of Surgery of the Hand (2) ChatGPT-modified information achieved a mean
(2) Use multiple readability formulas to compare sixth-grade readability level on multiple readability tests
pre and post-ChatGPT modified readability scores
of hand surgery content
[20]
Evaluation of AI- Berry et al. Both patients and plastic surgeons 2024 Journal of (1) Compare the quality of ChatGPT-generated (1) Surgeons preferred ChatGPT’s responses 70.7% of
generated patient- prefer artificial intelligence - Reconstructive responses to microsurgery questions against those the time and rated ChatGPT higher in terms of
facing information generated microsurgical Microsurgery provided by the ASRM comprehensiveness (P < 0.001) and clarity (P < 0.05)
information (2) Assess the comprehensiveness, clarity, and (2) ChatGPT’s responses had significantly worse
readability of ChatGPT’s responses against ASRM readability scores across multiple readability formulas
content compared to ASRM content
Grippaudo Quality of the information provided 2024 JPRAS Open (1) Assess the quality of ChatGPT-generated (1) ChatGPT produced high-quality information for
[21]
et al. by ChatGPT for patients in breast breast plastic surgery information using the EQIP breast reconstruction (19/36), reduction (19/36), and
plastic surgery: are we already in scale augmentation (20/36), scoring above 18 on the EQIP
the future? (2) Evaluate ChatGPT’s generated patient-facing scale
breast surgery information across content, (2) ChatGPT information lacked identification since the
identification, and structure data information did not contain bibliography references or
proper validation
[22]
Seth et al. Evaluating chatbot efficacy for 2023 Aesthetic Surgery (1) Evaluate ChatGPT’s ability to create accurate, (1) ChatGPT provided comprehensive and grammatically
answering frequently asked Journal informative, and accessible breast augmentation accurate responses to breast augmentation questions
questions in plastic surgery: a information (2) ChatGPT-generated information lacked personalized
ChatGPT case study focused on (2) Qualitatively compare ChatGPT’s responses to advice
breast augmentation breast augmentation questions with established
literature
[23]
Xie et al. Aesthetic surgery advice and 2023 Aesthetic Plastic (1) Investigate ChatGPT’s ability to generate (1) ChatGPT provided comprehensive and coherent
counseling from artificial Surgery accurate, informative, and accessible responses to answers to rhinoplasty questions
intelligence: a rhinoplasty rhinoplasty questions sourced from the ASPS (2) ChatGPT lacked the ability to generate personalized
consultation with ChatGPT website advice, limiting its usefulness in patient consultations
(2) Evaluate the quality of ChatGPT’s responses in
a clinical setting