USC Institute of Urology and Catherine & Joseph Aresty Department of Urology, University of Southern California, Los Angeles, California.
Artificial Intelligence Center (UroAI) at USC Institute of Urology, University of Southern California, Los Angeles, California.
J Urol. 2024 Dec;212(6):873-881. doi: 10.1097/JU.0000000000004199. Epub 2024 Aug 14.
This cross-sectional study assessed a generative artificial intelligence platform to automate the creation of accurate, appropriate, and compelling social media (SoMe) posts from urological journal articles.
One hundred SoMe posts from the top 3 journals in urology X (formerly Twitter) profiles were collected from August 2022 to October 2023. A freeware generative pre-trained transformer (GPT) tool was developed to autogenerate SoMe posts, which included title summarization, key findings, pertinent emojis, hashtags, and digital object identifier links to the article. Three physicians independently evaluated GPT-generated posts for achieving tetrafecta of accuracy and appropriateness criteria. Fifteen scenarios were created from 5 randomly selected posts from each journal. Each scenario contained both the original and the GPT-generated post for the same article. Five questions were formulated to investigate the posts' likability, shareability, engagement, understandability, and comprehensiveness. The paired posts were then randomized and presented to blinded academic authors and general public through Amazon Mechanical Turk (AMT) responders for preference evaluation.
Median time for post autogeneration was 10.2 seconds (interquartile range 8.5-12.5). Of the 150 rated GPT-generated posts, 115 (76.6%) met the correctness tetrafecta: 144 (96%) accurately summarized the title, 147 (98%) accurately presented the articles' main findings, 131 (87.3%) appropriately used emojis, and 138 (92%) appropriately used hashtags. A total of 258 academic urologists and 493 AMT responders answered the surveys, wherein the GPT-generated posts consistently outperformed the original journals' posts for both academicians and AMT responders ( < .05).
Generative artificial intelligence can automate the creation of SoMe posts from urology journal abstracts that are both accurate and preferable by the academic community and general public.
本横断面研究评估了一种生成式人工智能平台，以从泌尿科期刊文章中自动创建准确、恰当且引人入胜的社交媒体 (SoMe) 帖子。
从 2022 年 8 月至 2023 年 10 月，从泌尿科 X（前 Twitter）档案的前 3 本期刊中收集了 100 篇 SoMe 帖子。开发了一个免费的生成式预训练转换器 (GPT) 工具来自动生成 SoMe 帖子，其中包括标题总结、主要发现、相关表情符号、话题标签和文章的数字对象标识符链接。三位医生独立评估了 GPT 生成的帖子是否符合准确性和恰当性标准的四件套。从每个期刊中随机选择 5 个帖子创建了 15 个场景。每个场景都包含同一篇文章的原始帖子和 GPT 生成的帖子。提出了 5 个问题来调查帖子的受欢迎程度、可分享性、参与度、理解度和全面性。然后对配对帖子进行随机化，并通过亚马逊机械土耳其人 (AMT) 响应者向盲目的学术作者和公众展示，以进行偏好评估。
帖子自动生成的中位数时间为 10.2 秒（四分位距 8.5-12.5）。在 150 个评价的 GPT 生成的帖子中，有 115 个（76.6%）符合正确性四件套标准：144 个（96%）准确地总结了标题，147 个（98%）准确地呈现了文章的主要发现，131 个（87.3%）恰当地使用了表情符号，138 个（92%）恰当地使用了话题标签。共有 258 名学术泌尿科医生和 493 名 AMT 响应者回答了调查，其中 GPT 生成的帖子在学术社区和普通公众中都比原始期刊的帖子表现更好（<0.05）。
生成式人工智能可以从泌尿科期刊摘要中自动创建准确且受学术界和普通公众欢迎的社交媒体帖子。
