ChatGPT与谷歌网络搜索在解剖全肩关节置换术和反向全肩关节置换术中的应用:从患者的角度

Q4 Medicine
Alexander K. Hahn MD , Gregory M. Connors BS , Martinus Megalla MD , Zachary T. Grace MD , Brett J. Croen MD , Michael R. Mancini MD , Matthew R. LeVasseur MD , Corey R. Dwyer MD , Katherine J. Coyner MD, MBA
{"title":"ChatGPT与谷歌网络搜索在解剖全肩关节置换术和反向全肩关节置换术中的应用:从患者的角度","authors":"Alexander K. Hahn MD ,&nbsp;Gregory M. Connors BS ,&nbsp;Martinus Megalla MD ,&nbsp;Zachary T. Grace MD ,&nbsp;Brett J. Croen MD ,&nbsp;Michael R. Mancini MD ,&nbsp;Matthew R. LeVasseur MD ,&nbsp;Corey R. Dwyer MD ,&nbsp;Katherine J. Coyner MD, MBA","doi":"10.1053/j.sart.2024.11.005","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><div>Artificial intelligence, which encompasses machine learning and natural language processing has made strides in recent years. Among these models, ChatGPT has gained attention for its abilities to allow for interactive responses. Given the rising popularity of ChatGPT, and its ability to understand medical complexity, this study aimed to evaluate ChatGPT’s utility in answering medical questions vs. the most widely used search engine in the United States, Google. The goal of this study is to replicate a patient’s web search to assess the utility of ChatGPT vs. Google regarding the most popular questions asked by patients undergoing a total shoulder replacement and reverse total shoulder replacement.</div></div><div><h3>Methods</h3><div>A Google web search was performed with the following search terms: “total shoulder replacement” and “reverse total shoulder replacement” and the top 10 frequently asked questions (FAQs) for each term were recorded along with the source. ChatGPT was then asked for the top 10 FAQs searched for total shoulder replacement and reverse total shoulder replacement and responses recorded. A Google web search was then completed using the same search terms and the top 10 FAQs, which included a numerical response were recorded. These questions were then placed into ChatGPT. Both the Google web search and ChatGPT responses to the questions were recorded, along with the source that provided the answer.</div></div><div><h3>Results</h3><div>Eight of the 20 questions were similar when performing a Google Web search and ChatGPT for the search terms “total shoulder replacement.” Ten of the 20 questions were similar when using the search term “reverse shoulder replacement”. The most common subcategory was “evaluation of surgery” for Google (6 of 20) and “technical” for ChatGPT (5 of 20). The most common sources for answers to the top 10 FAQs for Google for both total shoulder arthroplasty and reverse total shoulder arthroplasty were academic sources (9 of 20).</div></div><div><h3>Conclusion</h3><div>Google and ChatGPT users ask similar types of questions and reveal similar answers. ChatGPT typically offers more detailed answers but is less transparent at providing its source and generates improper sources up to 47% of the time. However, for properly cited references, ChatGPT typically uses credible government or academic sources compared to Google. Since Google is currently more familiar for its users than ChatGPT, it may be easier to use for patients wanting fast, simple answers to their questions, while ChatGPT may be superior for patients who may have multiple questions that want longer answers.</div></div>","PeriodicalId":39885,"journal":{"name":"Seminars in Arthroplasty","volume":"35 2","pages":"Pages 158-165"},"PeriodicalIF":0.0000,"publicationDate":"2025-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Use of ChatGPT vs. Google web search in anatomic total shoulder arthroplasty and reverse total shoulder arthroplasty: from a patient’s perspective\",\"authors\":\"Alexander K. Hahn MD ,&nbsp;Gregory M. Connors BS ,&nbsp;Martinus Megalla MD ,&nbsp;Zachary T. Grace MD ,&nbsp;Brett J. Croen MD ,&nbsp;Michael R. Mancini MD ,&nbsp;Matthew R. LeVasseur MD ,&nbsp;Corey R. Dwyer MD ,&nbsp;Katherine J. Coyner MD, MBA\",\"doi\":\"10.1053/j.sart.2024.11.005\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><h3>Background</h3><div>Artificial intelligence, which encompasses machine learning and natural language processing has made strides in recent years. Among these models, ChatGPT has gained attention for its abilities to allow for interactive responses. Given the rising popularity of ChatGPT, and its ability to understand medical complexity, this study aimed to evaluate ChatGPT’s utility in answering medical questions vs. the most widely used search engine in the United States, Google. The goal of this study is to replicate a patient’s web search to assess the utility of ChatGPT vs. Google regarding the most popular questions asked by patients undergoing a total shoulder replacement and reverse total shoulder replacement.</div></div><div><h3>Methods</h3><div>A Google web search was performed with the following search terms: “total shoulder replacement” and “reverse total shoulder replacement” and the top 10 frequently asked questions (FAQs) for each term were recorded along with the source. ChatGPT was then asked for the top 10 FAQs searched for total shoulder replacement and reverse total shoulder replacement and responses recorded. A Google web search was then completed using the same search terms and the top 10 FAQs, which included a numerical response were recorded. These questions were then placed into ChatGPT. Both the Google web search and ChatGPT responses to the questions were recorded, along with the source that provided the answer.</div></div><div><h3>Results</h3><div>Eight of the 20 questions were similar when performing a Google Web search and ChatGPT for the search terms “total shoulder replacement.” Ten of the 20 questions were similar when using the search term “reverse shoulder replacement”. The most common subcategory was “evaluation of surgery” for Google (6 of 20) and “technical” for ChatGPT (5 of 20). The most common sources for answers to the top 10 FAQs for Google for both total shoulder arthroplasty and reverse total shoulder arthroplasty were academic sources (9 of 20).</div></div><div><h3>Conclusion</h3><div>Google and ChatGPT users ask similar types of questions and reveal similar answers. ChatGPT typically offers more detailed answers but is less transparent at providing its source and generates improper sources up to 47% of the time. However, for properly cited references, ChatGPT typically uses credible government or academic sources compared to Google. Since Google is currently more familiar for its users than ChatGPT, it may be easier to use for patients wanting fast, simple answers to their questions, while ChatGPT may be superior for patients who may have multiple questions that want longer answers.</div></div>\",\"PeriodicalId\":39885,\"journal\":{\"name\":\"Seminars in Arthroplasty\",\"volume\":\"35 2\",\"pages\":\"Pages 158-165\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2025-01-03\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Seminars in Arthroplasty\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1045452724001159\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"Medicine\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Seminars in Arthroplasty","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1045452724001159","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Medicine","Score":null,"Total":0}
引用次数: 0

摘要

包括机器学习和自然语言处理在内的人工智能近年来取得了长足的进步。在这些模型中,ChatGPT因其允许交互响应的能力而受到关注。鉴于ChatGPT的日益普及,以及它理解医疗复杂性的能力,本研究旨在评估ChatGPT在回答医疗问题方面的效用,与美国使用最广泛的搜索引擎b谷歌相比。本研究的目的是复制患者的网络搜索,以评估ChatGPT与谷歌在接受全肩关节置换术和反向全肩关节置换术患者最常见的问题方面的效用。方法采用b谷歌网络检索“全肩关节置换术”和“反向全肩关节置换术”,并记录各检索词的前10个常见问题(FAQs)及其来源。然后询问ChatGPT关于全肩关节置换术和反向全肩关节置换术的十大常见问题,并记录其回答。然后使用相同的搜索词完成b谷歌网络搜索,并记录前10个常见问题,其中包括数字回答。然后将这些问题放入ChatGPT中。b谷歌网络搜索和ChatGPT对问题的回答都被记录下来,以及提供答案的来源。结果在b谷歌网络搜索和ChatGPT搜索“全肩关节置换术”时,20个问题中有8个相似。在使用“反向肩关节置换术”这一搜索词时,20个问题中有10个是相似的。最常见的子分类是谷歌的“手术评估”(6 / 20)和ChatGPT的“技术评估”(5 / 20)。对于全肩关节置换术和反向全肩关节置换术的谷歌前10个常见问题的答案,最常见的来源是学术来源(20个中有9个)。结论:google和ChatGPT用户提出的问题类型相似,给出的答案也相似。ChatGPT通常会提供更详细的答案,但在提供来源方面不太透明,并且在47%的情况下会产生不正确的来源。然而,对于正确引用的参考文献,ChatGPT通常使用可信的政府或学术来源,而不是谷歌。由于谷歌目前比ChatGPT更为用户所熟悉,对于想要快速、简单的问题答案的患者来说,谷歌可能更容易使用,而ChatGPT可能更适合那些有多个问题、想要更长的答案的患者。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Use of ChatGPT vs. Google web search in anatomic total shoulder arthroplasty and reverse total shoulder arthroplasty: from a patient’s perspective

Background

Artificial intelligence, which encompasses machine learning and natural language processing has made strides in recent years. Among these models, ChatGPT has gained attention for its abilities to allow for interactive responses. Given the rising popularity of ChatGPT, and its ability to understand medical complexity, this study aimed to evaluate ChatGPT’s utility in answering medical questions vs. the most widely used search engine in the United States, Google. The goal of this study is to replicate a patient’s web search to assess the utility of ChatGPT vs. Google regarding the most popular questions asked by patients undergoing a total shoulder replacement and reverse total shoulder replacement.

Methods

A Google web search was performed with the following search terms: “total shoulder replacement” and “reverse total shoulder replacement” and the top 10 frequently asked questions (FAQs) for each term were recorded along with the source. ChatGPT was then asked for the top 10 FAQs searched for total shoulder replacement and reverse total shoulder replacement and responses recorded. A Google web search was then completed using the same search terms and the top 10 FAQs, which included a numerical response were recorded. These questions were then placed into ChatGPT. Both the Google web search and ChatGPT responses to the questions were recorded, along with the source that provided the answer.

Results

Eight of the 20 questions were similar when performing a Google Web search and ChatGPT for the search terms “total shoulder replacement.” Ten of the 20 questions were similar when using the search term “reverse shoulder replacement”. The most common subcategory was “evaluation of surgery” for Google (6 of 20) and “technical” for ChatGPT (5 of 20). The most common sources for answers to the top 10 FAQs for Google for both total shoulder arthroplasty and reverse total shoulder arthroplasty were academic sources (9 of 20).

Conclusion

Google and ChatGPT users ask similar types of questions and reveal similar answers. ChatGPT typically offers more detailed answers but is less transparent at providing its source and generates improper sources up to 47% of the time. However, for properly cited references, ChatGPT typically uses credible government or academic sources compared to Google. Since Google is currently more familiar for its users than ChatGPT, it may be easier to use for patients wanting fast, simple answers to their questions, while ChatGPT may be superior for patients who may have multiple questions that want longer answers.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Seminars in Arthroplasty
Seminars in Arthroplasty Medicine-Surgery
CiteScore
1.00
自引率
0.00%
发文量
104
期刊介绍: Each issue of Seminars in Arthroplasty provides a comprehensive, current overview of a single topic in arthroplasty. The journal addresses orthopedic surgeons, providing authoritative reviews with emphasis on new developments relevant to their practice.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信