Generative Artificial Intelligence Biases, Limitations and Risks in Nuclear Medicine: An Argument for Appropriate Use Framework and Recommendations.

IF 4.6 2区 医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING Seminars in nuclear medicine Pub Date : 2024-06-07 DOI:10.1053/j.semnuclmed.2024.05.005
Geoffrey M Currie, K Elizabeth Hawk, Eric M Rohren
{"title":"Generative Artificial Intelligence Biases, Limitations and Risks in Nuclear Medicine: An Argument for Appropriate Use Framework and Recommendations.","authors":"Geoffrey M Currie, K Elizabeth Hawk, Eric M Rohren","doi":"10.1053/j.semnuclmed.2024.05.005","DOIUrl":null,"url":null,"abstract":"<p><p>Generative artificial intelligence (AI) algorithms for both text-to-text and text-to-image applications have seen rapid and widespread adoption in the general and medical communities. While limitations of generative AI have been widely reported, there remain valuable applications in patient and professional communities. Here, the limitations and biases of both text-to-text and text-to-image generative AI are explored using purported applications in medical imaging as case examples. A direct comparison of the capabilities of four common text-to-image generative AI algorithms is reported and recommendations for the most appropriate use, DALL-E 3, justified. The risks use and biases are outlined, and appropriate use guidelines framed for use of generative AI in nuclear medicine. Generative AI text-to-text and text-to-image generation includes inherent biases, particularly gender and ethnicity, that could misrepresent nuclear medicine. The assimilation of generative AI tools into medical education, image interpretation, patient education, health promotion and marketing in nuclear medicine risks propagating errors and amplification of biases. Mitigation strategies should reside inside appropriate use criteria and minimum standards for quality and professionalism for the application of generative AI in nuclear medicine.</p>","PeriodicalId":21643,"journal":{"name":"Seminars in nuclear medicine","volume":null,"pages":null},"PeriodicalIF":4.6000,"publicationDate":"2024-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Seminars in nuclear medicine","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1053/j.semnuclmed.2024.05.005","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}
引用次数: 0

Abstract

Generative artificial intelligence (AI) algorithms for both text-to-text and text-to-image applications have seen rapid and widespread adoption in the general and medical communities. While limitations of generative AI have been widely reported, there remain valuable applications in patient and professional communities. Here, the limitations and biases of both text-to-text and text-to-image generative AI are explored using purported applications in medical imaging as case examples. A direct comparison of the capabilities of four common text-to-image generative AI algorithms is reported and recommendations for the most appropriate use, DALL-E 3, justified. The risks use and biases are outlined, and appropriate use guidelines framed for use of generative AI in nuclear medicine. Generative AI text-to-text and text-to-image generation includes inherent biases, particularly gender and ethnicity, that could misrepresent nuclear medicine. The assimilation of generative AI tools into medical education, image interpretation, patient education, health promotion and marketing in nuclear medicine risks propagating errors and amplification of biases. Mitigation strategies should reside inside appropriate use criteria and minimum standards for quality and professionalism for the application of generative AI in nuclear medicine.

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
生成式人工智能在核医学中的偏差、局限性和风险:核医学中的偏见、局限性和风险:适当使用框架和建议论证》。
用于文本到文本和文本到图像应用的生成式人工智能(AI)算法在普通和医疗界得到了迅速而广泛的应用。虽然生成式人工智能的局限性已被广泛报道,但在患者和专业团体中仍有宝贵的应用价值。在此,我们以医学影像领域的所谓应用为例,探讨了文本到文本和文本到图像生成式人工智能的局限性和偏差。报告对四种常见的文本到图像生成式人工智能算法的能力进行了直接比较,并对最合适的使用建议 DALL-E 3 进行了论证。概述了使用风险和偏差,并为在核医学中使用生成式人工智能制定了适当的使用指南。人工智能文本到文本和文本到图像的生成包含固有的偏见,尤其是性别和种族偏见,可能会误导核医学。将人工智能生成工具融入核医学的医学教育、图像解读、患者教育、健康宣传和市场营销中,有可能传播错误和扩大偏见。缓解策略应包含适当的使用标准,以及核医学中应用生成式人工智能的质量和专业性最低标准。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Seminars in nuclear medicine
Seminars in nuclear medicine 医学-核医学
CiteScore
9.80
自引率
6.10%
发文量
86
审稿时长
14 days
期刊介绍: Seminars in Nuclear Medicine is the leading review journal in nuclear medicine. Each issue brings you expert reviews and commentary on a single topic as selected by the Editors. The journal contains extensive coverage of the field of nuclear medicine, including PET, SPECT, and other molecular imaging studies, and related imaging studies. Full-color illustrations are used throughout to highlight important findings. Seminars is included in PubMed/Medline, Thomson/ISI, and other major scientific indexes.
期刊最新文献
Letter From the Editors. Clinical Explorations of [68Ga] Ga-FAPI-04 and [18F] FDG Dual-Tracer Total-body PET/CT and PET/MR Imaging. Emerging TSPO-PET Radiotracers for Imaging Neuroinflammation: A Critical Analysis. Highlighting New Research Trends on Zirconium-89 Radiopharmaceuticals Beyond Antibodies. Update on PSMA-based Prostate Cancer Imaging.
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1