乳腺影像人工智能相关文章的作者性别分布
该研究于2022年7月按PRISMA指南检索所有公开的AI与乳腺影像交叉的同行评议文章,纳入115项研究,通过邮件自我声明调查确定第一作者和资深作者性别。女性第一作者占35.7%(41/115)、女性资深作者占37.4%(43/115),逻辑回归显示女性资深作者随时间显著增加而女性第一作者无变化。作者认为乳腺影像AI研究存在显著作者性别差距,需进一步研究差距原因及潜在影响。
要点
- 乳腺影像AI相关文章中女性第一作者占35.7%(41/115)、女性资深作者占37.4%(43/115)。
- 女性资深作者比例随时间显著增加,女性第一作者比例无显著变化。
- 期刊影响因子与女性作者身份无关联,部分国家女性作者比例超过50%。
核心结论
- 乳腺影像人工智能研究存在显著的作者性别差距。
- 乳腺影像AI相关研究中资深作者女性比例随时间呈上升趋势,作者认为这对该领域更多女性声音是积极信号。
- 作者指出需进一步研究理解这一差距背后的原因及任何潜在影响。
研究设计与样本
- 研究于2022年7月按PRISMA指南进行策略性检索,纳入所有现有且公开的同行评议AI与乳腺影像交叉文章,共115项研究。
- 主要结局为第一作者和资深作者性别,通过邮件自我声明调查辅助判定。
- 次要结局包括文章所属国家、期刊影响因子和发表年份,比较采用逻辑回归模型和方差分析。
诊断标准与关键数值
- 女性第一作者占35.7%(41/115),女性资深作者占37.4%(43/115)。
- 逻辑回归模型显示女性资深作者随时间显著增加,女性第一作者无变化。
- 影响因子与女性作者身份无关联,部分国家女性作者比例超过50%。
临床应用与阅片要点
- 原文未据此给出临床推荐。
局限性
- 需进一步研究以理解这一差距背后的原因及任何潜在影响。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
RATIONALE AND OBJECTIVES: The purpose of this study is to investigate the variance of women authors, specifically first and senior authorship among peer-reviewed artificial intelligence-related articles with a specific focus in breast imaging.
MATERIALS AND METHODS: A strategic search was conducted in July 2022 according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to capture all existing and publicly available peer-reviewed articles intersecting AI and breast imaging. Primary outcomes were first and senior authors' gender, which were assigned with the aid of an emailed self-declaration survey. Secondary outcomes included country of article, journal impact factor, and year of publication. Comparisons were made using logistic regression models and analysis of variances.
RESULTS: 115 studies were included in the analysis. Women authors represented 35.7% (41/115) and 37.4% (43/115) of first and senior authors, respectively. Logistic regression modelling showed a significant increase in women senior authors over time but no changes in women first authors. Impact factor was not associated with female authorship and certain countries had women authorship reach over 50%.
CONCLUSION: This study demonstrates that there is a significant authorship gender gap in artificial intelligence breast imaging research. An increasing temporal trend of senior authors in breast imaging AI-related research is a promising prognosis for more women voices in this field. Further study needs to be done to understand the reasons behind this gap and any potential implications.
引用
- Shawn Yuan PH, Yan TD, Sharma S, et al. Authorship gender among articles about artificial intelligence in breast imaging. Eur J Radiol. 2024. PMID: 38492508. doi:10.1016/j.ejrad.2024.111428
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。