文献速递乳腺乳腺X线

人工智能筛查乳腺X线摄影漏诊的浸润性乳腺癌

Invasive Breast Cancers Missed by AI Screening of Mammograms.
PubMed 文献 · Radiology · 2025-01-01

本回顾性研究纳入2014年1月至2020年12月连续诊断的1097个乳腺癌,评估商用人工智能软件在乳腺X线摄影上的假阴性率及其与分子亚型的关系。人工智能漏诊14%(154/1097),HER2富集型假阴性率最低(9%),漏诊癌多为可操作病灶(61.7%),漏诊原因包括致密乳腺、非乳腺区位置、结构扭曲和无定形微钙化。

要点

  • 人工智能在乳腺X线摄影上漏诊14%的浸润性乳腺癌,HER2富集型假阴性率最低(9%)。
  • 漏诊癌中61.7%为可操作病灶,阅片时需关注致密乳腺和非乳腺区位置。
  • 结构扭曲和无定形微钙化是人工智能漏诊的常见原因。

核心结论

  • 为减少人工智能在乳腺X线摄影上的漏诊,应关注管腔型癌、致密乳腺、非乳腺区位置、结构扭曲和无定形钙化。

研究设计与样本

  • 回顾性研究,纳入2014年1月至2020年12月连续诊断的1082例女性共1097个乳腺癌(平均年龄54.3岁±11),使用商用人工智能软件阅读乳腺X线摄影并获得异常评分,以参考标准定义人工智能漏诊。

关键影像表现/征象

  • 人工智能漏诊的癌与检出癌相比,与更年轻年龄、肿瘤大小≤2 cm、更低组织学分级、更少淋巴结转移、更多BI-RADS 4类表现、更低Ki-67表达及非乳腺区位置相关(均P < .05)。
  • 盲法阅片中将61.7%(95/154)的人工智能漏诊癌归类为可操作病灶。
  • 漏诊原因包括致密乳腺(n = 56)、非乳腺区位置(n = 22)、结构扭曲(n = 12)和无定形微钙化(n = 5)。

诊断标准与关键数值

  • 人工智能漏诊癌定义为人工智能未识别出与参考标准精确匹配位置的癌。
  • 总体假阴性率为14%(154/1097)。
  • HER2富集型假阴性率9%(36/398),管腔型17.2%(106/616),三阴性14.5%(12/83),P = .001。
  • 盲法阅片中61.7%(95/154)的人工智能漏诊癌为可操作病灶。

临床应用与阅片要点

  • 阅片时对管腔型癌、致密乳腺、非乳腺区位置、结构扭曲和无定形钙化应给予关注以减少人工智能漏诊。
  • 盲法阅片显示61.7%的人工智能漏诊癌为可操作病灶,提示部分漏诊癌在人工阅片时可能被发现。

局限性

  • 原文未报告局限性。

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

Background Little is known about the features of invasive breast cancers missed by artificial intelligence (AI) on mammograms. Purpose To assess the false-negative rate (FNR) of AI mammogram evaluation according to molecular subtype and to investigate the features of and reasons for AI-missed cancers. Materials and Methods This retrospective study identified consecutive patients diagnosed with breast cancer between January 2014 and December 2020. Commercial AI software was used to read the mammograms, and abnormality score (AS) was acquired. AI-missed cancers were defined as those for which AI did not identify a precise location matching the reference standard. The FNR was calculated by counting AI-missed cancers according to molecular subtype (hormone receptor-positive [luminal] vs human epidermal growth factor receptor 2 [HER2]-enriched vs triple-negative). Three blinded radiologists classified AI-missed cancers as either actionable or under threshold, and reasons for misses were determined through nonblinded reviews. Features were compared according to AI detection with the χ2 test. Results A total of 1082 consecutive women diagnosed with 1097 cancers (mean age, 54.3 years ± 11 [SD]) were included. AI missed 14% (154 of 1097) of cancers. The FNR was lowest in the HER2-enriched subtype (9% [36 of 398] in the HER2-enriched subtype, 17.2% [106 of 616] in the luminal subtype, and 14.5% [12 of 83] in the triple-negative subtype; P = .001). Compared with AI-detected cancers, AI-missed cancers were associated with younger age, a tumor size less than or equal to 2 cm, a lower histologic grade, fewer lymph node metastases, more Breast Imaging Reporting and Data System category 4 findings, lower Ki-67 expression, and nonmammary zone locations (all, P < .05). In blinded reviews, 61.7% (95 of 154) of AI-missed cancers were actionable; the reasons for misses were dense breasts (n = 56), nonmammary zone locations (n = 22), architectural distortions (n = 12), and amorphous microcalcifications (n = 5). Conclusion To reduce AI-missed cancers on mammograms, attention should be given to luminal cancer, dense breasts, nonmammary zone locations, architectural distortions, and amorphous calcifications. Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Mullen in this issue.

来源:Radiology,查看原文。版权归原作者所有。

引用

  1. Woo OH, Song SE, Choe SJ, et al. Invasive Breast Cancers Missed by AI Screening of Mammograms. Radiology. 2025. PMID: 40553002. doi:10.1148/radiol.242408

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。