文献速递乳腺US

人工智能决策支持系统对不同经验水平放射科医生乳腺超声诊断的影响:一项三级中心的前瞻性研究

Impact of an artificial intelligence decision support system among radiologists with different levels of experience in breast ultrasound: A prospective study in a tertiary center.
PubMed 文献 · Eur J Radiol · 2025-02-25

这项前瞻性观察性单中心研究纳入222例患者共226个可疑乳腺病灶,比较三名不同经验放射科医生在应用Koios DS前后BI-RADS分类的诊断效能。应用Koios DS后所有医生AUC显著提高(p<0.001),低年资乳腺放射科医生AUC提高0.078,普通放射科医生提高0.062,高年资乳腺放射科医生提高0.045。高年资和低年资乳腺放射科医生推荐的不必要活检数量显著减少。

要点

  • Koios DS可提高不同经验放射科医生乳腺超声诊断的AUC,低年资者提升幅度更大。
  • 应用Koios DS后高年资和低年资乳腺放射科医生不必要活检比例显著降低。
  • 普通放射科医生不必要活检变化未达统计学显著。

核心结论

  • 应用Koios DS提高了放射科医生的诊断效能,尤其是经验较少者,并可能减少不必要活检。

研究设计与样本

  • 前瞻性观察性单中心研究,纳入连续222例患者共226个可疑乳腺病灶,以超声引导下空心针活检病理为参考标准。
  • 三名放射科医生(高年资乳腺放射科医生20年、低年资乳腺放射科医生3年、普通放射科医生不足1年)独立评估病灶,在应用Koios DS前后分别赋予BI-RADS分类。
  • 采用DeLong检验和McNemar检验比较应用Koios DS前后的AUC和不必要活检数量。

诊断标准与关键数值

  • 226个病灶中良性89/226(39.4%),恶性137/226(60.6%)。
  • 应用Koios DS后所有放射科医生AUC显著提高(p<0.001):低年资乳腺放射科医生从0.786升至0.864(ΔAUC 0.078),普通放射科医生从0.719升至0.781(ΔAUC 0.062),高年资乳腺放射科医生从0.823升至0.868(ΔAUC 0.045)。
  • 高年资乳腺放射科医生推荐的不必要活检从41/89(46.1%)降至30/89(33.7%),p<0.001。
  • 低年资乳腺放射科医生推荐的不必要活检从46/89(51.7%)降至29/89(32.6%),p=0.001。

临床应用与阅片要点

  • Koios DS可作为乳腺超声BI-RADS分类的辅助决策工具,尤其有助于经验较少的放射科医生提高诊断效能。
  • 应用Koios DS可能减少高年资和低年资乳腺放射科医生推荐的不必要活检,但普通放射科医生的不必要活检变化未达统计学显著。

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

PURPOSE: To assess the impact of an artificial intelligence decision support system (Koios DS) on the diagnostic performance of radiologists with different experience in breast ultrasound and to evaluate its potential to reduce unnecessary biopsies.

METHODS: This observational, prospective, single-centre study included consecutive patients scheduled for ultrasound-guided core-needle biopsy of suspicious breast lesions. Three radiologists with different experience in breast ultrasound (senior breast radiologist: 20 years; junior breast radiologist: 3 years; general radiologist: less than 1 year) independently evaluated the lesions, assigning BI-RADS categories before and after Koios DS application. Biopsy reports served as the reference standard. AUCs and the number of unnecessary biopsies before and after implementing Koios DS were compared using DeLong and McNemar's tests.

RESULTS: A total of 222 patients (median age 58 years, interquartile range 46-72 years) with 226 lesions were included, 89/226 (39.4 %) benign and 137/226 (60.6 %) malignant. The application of Koios DS significantly improved (p < 0.001) the AUC of all radiologists, with a 0.078 AUC Δ for the junior breast radiologist (from 0.786 to 0.864), a 0.062 AUC Δ for the general radiologist (from 0.719 to 0.781), and a 0.045 AUC Δ for the senior breast radiologist (from 0.823 to 0.868). Koios DS would have significantly reduced the number of unnecessary biopsies recommended by the senior breast radiologist (from 41/89 [46.1 %] to 30/89 [33.7 %], p < 0.001) and by the junior breast radiologist (from 46/89 [51.7 %] to 29/89 [32.6 %], p = 0.001).

CONCLUSION: The application of Koios DS improved the radiologists' diagnostic performance, particularly for less experienced ones, and could potentially reduce unnecessary biopsies.

来源:Eur J Radiol,查看原文。版权归原作者所有。

引用

  1. Irmici G, Cozzi A, Depretto C, et al. Impact of an artificial intelligence decision support system among radiologists with different levels of experience in breast ultrasound: A prospective study in a tertiary center. Eur J Radiol. 2025. PMID: 40031378. doi:10.1016/j.ejrad.2025.112012

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。