文献速递乳腺乳腺X线

深度学习模型对筛查乳腺X线摄影进行分流的模拟研究

A Deep Learning Model to Triage Screening Mammograms: A Simulation Study.
PubMed 文献 · Radiology · 2019-08-06

本回顾性研究纳入2009年1月至2016年12月66 661名女性的223 109次连续筛查乳腺X线摄影,开发深度学习模型将部分乳腺X线摄影分流为无癌并模拟分流工作流程。模拟流程中放射科医生阅读80.7%的乳腺X线摄影,特异度从93.5%提高至94.2%(P=0.002),敏感度90.1%在5%非劣效界值下非劣于90.6%(P<0.001)。

要点

  • 深度学习分流模型可将部分筛查乳腺X线摄影判为无癌,模拟流程中放射科医生仅需阅读80.7%的乳腺X线摄影。
  • 模拟分流流程特异度94.2% vs 全部阅读93.5%(P=0.002),敏感度90.1% vs 90.6%在5%界值下非劣(P<0.001)。
  • 该模型有潜力减少放射科医生工作量并显著提高特异度而不损害敏感度。

核心结论

  • 该深度学习模型有潜力减少放射科医生工作量并显著提高特异度而不损害敏感度。

研究设计与样本

  • 回顾性研究,纳入2009年1月至2016年12月66 661名女性的223 109次连续筛查乳腺X线摄影,癌症结局通过链接区域肿瘤登记库获得。
  • 按患者将队列分为训练集212 272次(56 831名患者)、验证集25 999次(7021名患者)和测试集26 540次(7176名患者)。
  • 开发深度学习模型将乳腺X线摄影分流为无癌,并在测试集上评估;模拟分流工作流程中放射科医生跳过被分流为无癌的乳腺X线摄影(将其判为无癌阴性),对未被分流为无癌的乳腺X线摄影使用原阅片医生的评估进行阅读。
  • 统计量基于5000个bootstrap样本计算置信区间(CI);特异度采用双侧t检验比较(P<0.05),敏感度采用单侧t检验比较,非劣效界值为5%(P<0.05)。

诊断标准与关键数值

  • 测试集7176名女性(平均年龄57.8岁±10.9[标准差])。
  • 全部阅读时放射科医生敏感度90.6%(173/191;95%CI:86.6%,94.7%),特异度93.5%(24 625/26 349;95%CI:93.3%,93.9%)。
  • 深度学习模拟分流流程中放射科医生敏感度90.1%(172/191;95%CI:86.0%,94.3%),特异度94.2%(24 814/26 349;95%CI:94.0%,94.6%),阅读80.7%(21 420/26 540)的乳腺X线摄影。
  • 模拟流程改善特异度(P=0.002),敏感度在5%界值下非劣(P<0.001)。

临床应用与阅片要点

  • 深度学习分流模型可将部分筛查乳腺X线摄影判为无癌,模拟流程中放射科医生仅阅读80.7%的乳腺X线摄影,特异度94.2% vs 全部阅读93.5%(P=0.002)。
  • 模拟流程敏感度90.1% vs 全部阅读90.6%,在5%非劣效界值下非劣(P<0.001)。
  • 该模型有潜力减少放射科医生工作量并显著提高特异度而不损害敏感度。

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 Recent deep learning (DL) approaches have shown promise in improving sensitivity but have not addressed limitations in radiologist specificity or efficiency. Purpose To develop a DL model to triage a portion of mammograms as cancer free, improving performance and workflow efficiency. Materials and Methods In this retrospective study, 223 109 consecutive screening mammograms performed in 66 661 women from January 2009 to December 2016 were collected with cancer outcomes obtained through linkage to a regional tumor registry. This cohort was split by patient into 212 272, 25 999, and 26 540 mammograms from 56 831, 7021, and 7176 patients for training, validation, and testing, respectively. A DL model was developed to triage mammograms as cancer free and evaluated on the test set. A DL-triage workflow was simulated in which radiologists skipped mammograms triaged as cancer free (interpreting them as negative for cancer) and read mammograms not triaged as cancer free by using the original interpreting radiologists' assessments. Sensitivities, specificities, and percentage of mammograms read were calculated, with and without the DL-triage-simulated workflow. Statistics were computed across 5000 bootstrap samples to assess confidence intervals (CIs). Specificities were compared by using a two-tailed t test (P < .05) and sensitivities were compared by using a one-sided t test with a noninferiority margin of 5% (P < .05). Results The test set included 7176 women (mean age, 57.8 years ± 10.9 [standard deviation]). When reading all mammograms, radiologists obtained a sensitivity and specificity of 90.6% (173 of 191; 95% CI: 86.6%, 94.7%) and 93.5% (24 625 of 26 349; 95% CI: 93.3%, 93.9%). In the DL-simulated workflow, the radiologists obtained a sensitivity and specificity of 90.1% (172 of 191; 95% CI: 86.0%, 94.3%) and 94.2% (24 814 of 26 349; 95% CI: 94.0%, 94.6%) while reading 80.7% (21 420 of 26 540) of the mammograms. The simulated workflow improved specificity (P = .002) and obtained a noninferior sensitivity with a margin of 5% (P < .001). Conclusion This deep learning model has the potential to reduce radiologist workload and significantly improve specificity without harming sensitivity. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Kontos and Conant in this issue.

来源:Radiology,查看原文。版权归原作者所有。

引用

  1. Yala A, Schuster T, Miles R, et al. A Deep Learning Model to Triage Screening Mammograms: A Simulation Study. Radiology. 2019. PMID: 31385754. doi:10.1148/radiol.2019182908

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。