机器学习在乳腺X线筛查工作流程中的应用:系统综述与Meta分析
该系统综述与Meta分析检索2012年1月至2020年9月数据库,纳入14篇文章共15项研究,评估独立机器学习算法在乳腺X线筛查工作流程中的性能;汇总敏感度、特异度和AUC分别为75.4%、90.6%和0.89,读者分别为73.0%、88.6%和0.85;结论认为算法达到或超过读者检出性能并提高效率,但证据来自少量回顾性研究,需进一步严格独立外部前瞻性测试。
要点
- 分诊研究中,17%-91%的正常乳腺X线片可被适配筛查流程判读,而估计漏诊0%-7%的癌症。
- 算法汇总敏感度75.4%、特异度90.6%、AUC 0.89,读者分别为73.0%、88.6%、0.85。
- 证据来自少量回顾性研究,需进一步严格独立外部前瞻性测试以支持结论。
核心结论
- 独立机器学习算法在乳腺X线筛查工作流程中达到甚至超过人类读者检出性能并提高效率,但证据来自少量回顾性研究,需进一步严格独立外部前瞻性测试以支持这些主张。
研究设计与样本
- 系统综述与Meta分析,检索Ovid Embase、Ovid Medline、Cochrane Central Register of Controlled Trials、Scopus和Web of Science数据库2012年1月至2020年9月发表的文献,纳入14篇文章共15项研究(独立检测8项、分诊7项),使用QUADAS-2和PROBAST进行质量评估,使用CLAIM评估报告,主要Meta分析纳入表现最佳算法及对应读者性能,采用双变量模型计算AUC汇总估计。
诊断标准与关键数值
- 分诊研究报告可被适配筛查判读的正常乳腺X线片比例为17%-91%,估计漏诊癌症比例为0%-7%;主要Meta分析纳入3个国家估计185 252例、超过39名读者,算法汇总敏感度75.4%(95%CI:65.6,83.2;P=.11)、特异度90.6%(95%CI:82.9,95.0;P=.40)、AUC 0.89(95%CI:0.84,0.98),读者汇总敏感度73.0%(95%CI:60.7,82.6)、特异度88.6%(95%CI:72.4,95.8)、AUC 0.85(95%CI:0.78,0.97)。
临床应用与阅片要点
- 原文结论认为独立机器学习算法在乳腺X线筛查工作流程中达到甚至超过人类读者检出性能并提高效率,但证据来自少量回顾性研究,需进一步严格独立外部前瞻性测试以支持这些主张。
局限性
- 原文结论指出证据来自少量回顾性研究。
- 原文结论指出需进一步严格独立外部前瞻性测试机器学习算法以在预设阈值下评估性能,从而支持这些主张。
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 Advances in computer processing and improvements in data availability have led to the development of machine learning (ML) techniques for mammographic imaging. Purpose To evaluate the reported performance of stand-alone ML applications for screening mammography workflow. Materials and Methods Ovid Embase, Ovid Medline, Cochrane Central Register of Controlled Trials, Scopus, and Web of Science literature databases were searched for relevant studies published from January 2012 to September 2020. The study was registered with the PROSPERO International Prospective Register of Systematic Reviews (protocol no. CRD42019156016). Stand-alone technology was defined as a ML algorithm that can be used independently of a human reader. Studies were quality assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 and the Prediction Model Risk of Bias Assessment Tool, and reporting was evaluated using the Checklist for Artificial Intelligence in Medical Imaging. A primary meta-analysis included the top-performing algorithm and corresponding reader performance from which pooled summary estimates for the area under the receiver operating characteristic curve (AUC) were calculated using a bivariate model. Results Fourteen articles were included, which detailed 15 studies for stand-alone detection (n = 8) and triage (n = 7). Triage studies reported that 17%-91% of normal mammograms identified could be read by adapted screening, while "missing" an estimated 0%-7% of cancers. In total, an estimated 185 252 cases from three countries with more than 39 readers were included in the primary meta-analysis. The pooled sensitivity, specificity, and AUC was 75.4% (95% CI: 65.6, 83.2; P = .11), 90.6% (95% CI: 82.9, 95.0; P = .40), and 0.89 (95% CI: 0.84, 0.98), respectively, for algorithms, and 73.0% (95% CI: 60.7, 82.6), 88.6% (95% CI: 72.4, 95.8), and 0.85 (95% CI: 0.78, 0.97), respectively, for readers. Conclusion Machine learning (ML) algorithms that demonstrate a stand-alone application in mammographic screening workflows achieve or even exceed human reader detection performance and improve efficiency. However, this evidence is from a small number of retrospective studies. Therefore, further rigorous independent external prospective testing of ML algorithms to assess performance at preassigned thresholds is required to support these claims. ©RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Whitman and Moseley in this issue.
引用
- Hickman SE, Woitek R, Le EPV, et al. Machine Learning for Workflow Applications in Screening Mammography: Systematic Review and Meta-Analysis. Radiology. 2021. PMID: 34665034. doi:10.1148/radiol.2021210391
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。