机器学习在 MRI 识别临床上有意义的前列腺癌中的应用：一项荟萃分析。
Machine learning for the identification of clinically significant prostate cancer on MRI: a meta-analysis.
机构信息
Department of Advanced Biomedical Sciences, University of Naples "Federico II", Via Pansini 5, 80131, Naples, Italy.
出版信息
Eur Radiol. 2020 Dec;30(12):6877-6887. doi: 10.1007/s00330-020-07027-w. Epub 2020 Jun 30.
OBJECTIVES
The aim of this study was to systematically review the literature and perform a meta-analysis of machine learning (ML) diagnostic accuracy studies focused on clinically significant prostate cancer (csPCa) identification on MRI.
METHODS
Multiple medical databases were systematically searched for studies on ML applications in csPCa identification up to July 31, 2019. Two reviewers screened all papers independently for eligibility. The area under the receiver operating characteristic curves (AUC) was pooled to quantify predictive accuracy. A random-effects model estimated overall effect size while statistical heterogeneity was assessed with the I value. A funnel plot was used to investigate publication bias. Subgroup analyses were performed based on reference standard (biopsy or radical prostatectomy) and ML type (deep and non-deep).
RESULTS
After the final revision, 12 studies were included in the analysis. Statistical heterogeneity was high both in overall and in subgroup analyses. The overall pooled AUC for ML in csPCa identification was 0.86, with 0.81-0.91 95% confidence intervals (95%CI). The biopsy subgroup (n = 9) had a pooled AUC of 0.85 (95%CI = 0.79-0.91) while the radical prostatectomy one (n = 3) of 0.88 (95%CI = 0.76-0.99). Deep learning ML (n = 4) had a 0.78 AUC (95%CI = 0.69-0.86) while the remaining 8 had AUC = 0.90 (95%CI = 0.85-0.94).
CONCLUSIONS
ML pipelines using prostate MRI to identify csPCa showed good accuracy and should be further investigated, possibly with better standardisation in design and reporting of results.
KEY POINTS
• Overall pooled AUC was 0.86 with 0.81-0.91 95% confidence intervals. • In the reference standard subgroup analysis, algorithm accuracy was similar with pooled AUCs of 0.85 (0.79-0.91 95% confidence intervals) and 0.88 (0.76-0.99 95% confidence intervals) for studies employing biopsies and radical prostatectomy, respectively. • Deep learning pipelines performed worse (AUC = 0.78, 0.69-0.86 95% confidence intervals) than other approaches (AUC = 0.90, 0.85-0.94 95% confidence intervals).
目的
本研究旨在系统地回顾文献，并对基于机器学习（ML）的前列腺癌（PCa）诊断准确性研究进行荟萃分析，重点关注 MRI 上的临床显著 PCa（csPCa）识别。
方法
截至 2019 年 7 月 31 日，我们系统地检索了多个医学数据库，以获取有关 ML 在 csPCa 识别中应用的研究。两名评审员独立筛选所有论文以确定其合格性。通过受试者工作特征曲线下面积（AUC）来量化预测准确性。采用随机效应模型估计总体效应大小，同时采用 I 值评估统计异质性。采用漏斗图来调查发表偏倚。根据参考标准（活检或根治性前列腺切除术）和 ML 类型（深度和非深度）进行亚组分析。
结果
经过最终修订，共有 12 项研究纳入分析。总体和亚组分析中均存在高度统计学异质性。ML 在 csPCa 识别中的总体汇总 AUC 为 0.86，95%置信区间（95%CI）为 0.81-0.91。活检亚组（n=9）的汇总 AUC 为 0.85（95%CI=0.79-0.91），根治性前列腺切除术亚组（n=3）的汇总 AUC 为 0.88（95%CI=0.76-0.99）。深度学习 ML（n=4）的 AUC 为 0.78（95%CI=0.69-0.86），其余 8 项的 AUC 为 0.90（95%CI=0.85-0.94）。
结论
使用前列腺 MRI 识别 csPCa 的 ML 管道具有较好的准确性，应进一步研究，可能需要在设计和结果报告方面进行更好的标准化。
关键要点
总体汇总 AUC 为 0.86，95%置信区间为 0.81-0.91。
在参考标准亚组分析中，活检和根治性前列腺切除术研究的算法准确性相似，汇总 AUC 分别为 0.85（95%CI=0.79-0.91）和 0.88（95%CI=0.76-0.99）。
深度学习管道的表现不如其他方法（AUC=0.90，0.85-0.94），其 AUC 为 0.78（95%CI=0.69-0.86）。
Zotero 插件