From Hanoi Medical University Hospital, Hanoi, Vietnam (A.H.D.); Inserm, U1032, LabTau, Lyon, France (A.H.D., R.S., S.C., O.R.); CNRS, UMR 5553, Grenoble, France (C.M.); Université Joseph Fourier, Laboratoire d'Ecologie Alpine, Grenoble, France (C.M.); Hospices Civils de Lyon, Department of Urinary and Vascular Imaging, Hôpital Edouard Herriot, Lyon, France (P.C.M., F.B., G.P., O.R.); Université de Lyon, Lyon, France; Université Lyon 1, Faculté de Médecine Lyon Est, Lyon, France (P.C.M., S.C., M.C., O.R.); Hospices Civils de Lyon, Department of Pathology, Hôpital Edouard Herriot, Lyon, France (F.M.L.); Hospices Civils de Lyon, Department of Urology, Centre Hospitalier Lyon Sud, Pierre Bénite, France (A.R.); and Hospices Civils de Lyon, Department of Urology, Hôpital Edouard Herriot, Lyon, France (S.C., M.C.).
Radiology. 2018 May;287(2):525-533. doi: 10.1148/radiol.2017171265. Epub 2018 Jan 22.
Purpose To determine the performance of a computer-aided diagnosis (CAD) system trained at characterizing cancers in the peripheral zone (PZ) with a Gleason score of at least 7 in patients referred for multiparametric magnetic resonance (MR) imaging before prostate biopsy. Materials and Methods Two institutional review board-approved prospective databases of patients who underwent multiparametric MR imaging before prostatectomy (database 1) or systematic and targeted biopsy (database 2) were retrospectively used. All patients gave informed consent for inclusion in the databases. A CAD combining the 10th percentile of the apparent diffusion coefficient and the time to peak of enhancement was trained to detect cancers in the PZ with a Gleason score of at least 7 in 106 patients from database 1. The CAD was tested in 129 different patients from database 2. All targeted lesions were prospectively scored at biopsy by using a five-level Likert score. The CAD scores were retrospectively calculated. Biopsy results were used as the reference standard. Areas under the receiver operating characteristic curves (AUCs) were computed for CAD and Likert scores by using binormal smoothing for per-lesion and per-lobe analyses, and a density function for per-patient analysis. Results The CAD outperformed the Likert score in the overall population and all subgroups, except in the transition zone. The difference was statistically significant for the overall population (AUC, 0.95 [95% confidence interval {CI}: 0.90, 0.98] vs 0.88 [95% CI: 0.68, 0.96]; P = .02) at per-patient analysis, and for less-experienced radiologists (<1 year) at per-lesion (AUC, 0.90 [95% CI: 0.81, 0.95] vs 0.83 [95% CI: 0.73, 0.90]; P = .04) and per-lobe (AUC, 0.92 [95% CI: 0.80, 0.96] vs 0.84 [95% CI: 0.72, 0.91]; P = .04) analysis. Conclusion The CAD outperformed the Likert score prospectively assigned at biopsy in characterizing cancers with a Gleason score of at least 7. RSNA, 2018 Online supplemental material is available for this article.
目的 旨在评估一种计算机辅助诊断（CAD）系统的性能，该系统通过对接受前列腺多参数磁共振成像（mpMRI）检查且拟行前列腺穿刺活检的患者外周带（PZ）内至少 7 分的癌进行特征描述来进行训练。
材料与方法 本研究回顾性分析了两个经机构审查委员会批准的数据库中的患者资料，这些患者分别接受了前列腺切除术（数据库 1）或系统和靶向活检（数据库 2）前的 mpMRI 检查。所有患者均签署了纳入研究的知情同意书。通过对数据库 1 中 106 例患者的第 10 个百分位数的表观扩散系数和增强峰值时间进行组合，训练出一种 CAD，以检测 PZ 内的 Gleason 评分至少为 7 分的癌症。该 CAD 在数据库 2 中的 129 例不同患者中进行了测试。所有靶向病变均在活检时通过 5 级 Likert 评分进行前瞻性评分。对 CAD 评分进行回顾性计算。以活检结果为参考标准。通过二项正态分布平滑法进行基于病变和基于叶的分析，通过密度函数进行基于患者的分析，计算 CAD 和 Likert 评分的受试者工作特征曲线（ROC）下面积（AUC）。
结果 在总体人群和所有亚组中，CAD 的表现优于 Likert 评分，除了移行带。在总体人群中，差异具有统计学意义（AUC：0.95[95%置信区间：0.90，0.98]与 0.88[95%置信区间：0.68，0.96]；P =.02），在经验较少的放射科医生（<1 年）中，基于病变的 AUC 也存在显著差异（0.90[95%置信区间：0.81，0.95]与 0.83[95%置信区间：0.73，0.90]；P =.04），基于叶的 AUC 也存在显著差异（0.92[95%置信区间：0.80，0.96]与 0.84[95%置信区间：0.72，0.91]；P =.04）。
结论 在对拟行前列腺穿刺活检的患者的 PZ 内至少 7 分的癌进行特征描述方面，CAD 优于前瞻性分配的 Likert 评分。
RSNA，2018 在线补充材料可供本文参考。
Zotero 插件