Departments of Diagnostic Radiology and Nuclear Medicine (I.Y., Y.S., U.T.), Comprehensive Reproductive Medicine (N.O., N.M., K.W., A.W.), Oral and Maxillofacial Radiology (J.S.), and Human Pathology (D.K.), Graduate School, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan.
Radiol Imaging Cancer. 2020 May 22;2(3):e190085. doi: 10.1148/rycan.2020190085. eCollection 2020 May.
To determine the feasibility of texture analysis of apparent diffusion coefficient (ADC) maps and to assess the performance of texture analysis and ADC to predict histologic grade, parametrial invasion, lymph node metastasis, International Federation of Gynecology and Obstetrics (FIGO) stage, recurrence, and recurrence-free survival (RFS) in patients with cervical carcinoma.
This retrospective study included 58 patients with cervical carcinoma who were examined with a 1.5-T MRI system and diffusion-weighted imaging with values of 0 and 1000 sec/mm. Software with volumes of interest on ADC maps was used to extract 45 texture features, including higher-order texture features. Receiver operating characteristic (ROC) analysis was performed to compare the diagnostic performance of ADC map random forest models and of ADC values. Dunnett test, Spearman rank correlation coefficient, Kaplan-Meier analyses, log-rank test, and Cox proportional hazards regression analyses were also used for statistical analyses.
The ADC map random forest models showed a significantly larger area under the ROC curve (AUC) than the AUC of ADC values for predicting high-grade cervical carcinoma ( = .0036), but not for parametrial invasion, lymph node metastasis, stages III-IV, and recurrence ( = .0602, .3176, .0924, and .5633, respectively). The random forest models predicted that the mean RFS rates were significantly shorter for high-grade cervical carcinomas, parametrial invasion, lymph node metastasis, stages III-IV, and recurrence ( = .0405, < .0001, .0344, .0001, and .0015, respectively); the random forest models for parametrial invasion and stages III-IV were more useful than ADC values ( = .0018) for predicting RFS.
The ADC map random forest models were more useful for noninvasively evaluating histologic grade, parametrial invasion, lymph node metastasis, FIGO stage, and recurrence and for predicting RFS in patients with cervical carcinoma than were ADC values. Comparative Studies, Genital/Reproductive, MR-Diffusion Weighted Imaging, MR-Imaging, Neoplasms-Primary, Pathology, Pelvis, Tissue Characterization, Uterus© RSNA, 2020See also the commentary by Reinhold and Nougaret in this issue.
确定表观扩散系数（ADC）图纹理分析的可行性，并评估纹理分析和 ADC 预测宫颈癌组织学分级、宫旁侵犯、淋巴结转移、国际妇产科联合会（FIGO）分期、复发和无复发生存（RFS）的性能。
本回顾性研究纳入 58 例宫颈癌患者，均在 1.5T MRI 系统上进行检查，扩散加权成像 b 值为 0 和 1000 sec/mm²。采用 ADC 图容积感兴趣区软件提取 45 个纹理特征，包括高阶纹理特征。采用受试者工作特征（ROC）分析比较 ADC 图随机森林模型和 ADC 值的诊断性能。Dunnett 检验、Spearman 秩相关系数、Kaplan-Meier 分析、log-rank 检验和 Cox 比例风险回归分析用于统计学分析。
ADC 图随机森林模型预测高级别宫颈癌的 ROC 曲线下面积（AUC）显著大于 ADC 值（=0.0036），但对宫旁侵犯、淋巴结转移、FIGO 分期 III-IV 和复发的 AUC 无显著差异（=0.0602、0.3176、0.0924 和 0.5633）。随机森林模型预测，高级别宫颈癌、宫旁侵犯、淋巴结转移、FIGO 分期 III-IV 和复发患者的平均 RFS 率显著缩短（=0.0405、<0.0001、0.0344、0.0001 和 0.0015）；宫旁侵犯和 FIGO 分期 III-IV 的随机森林模型预测 RFS 的效能优于 ADC 值（=0.0018）。
与 ADC 值相比，ADC 图随机森林模型在评估宫颈癌组织学分级、宫旁侵犯、淋巴结转移、FIGO 分期和复发，预测 RFS 方面更有用。比较研究，生殖器官/生殖，MR 扩散加权成像，MR 成像，肿瘤-原发性，病理学，骨盆，组织特征，子宫©RSNA，2020 也见本期 Reinhold 和 Nougaret 的评论。
