文献速递神经MR

联合MRI与组织病理学影像特征预测胶质瘤患者总生存期

Combining MRI and Histologic Imaging Features for Predicting Overall Survival in Patients with Glioma.
PubMed 文献 · Radiol Imaging Cancer · 2021-01-01

本研究利用癌症影像档案库中171例胶质瘤患者的 multiparametric MR 与数字病理图像提取特征,比较仅MRI、仅病理及两者联合的Cox与SVM模型预测总生存期的效能。联合模型在全部患者、高级别及低级别胶质瘤中C-index均高于单一模型,二分类AUC亦更高。

要点

  • 联合MRI与数字病理特征的Cox模型预测总生存期的C-index在全部患者为0.79,高于仅MRI的0.70和仅病理的0.67。
  • 联合模型在低级别胶质瘤中C-index达0.88,高于仅MRI的0.62和仅病理的0.62。
  • 联合SVM模型二分类AUC为0.86,高于仅MRI的0.68和仅病理的0.72。

核心结论

  • 联合MR与组织病理学图像特征的模型预测总生存期的准确性高于单独使用MR或组织病理学图像特征的模型。

研究设计与样本

  • 研究论文;从癌症影像档案库获取171例胶质瘤患者的multiparametric MR与组织病理学图像(原始图像采集于1983-2008年),提取强度、直方图与纹理等特征,采用Cox比例风险回归与支持向量机分类模型,在训练-测试拆分配置中评估。

关键影像表现/征象

  • 从MR与组织病理学图像的勾画肿瘤区域提取强度、直方图与纹理等工程特征。
  • 联合模型在全部患者、高级别胶质瘤与低级别胶质瘤中C-index均高于单一模态模型。

诊断标准与关键数值

  • 全部患者中,联合Cox模型C-index 0.79 vs 仅MRI 0.70(P=0.02)vs 仅病理0.67(P=0.01)。
  • 高级别胶质瘤中,联合Cox模型C-index 0.78 vs 仅MRI 0.68(P=0.03)vs 仅病理0.64(P=0.01)。
  • 低级别胶质瘤中,联合Cox模型C-index 0.88 vs 仅MRI 0.62(P=0.008)vs 仅病理0.62(P=0.006)。
  • 二分类中,联合SVM模型AUC 0.86(95%CI: 0.80, 0.95)vs 仅MRI 0.68(95%CI: 0.50, 0.81;P=0.01)vs 仅病理0.72(95%CI: 0.60, 0.85;P=0.04)。

临床应用与阅片要点

  • 原文未据此给出推荐。

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

Purpose To test the hypothesis that combined features from MR and digital histopathologic images more accurately predict overall survival (OS) in patients with glioma compared with MRI or histopathologic features alone. Materials and Methods Multiparametric MR and histopathologic images in patients with a diagnosis of glioma (high- or low-grade glioma [HGG or LGG]) were obtained from The Cancer Imaging Archive (original images acquired 1983-2008). An extensive set of engineered features such as intensity, histogram, and texture were extracted from delineated tumor regions in MR and histopathologic images. Cox proportional hazard regression and support vector machine classification (SVC) models were applied to (a) MRI features only (MRIcox/svc), histopathologic features only (HistoPathcox/svc), and (c) combined MRI and histopathologic features (MRI+HistoPathcox/svc) and evaluated in a split train-test configuration. Results A total of 171 patients (mean age, 51 years ± 15; 91 men) were included with HGG (n = 75) and LGG (n = 96). Median OS was 467 days (range, 3-4752 days) for all patients, 350 days (range, 15-1561 days) for HGG, and 595 days (range, 3-4752 days) for LGG. The MRI+HistoPathcox model demonstrated higher concordance index (C-index) compared with MRIcox and HistoPathcox models on all patients (C-index, 0.79 vs 0.70 [P = .02; MRIcox] and 0.67 [P = .01; HistoPathcox]), patients with HGG (C-index, 0.78 vs 0.68 [P = .03; MRIcox] and 0.64 [P = .01; HistoPathcox]), and patients with LGG (C-index, 0.88 vs 0.62 [P = .008; MRIcox] and 0.62 [P = .006; HistoPathcox]). In binary classification, the MRI+HistoPathsvc model (area under the receiver operating characteristic curve [AUC], 0.86 [95% CI: 0.80, 0.95]) had higher performance than the MRIsvc model (AUC, 0.68 [95% CI: 0.50, 0.81]; P = .01) and the HistoPathsvc model (AUC, 0.72 [95% CI: 0.60, 0.85]; P = .04). Conclusion The model combining features from MR and histopathologic images had higher accuracy in predicting OS compared with the models with MR or histopathologic images alone. Keywords: Survival Prediction, Gliomas, Digital Pathology Imaging, MR Imaging, Machine Learning Supplemental material is available for this article.

来源:Radiol Imaging Cancer,查看原文。版权归原作者所有。

引用

  1. Rathore S, Chaddad A, Iftikhar MA, et al. Combining MRI and Histologic Imaging Features for Predicting Overall Survival in Patients with Glioma. Radiol Imaging Cancer. 2021. PMID: 34296969. doi:10.1148/rycan.2021200108

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。