| 颜蕾.基于CT影像组学模型鉴别肾乏脂肪血管平滑肌脂肪瘤与均质肾透明细胞癌[J].中国医学影像技术,2020,36(5): |
| 基于CT影像组学模型鉴别肾乏脂肪血管平滑肌脂肪瘤与均质肾透明细胞癌 |
| CT-based radiomics model in differentiation of fat-poor renal angiomyolipoma and homogeneous-density clear cell renal cell carcinoma |
| 投稿时间：2019-08-06 修订日期：2020-05-18 |
| DOI： |
| 中文关键词: 肾透明细胞癌 血管平滑肌脂肪瘤 CT增强扫描 影像组学 |
| 英文关键词:Clear cell renal cell carcinoma Angiomyolipoma Enhanced CT scan Radiomics |
| 基金项目:青岛市市南区科技计划项目 |
|
| 摘要点击次数: 2162 |
| 全文下载次数: 365 |
| 中文摘要: |
| 目的 基于增强CT影像组学特征联合临床特征建立综合模型，并验证其在术前鉴别肾乏脂肪血管平滑肌脂肪瘤(fp-AML)与均质肾透明细胞癌(hd-CCRCC)的效能。方法 回顾性分析2010年7月至2017年12月经病理证实的32例fp-AML与39例hd-ccRCC患者的临床资料及术前增强CT图像。在皮质期、实质期及排泄期图像手工进行肿瘤ROI的勾画，并提取影像特征。通过计算观察者间及观察者自身ICCs排除观察者勾画ROI的主观差异，采用LASSO回归进行特征选择，通过Logistic多元回归分析构建回归方程并计算皮质期、实质期、排泄期及三期联合的影像组学得分。将临床特征及影像组学得分通过Logistic多元回归分析建立综合模型，并绘制列线图。通过Hosmer-Lemeshow拟合优度检验评价列线图的拟合度，并绘制校准曲线。通过ROC曲线分析检测列线图的鉴别效能。通过决策曲线评价列线图鉴别fp-AML和hd-ccRCC的净获益。结果 每期图像提取包括强度、形状、纹理、图像滤波在内的共1029个特征，将符合观察者间及观察者自身ICC均＞0.75的影像组学特征进行LASSO特征选择后，皮质期、实质期、排泄期和三期联合分别得6个、6个、5个和7个有鉴别意义的特征，影像组学得分AUC分别为0.83(95%CI：0.73～0.92)、0.80(95%CI：0.70～0.91)、0.78(95%CI：0.68～0.89)和0.86(95%CI：0.77～0.95)。基于三期联合影像组学得分和临床特征的列线图AUC为0.90(95%CI：0.81～0.99)，决策曲线表明利用列线图术前鉴别fp-AML和hd-ccRCC可获得较为满意的诊断净获益。结论 基于增强CT影像组学特征联合临床特征建立的综合模型，通过列线图表示，在术前鉴别fp-AML与hd-ccRCC中具有较高的诊断效能，有助于肾肿瘤的术前定性诊断。 |
| 英文摘要: |
| Objective To evaluate the value of an enhanced CT-based radiomics nomogram incorporated with radiomics signatures and clinical factors for differentiating fp-AML from hd-ccRCC before surgery. Methods From July 2010 to December 2017, the clinical data and enhanced CT images of 71 patients with pathologically proven fp-AML (n=32) and hd-ccRCC (n=39) were retrospectively collected. The 3-dimensional regions of interest (ROIs) were contoured manually at cortical, nephrographic and excretory phases (CP, NP and EP), and the radiomics features were extracted. Inter and intra class correlation coefficients (ICCs) were used to exclude the inter-observer and intra-observer difference of ROI feature extraction. The least absolute shrinkage and selection operator (LASSO) regression method were used to select radiomics features. A regression formula was constructed by multivariate logistic regression analysis. Radiomics scores of CP, NP, EP and the three phases were calculated. A combined radiomics nomogram was developed by incorporating the clinical factors and radiomics score, by using a multivariate Cox regression model. The Hosmer–Lemeshow test was performed to assess the goodness-of-fit of the nomogram. The calibration of the nomogram was assessed by using calibration curves. The differential effectiveness of the radiomics nomogram was evaluated on the basis of ROC curves. The decision curve analysis (DCA) was performed to evaluate the net benefits of the nomogram for differentiating fp-AML from hd-ccRCC. Results 1029 features including intensity, shape, texture and wavelet features were extracted from each phase. The radiomics features with ICCs greater than 0.75 were enrolled into the LASSO Cox regression model. 6, 6, 5 and 7 optimal features, respectively extracted from CP, NP, EP and 3 phases, were selected and the area under the curve (AUC) was 0.83 (95%CI [confidence interval]: 0.73～0.92), 0.80 (95%CI: 0.70～0.91), 0.78 (95%CI: 0.68～0.89) and 0.86 (95%CI: 0.77～0.95), respectively. The AUC of the nomogram based on the radiomics score of three phases and the clinical factors was 0.90 (95%CI：0.81～0.99). The DCA indicated that the radiomics nomogram had a satisfactory overall net benefit for differentiating fp-AML from hd-ccRCC before surgery. Conclusion The enhanced CT-based radiomics nomogram, which incorporates the radiomics signature and clinical factors, shows favorable predictive value for differentiating fp-AML from hd-ccRCC, which might assist clinicians in accurate diagnosis before surgery. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
