基于BOLD-MRI的影像组学鉴别肾良恶性肿瘤

Application of BOLD-MRI-based radiomics in differentiating malignant from benign renal tumors

目的:血氧水平依赖磁共振成像(blood oxygen level dependent magnetic resonance imaging,BOLD-MRI)技术是一种反映组织和细胞血氧水平的无创磁共振成像方法,本研究探讨基于BOLD-MRI的影像组学鉴别肾良、恶性肿瘤的价值。方法:回顾性分析经手术病理证实并于术前2周内行肾T_(1)加权成像(T_(1) weighted imaging,T_(1)WI)、T_(2)加权成像(T_(2) weighted imaging,T_(2)WI)及BOLD-MRI扫描的141例肾肿瘤患者的资料。其中男74例,女67例;年龄26~78(中位年龄56)岁;恶性肿瘤118例,良性肿瘤23例。将良、恶性肿瘤分成训练组(良性17例,恶性83例)和验证组(良性6例,恶性35例)。由两位未知肿瘤病理类型的影像科医师(A、B)独立勾画肿瘤最大层面的边缘,生成感兴趣区(regions of interest,ROI),医师B间隔1个月再次勾画ROI,用组内相关系数(intra-class correlation coefficient,ICC)评估观察者间及观察者内一致性(ICC>0.75为一致性较好)。将T_(2)^(*)Mapping图像及ROI图像导入Artificial Intelligence Kit软件,从每个ROI中提取出基于形态学、直方图、灰度共生矩阵、灰度游程长度矩阵、灰度区域大小矩阵、灰度依赖性矩阵的396个影像组学特征。对训练组采用最小冗余最大相关性(minimum redundancy maximum relevance,mRMR)算法筛选出冗余度最低相关性最高的特征,再使用最小绝对收缩与选择算子(least absolute shrinkage and selection operator,LASSO)法筛选出预测价值最高的特征,采用多因素logistic回归构建影像组学模型。计算每个病例的影像组学分值(Radscore)。采用Wilcoxon检验比较训练组和验证组肾良、恶性肿瘤之间Radscore的差异;使用受试者工作特征(receiver operating characteristic,ROC)曲线和留组交叉验证(leave group out cross validation,LGOCV)法分析和评估模型鉴别肾良、恶性肿瘤的诊断效能;采用决策曲线分析评估模型的临床应用价值。结果:良、恶性肿瘤两组患者年龄差异有统计学意义(t=4.383,P<0.001);性别构成差异无统计学意义(χ^(2)=3.452,P=0.063);两组肿瘤最大径差异无统计学意义(t=1.432,P=0.154)。医师A和医师B各自独立测量的组间ICC值为0.71~0.87,医师B两次测量的组内ICC为0.76~0.91。采用mRMR算法筛选出30个冗余度最低相关性最高的特征,采用LASSO法筛选出12个预测价值最高的特征。训练组和验证组恶性肿瘤Radscore均高于良性肿瘤(分别P<0.001和P=0.006)。影像组学模型鉴别训练组、验证组肾良、恶性肿瘤的ROC曲线下面积分别为0.881、0.706,准确度分别为82.93%、79.00%,敏感度分别为82.86%、77.11%,特异度分别为83.33%、88.24%。决策曲线分析结果显示当阈值高于0.3时,影像组学模型的净收益率高于被识别为“全部恶性”或“全部良性”的净收益率。结论:基于BOLD-MRI影像组学可作为一种无创预测肾良、恶性肿瘤的工具。

Objective: Blood oxygen level dependent magnetic resonance imaging(BOLD-MRI) is a kind of non-invasive MRI technology which reflects the tissue blood oxyen levels. This stuy aims to explore the value of radiomics based on BOLD-MRI in differentiating malignant from benign renal tumors.Methods: A total of 141 patients with renal tumors confirmed by pathology were retrospectively analyzed. Seventy-four men and sixty-seven women, aged 26-78 years, with a median age of 56, were included. In all patients, 118 with malignant tumors and 23 with benign tumors were confirmed. All the patients underwent renal T1 weighted imaging(T1 WI), T2 weighted imaging(T2 WI), and BOLD-MRI scan within 2 weeks before surgery.The patients were randomly assigned into a training group(benign, n=17;malignant, n=83)and a test group(benign, n=6;malignant, n=35). Two radiologists(A and B), who were blind to the pathological results, delineated the regions of interest(ROI) on the maximum axial slices of the tumors. Radiologist B delineated the ROI again at an interval of one month. The intra-class correlation coefficient(ICC) was used to evaluate inter-observer and intra-observer repeatability and ICC>0.75 represented as a good consistency. All the T2*Mapping images and the related ROI files were loaded into the Artificial Intelligence Kit software. A total of 396 texture features, which were calculated based on morphology,histogram, gray level co-occurrence matrix, gray-scale run length matrix, gray-scale area size matrix and gray-scale dependent matrix, were extracted from each ROI. The lowest redundancy and the highest correlation were filtered using minimum redundancy maximum relevance(m RMR) algorithm. Then least absolute shrinkage and selection operator(LASSO) algorithm was used to screened out the most predictive features. Multivariate logistic regression was performed to develop the prediction model after feature selection.The radiomics signature score(Radscore) of each case was calculated. The Wilcoxon test was used to compare the difference in the Radscore between benign and malignant renal tumors in the training and test groups. The diagnostic performance of the model in differentiating malignant from benign renal tumors was evaluated with receiver operating characteristic(ROC) curve and leave group out cross validation. The clinical application value of the model was evaluated by decision curve analysis(DCA).Results: There was significant difference in the age between the patients with benign and those with malignant tumors(t=4.383, P<0.001). There were no significant differences in gender composition and in the largest tumor diameter between the 2 groups(χ2=3.452,P=0.063;t=1.432, P=0.154). The ICC values of all the texture features for the inter-observer repeatability were ranged from 0.71 to 0.87, and the ICC values for the intra-observer repeatability were ranged from 0.76 to 0.91. Thirty features with the lowest redundancy and the highest correlation were screened out. The most predictive 12 features were filtered out.The Radscores of malignant tumors in the training and test groups were higher than those of benign tumors(P<0.001 and P=0.006, respectively). The areas under the ROC curve of the model developed by multivariable logistic regression for differentiating malignant from benign renal tumors in the training and test groups were 0.881 and 0.706, with the accuracy at 82.93% and 79.00%, the sensitivity at 82.86% and 77.11%, and the specificities at83.33% and 88.24%, respectively. The results of decision curve analysis showed that the net benefit of the radiomics model was higher than that of"all malignant"or"all benign"when the threshold was higher than 0.3.Conclusion: BOLD-MRI-based radiomics can be a reliable non-invasive approach for differentiating renal malignant tumors from benign tumors.