论文部分内容阅读
目的:探讨在n 18F-脱氧葡萄糖(FDG) PET/CT图像中采用不同方法勾画胰腺导管腺癌(PDAC)肿瘤区域对使用影像组学特征预测病理分化程度的影响。n 方法:回顾性分析2010年9月至2016年1月间于北京协和医院经病理证实的72例PDAC患者(男46例、女26例，年龄：25～87岁)的术前n 18F-FDG PET/CT图像及病理资料。根据PDAC病理分化程度将患者分为高分化和非高分化组。入组患者按3∶1的比例随机划分至训练集和验证集。所有病例由2位医师手动勾画感兴趣区(ROI；记为ROI_M1和ROI_M2)，再分别基于标准摄取值(SUV)梯度(记为ROI_G)和40%最大SUV(SUVn max)阈值(记为ROI_S)半自动勾画ROI。计算并比较4种勾画结果的体积、戴斯相似性系数(DSC)。从PET/CT原始和预处理图像中提取形状、一阶、纹理等特征，并以组间相关系数(ICC)评估每个特征在不同勾画结果间的一致性。使用Kruskal-Wallis秩和检验、两独立样本n t检验或n z检验分析数据。采用受试者工作特征曲线下面积评估模型准确性，并通过交叉验证评估模型泛化能力。n 结果:训练集共55例患者(高分化14例，非高分化41例)；验证集共17例患者(高分化4例，非高分化13例)。20个特征组中共筛选出44个对PDAC分化程度有预测价值的特征。ROI_M1、ROI_M2、ROI_G和ROI_S勾画的轮廓体积分别为10.29(4.01，19.43)、9.34(4.26，17.27)、11.86(5.52，19.74)和15.08(9.62，27.44) cmn 3，差异有统计学意义(n H＝18.641, n P0.05)；ROI_M2的预测模型准确性优于ROI_G(n z＝3.031, n P＝0.002)，但泛化能力不足(n t＝3.086, n P＝0.012)。n 结论:基于手动勾画构建的预测模型准确性较高，但模型性能不稳定；基于梯度的半自动勾画可以达到与手动勾画相似的准确性，且模型泛化能力更强。“,”Objective:To investigate the segmentation methods of pancreatic ductal adenocarcinoma (PDAC) tumor regions in n 18F-fluorodeoxyglucose (FDG) PET/CT images, as well as their impact on radiomic features-based pathological grade prediction.n Methods:A total of 72 patients (46 males, 26 females, age range: 25-87 years) with pathologically confirmed PDAC and a preoperative n 18F-FDG PET/CT scan in Peking Union Medical College Hospital between September 2010 and January 2016 were enrolled retrospectively. The cohort of patients was classified as well differentiated group and non-well differentiated group based on the pathological grade of PDAC, and patients were divided into training set and validation set in the ratio of 3∶1 randomly. Two physicians performed manual contours in the tumor region (referred as region of interest (ROI)_M1 and ROI_M2) and semi-automatic ROIs based on standardized uptake value (SUV) gradient edge search (referred as ROI_G) and 40% threshold applied to the maximum SUV (SUVn max; referred as ROI_S) were drawn. The four types of segmentation results were compared in terms of volume and Dice similarity coefficient (DSC). Shape, first-order, and texture features were extracted from PET/CT original and preprocessed images, and the interclass correlation coefficient (ICC) was used to assess each feature′s consistency across all segmentations. Kruskal-Wallis rank sum test, independent-sample n t test or n z test were used to analyze the data. The area under the receiver operating characteristic curve was used to assess model accuracy, and cross validation was used to assess generalization ability.n Results:There were 55 patients in the training set (14 well differentiated cases and 41 non-well differentiated cases) and 17 patients in the validation set (4 well differentiated cases and 13 non-well differentiated cases). A total of 44 selected features were predictive of the pathological grade of PDAC among 20 feature groups. There was significant difference among the volumes of ROI_M1, ROI_M2, ROI_G and ROI_S (10.29(4.01, 19.43), 9.34(4.26, 17.27), 11.86(5.52, 19.74) and 15.08(9.62, 27.44) cmn 3; n H=18.641, n P0.05). The accuracy of ROI_M2 was better than ROI_G (n z=3.031, n P=0.002), but the generalization ability of ROI_M2 was insufficient (n t=3.086, n P=0.012).n Conclusions:Although the manual contour prediction models are highly accurate, their performance are unstable. Semi-automatic contouring based on gradient can achieve comparable accuracy to manual contouring, and the model′s generalization ability is stronger.