CT影像组学预测磨玻璃结节肺腺癌浸润性的研究

Predicting the Invasiveness of Ground Glass Nodule Lung Adenocarcinoma Based on CT Radiomics Model

目的 探讨CT影像组学预测磨玻璃结节肺腺癌浸润性的价值。方法 搜集南通大学附属医院2021年1月至2022年6月经病理证实为非典型腺瘤样增生、原位腺癌、微浸润性腺癌和浸润性腺癌的患者资料,CT表现为磨玻璃结节且最大径≤3 cm,其中非典型腺瘤样增生和原位腺癌为非浸润组,微浸润性腺癌和浸润性腺癌为浸润组。共295例患者纳入研究,非浸润组86例,浸润组209例。按照7∶3比例分为训练集和验证集,用卡方检验、独立样本t检验及非参数秩和检验评价非浸润组和浸润组的临床、影像特征差异。利用3D-slicer软件勾画VOI、提取影像组学特征,利用R语言软件进行特征筛选、建立模型。绘制受试者工作特征曲线(ROC),用曲线下面积(AUC)、敏感度、特异度评估模型的预测性能。绘制临床决策曲线(DCA)评估模型的临床价值。结果 年龄、支气管充气征、最大截面长径、实性成分长径、肿瘤实变率(CTR)在非浸润组和浸润组中具有统计学差异(P<0.05)。多因素Logistic回归分析显示,最大截面长径是肺腺癌浸润性的独立预测因子(P<0.05),最佳截断值为11.545 mm。CT影像组学共提取851个特征,最终纳入11个特征用于构建模型,CT影像组学模型在训练集和验证集中的AUC分别为0.902和0.93。结论 基于CT影像组学建立的模型在预测肺腺癌浸润性中具有较好的效能,有助于临床决定随访管理、手术时机和手术方式等。

Objective To evaluate the value of CT radiomics in predicting the invasion of pulmonary adenocarcinoma with ground glass nodules.Methods The data of patients with atypical adenomatous hyperplasia,adenocarcinoma in situ,minimally invasive adenocarcinoma and invasive adenocarcinoma confirmed by pathology in the Affiliated Hospital of Nantong University from January 2021 to June 2022 were collected.The CT imaging findings were ground glass nodules with a maximum diameter of ≤3 cm.Among them,atypical adenomatous hyperplasia and adenocarcinoma in situ were non-invasive groups,and minimally invasive adenocarcinoma and invasive adenocarcinoma were invasive groups.A total of 295 patients were included in the study,86 in the non-invasive group and 209 in the invasive group.The training set and validation set were divided according to the ratio of 7:3.The differences of clinical and imaging data between the non-invasive groups and invasive groups were evaluated by chi square test,independent sample t-test and nonparametric rank sum test.3D slicer software is used to sketch VOI,extract image omics features,and R language software is used to screen features and establish models.Draw the ROC curve,evaluate the predictive performance of the model with the area under the curve(AUC) sensitivity and specificity,and evaluate the clinical value of the model with the clinical decision curve analysis(DCA).Results Age,air bronchogran,maximum section length diameter,solid component length and tumor consolidation rate(CTR)were statistically different between the non-invasive group and the invasive group(P<0.05).Multivariate Logistic regression analysis showed that maximum section length diameter was an independent predictor of invasive lung adenocarcinoma(P <0.05).The optimal cut-off value was 11.545 mm.A total of 851 features were extracted from CT radiomics,and finally11 features were included to construct the model.The AUC of CT radiomics model in training set and verification set were0.902 and 0.93 respectively.Conclusion The model based on CT radiomics has a good efficacy in predicting the invasion of lung adenocarcinoma,and is helpful for clinical decision on follow-up management,surgical timing and surgical methods.