基于T2WI和DWI影像组学在术前预测垂体腺瘤质地中的应用价值

Application value of radiomics based on T2WI and DWI in preoperative prediction of pituitary adenoma consistency

目的:探讨基于T2WI和DWI影像组学在术前无创预测垂体腺瘤质地中的应用价值。方法:回顾性分析经病理证实为垂体腺瘤的108例患者临床及术前MRI资料,术中2名神经外科医生评估肿瘤质地,将其分为质软组和质硬组。按7∶3随机分为训练组和验证组,在T2WI和DWI图像上手动勾画肿瘤实质区的体积作为感兴趣区容积(volume of interest,VOI),用FeAture Explorer软件提取特征,采用无监督特征选择(unsupervised feature selection,UFS)进行特征筛选,采用支持向量机(support vector machine,SVM)构建影像组学模型。通过受试者工作特征曲线下面积(area under curve,AUC)及校准曲线评估模型的效能。结果:在联合T2WI和DWI影像组学模型中,训练组预测垂体腺瘤质地的AUC为0.89,验证组的AUC为0.80。校准曲线显示模型预测值与实际值一致性较好。结论:联合T2WI和DWI影像组学模型具有较好的诊断效能,有助于术前预测垂体腺瘤的质地。

Objective:To explore the application value of radiomics based on T2⁃weighted imaging(T2WI)and diffusion⁃weighted imaging(DWI)in non⁃invasive preoperative prediction of pituitary adenoma consistency.Methods:The clinical and preoperative MRI data of 108 patients with pathologically confirmed pituitary adenoma were retrospectively analyzed.Two neurosurgeons evaluated tumor consistency intraoperatively and categorized them into soft and hard groups.Patients were randomly divided into a training set and a validation set in a 7∶3 ratio.Volume of interest(VOI)representing the tumor solid component were manually delineated on T2WI and DWI images.Radiomics features were extracted by FeAture Explorer software.Unsupervised feature selection(UFS)was applied for feature selection.Support vector machine(SVM)was used to conduct the radiomics models.Area under curve(AUC)and calibration curve were used to assess the performance of the models.Results:In the combined T2WI and DWI radiomics model,the AUC for predicting the consistency of pituitary adenoma was 0.89 in the training set and 0.80 in the validation set.The calibration curve showed a good consistency between predicted and actual values.Conclusion:The combined T2WI and DWI radiomics model demonstrates good diagnostic performance and aids in preoperative prediction of the consistency of pituitary adenoma.