CT影像组学联合临床及CT特征预测胸腺上皮肿瘤TNM分期

CT radiomics combined with clinical and CT features for predicting TNM stage of thymic epithelial tumor

目的观察CT影像组学联合临床及CT特征预测胸腺上皮肿瘤(TET)TNM分期的价值。方法回顾性分析216例经手术病理证实的单发TET患者,以其中151例TNMⅠ期为早期组,将27例TNMⅢ期及38例Ⅳ期归为晚期组(n=65)。采用单因素分析组间临床资料及胸部CT表现。分别基于平扫(NECT)及增强CT(CECT)提取并筛选最佳影像组学特征,建立预测TET TNM分期的影像组学模型RM_(NECT)、RM_(CECT),联合组间差异有统计学意义的临床及CT特征构建RM_(NECT-临床)、RM_(CECT-临床)及RM_(NECT-临床-CT)、RM_(CECT-临床-CT)。按7∶3比例将患者分为训练集(n=151)及验证集(n=65),采用重复5折交叉验证法于训练集训练模型,并于验证集验证其效能。结果组间临床症状及CT所示病灶周围脂肪浸润、纵隔淋巴结肿大、胸腔积液差异均有统计学意义(P均<0.05)。分别基于NECT及CECT筛选出2个及9个最佳影像组学特征,以之构建相应模型。验证集中,RM_(NECT-临床)-CT预测TET TNM分期的AUC(0.864)高于RM_(NECT)及RM_(NECT-临床)(AUC=0.634、0.721,Z=3.081、2.937,P=0.002、0.003),RM_(CECT-临床-CT)的AUC(0.920)高于RM_(CECT)及RM_(CECT-临床)(AUC=0.689、0.751,Z=2.698、2.390,P=0.007、0.017)。结论CT影像组学联合临床及CT特征能有效预测TET TNM分期。

Objective To explore the value of CT radiomics combined with clinical data and CT features for predicting TNM stage of thymic epithelial tumor(TET).Methods Data of 216 single TET patients confirmed by surgical pathology were retrospectively analyzed.Totally 151 cases with TNM stageⅠTET were divided into early group,while 27 with TNM stageⅢand 38 with TNM stageⅣTET were divided into late group(n=65).Univariate analysis was used to analyze clinical data and chest CT manifestations.Based on non-contrast-enhanced CT(NECT)and contrastenhanced CT(CECT),the best radiomics features were extracted and screened to establish radiomics models(RM_(NECT),RM_(CECT))for predicting TNM stage of TET.RM_(NECT-clinic),RM_(CECT-clinic),RM_(NECT-clinic)-CT and RM_(CECT-clinic-CT)were constructed based on combination of clinical and CT features being significantly different between groups,respectively.The patients were divided into training set(n=151)and validation set(n=65)at the ratio of 7∶3.The above models were trained in the training set using repeated 5-fold cross validation method,and their efficacy were verified in the validation set.Results Significant differences of clinical symptoms and CT manifestations including fat infiltration around the lesion,mediastinal lymph node enlargement and pleural effusion were found between groups(all P<0.05).Based on NECT and CECT,2 and 9 best radiomics features were selected to construct the corresponding models.In validation set,the area under the curve(AUC)of RM_(NECT-clinic-CT)for predicting TNM stage of TET(0.864)was higher than that of RM_(NECT)and RM_(NECT-clinic)(AUC=0.634,0.721,Z=3.081,2.937,P=0.002,0.003),while AUC of RM_(CECT-clinic)-CT(0.920)was also higher than that of RM_(CECT)and RM_(CECT-clinic)(AUC=0.689,0.751,Z=2.698,2.390,P=0.007,0.017).Conclusion CT radiomics combined with clinical data and CT features could effectively predict TNM stage of TET.