目的:基于术前CT构建预测肝门部胆管癌(pCCA)神经侵犯(PNI)的影像组学模型,并评价其效能.方法:回顾性分析本院 2013 年 2 月-2021 年 2 月 149 例经病确诊的 pCCA 患者的临床资料,其中PNI组患者 108 例,无PNI组患者 41 例.采用R语言将所有患者按 3:1 比例随机分为训练集和验证集.在静脉期图像上,沿肿瘤边缘在所有层面上手动勾画 3D感兴趣区(ROI),使用 3D Slicer提取影像组学特征.采用组内相关系数(ICC)、相关性分析去除冗余特征,采用随机森林算法(RF)对所有临床、影像组学特征进行重要性排序,并选取前 18 个重要特征构建 RF 模型.使用准确性、敏感性、特异性及受试者操作特征(ROC)曲线评价模型效能.结果:在训练集中,RF 模型的准确性、敏感性、特异性均为 100%,ROC 曲线下面积 AUC 为 1;在验证集中,RF 模型的准确性为 70.3%,敏感性为59.3%,特异性为 100%,AUC 为 0.846(0.713～0.979).结论:基于增强 CT 图像建立的影像组学模型可用于术前无创性预测pCCA患者的PNI状态.

目的 探讨微卫星高度不稳定型(MSI-H)胃癌的CT及临床病理学特征.方法 回顾性分析201例胃癌患者的临床资料,包括33例MSI-H,168例微卫星稳定型(MSS)/微卫星低度不稳定型(MSI-L).统计分析临床、CT及病理学特征与微卫星不稳定型(MSI)状态的相关性.结果 MSI-H组的性别(P=0.018)、肿瘤长径(P＜0.001)、肿瘤部位(P＜0.001)、程序性细胞死亡配体1(PD-L1)表达(P=0.007)与MSS/MSI-L组相比有显著统计学差异,单因素和多因素logistic回归分析显示性别(女性)、肿瘤部位(胃窦)、PD-L1阳性及肿瘤长径为MSI-H的独立预测因子.结论 性别(女性)、肿瘤部位(胃窦)、PD-L1阳性及肿瘤长径可作为MSI-H的独立预测因子,用以MSI-H胃癌的初始筛查.

To develop a CT-based radiomics model for preoperative prediction of lymph node (LN) metastasis in perihilar cholangiocarcinoma (pCCA). The study enrolled consecutive pCCA patients from three independent Chinese medical centers. The Boruta algorithm was applied to build the radiomics signature for the primary tumor and LN. The k-means algorithm was employed to cluster the selected LNs based on the radiomics signature LN. Support vector machines were used to construct the prediction models. The diagnostic efficiency was measured by the area under the receiver operating characteristic curve (AUC). The optimal model was evaluated in terms of calibration, clinical usefulness, and prognostic value. A total of 214 patients were included in the study (mean age: 61.6 years ± 9.4; 130 male). The selected LNs were classified into two clusters, which were significantly correlated with LN metastasis in all cohorts (p < 0.001). The model incorporated the clinical risk factors, radiomics signature primary tumor, and the LN cluster obtained the best discrimination, with AUC values of 0.981 (95 • The radiomics model based on contrast-enhanced CT is a useful tool for preoperative prediction of lymph node metastasis in perihilar cholangiocarcinoma. • Radiomics features extracted from lymph nodes show great potential for predicting lymph node metastasis. • The study is the first to identify a lymph node phenotype with a high probability of metastasis based on radiomics.

Background: Preoperative assessment of lymph node (LN) status is crucial for the optimal management of perihilar cholangiocarcinoma (pCCA).Methods: This retrospective study enrolled consecutive pCCA patients from three Chinese centers. Logistic regression was employed to identify clinical risk factors. The Boruta algorithm was applied to build radiomics signatures of the tumor and the selected LNs. Clustering was performed using the k-means algorithm based on the radiomics signature for the selected LNs. The prediction models were constructed using support vector machines. The diagnostic efficiency was quantified with the area under the receiver operating characteristic curve (AUC). The optimal model achieving the best AUC was further evaluated in calibration, clinical usefulness, and prognostic value.Findings: 214 patients were included in the study (61.6 years ± 9.4; 130 men). The selected LNs were classified into two clusters, which was significantly correlated with LN metastasis in all cohorts (P < 0.001). The model incorporated the clinical risk factors, the tumor radiomics, and the LN cluster obtained the best discrimination with AUC values of 0.981 (95% CI: 0.962, 1), 0.896 (95% CI: 0.810, 0.982), and 0.865 (95% CI: 0.768–0.961) in the training, internal validation, and external validation cohorts, respectively. Patients with a high risk of LN metastasis, predicted by the optimal model, had shorter overall survival than low-risk patients (median, 13.7 vs 27.3 months, P < 0.001).Interpretation: This study proposed a radiomics model with good performance to predict LN metastasis in pCCA. The radiomics model could inform treatment options and improve patient care.Funding Information: Supported by the Key Scientific Research Project of Higher Education in Henan Province (No.22A320057).Declaration of Interests: We declare no conflicts of interest.Ethics Approval Statement: This multicenter retrospective study was approved by the Ethics Committee (2021-KY-0778-001), and written informed consent was waived.