| 田琪,冯峰,陈巧玲,秦彩,周慧,李曼曼,邢金丽.CT影像组学列线图评估非小细胞肺癌程序性死亡受体1表达[J].中国医学影像技术,2023,39(4):543~548 |
| CT影像组学列线图评估非小细胞肺癌程序性死亡受体1表达 |
| CT radiomics nomogram for evaluating programmed death receptor 1 expression of non-small cell lung cancer |
| 投稿时间：2022-11-24 修订日期：2023-01-06 |
| DOI：10.13929/j.issn.1003-3289.2023.04.013 |
| 中文关键词: 癌,非小细胞肺 影像组学 列线图 程序性细胞死亡1受体 体层摄影术,X线计算机 |
| 英文关键词:carcinoma, non-small-cell lung radiomics nomogram programmed cell death 1 receptor tomography, X-ray computed |
| 基金项目:南通市科技项目(MS22021047)。 |
|
| 摘要点击次数: 1548 |
| 全文下载次数: 815 |
| 中文摘要: |
| 目的 观察CT影像组学列线图评估非小细胞肺癌(NSCLC)程序性死亡受体1(PD-1)表达的价值。方法 纳入143例NSCLC患者,其中PD-1阳性30例、阴性113例;按7 ∶ 3比例将其分为训练集(n=101)和验证集(n=42),比较PD-1阳性与阴性患者临床资料差异,以logistic回归分析筛选临床因素,构建临床模型;基于CT提取并筛选影像组学特征,建立影像组学模型;结合临床因素及影像组学特征构建CT影像组学列线图,分析各模型评估PD-1表达的效能。结果 针对训练集及验证集,临床模型评估NSCLC PD-1表达的曲线下面积(AUC)分别为0.79和0.74,影像组学模型的AUC分别为0.89和0.81,CT影像组学列线图的AUC分别为0.92及0.86。DeLong检验结果显示,仅临床模型与CT影像组学列线图评估训练集NSCLC PD-1表达的AUC差异有统计学意义(Z=2.47,P=0.01),其余AUC两两比较差异均无统计学意义(P均＞0.05)。结论 CT影像组学列线图有助于评估NSCLC PD-1表达。 |
| 英文摘要: |
| Objective To observe the value of CT radiomics nomogram for evaluating programmed death receptor 1 (PD-1) expression of non-small cell lung cancer (NSCLC). Methods A total of 143 patients with NSCLC were enrolled, including 30 PD-1 positive and 113 of PD-1 negative cases, and were divided into training set (n=101) and validation set (n=42) at the ratio of 7 ∶ 3. Clinical data were compared between PD-1 positive and negative patients, and logistic regression analysis was performed to screen clinical factors and establish clinical model. Radiomics features were extracted and screened based on CT, and radiomics model was established. Finally, CT radiomics nomogram was established combining clinical factors and radiomics features. The value of each model for evaluating PD-1 expression was analyzed. Results In training set and validation set, the area under the curve (AUC) of clinical model for evaluating PD-1 expression of NSCLC was 0.79 and 0.74, respectively, of radiomics model was 0.89 and 0.81, while of CT radiomics nomogram was 0.92 and 0.86, respectively. DeLong test results showed that significant difference of AUC existed only between clinical model and CT radiomics nomogram in training set (Z=2.47, P=0.01), while no significant difference was found among other AUC (all P＞0.05). Conclusion CT radiomics nomogram was helpful to evaluating PD-1 expression of NSCLC. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
