摘要
目的探讨基于治疗前CT影像组学模型预测肺癌患者化疗早期疗效的效能。方法回顾性搜集2017年1月至2022年1月202例确诊肺癌患者的临床影像资料,通过对比化疗半年后的CT图像与化疗前CT图像,根据实体肿瘤疗效评价标准(RECIST 1.1)分为部分缓解(PR)组100例和疾病进展(PD)组102例,将入组患者的胸部CT平扫DICOM文件导入开源图像分割软件ITK-SNAP软件,进行感兴趣区(ROI)的勾画,再进行影像学特征提取,按照7∶3的比例随机分配为训练组和验证组,采用LASSO法进行降维及构建影像组学模型,最终根据受试者工作特征曲线(ROC)的曲线下面积(AUC)、敏感度、特异度来评价模型的预测效能。结果训练组的AUC值、特异度、敏感度分别为0.737(95%CI:0.656~0.819)、0.71、0.67;验证组的AUC值、特异度、敏感度分别为0.722(95%CI:0.594~0.851)、0.77、0.53,模型达到了良好的预测效能。结论基于治疗前CT的影像组学模型可在化疗前对肺癌的早期疗效做出较准确预测。
Objective To investigate the application value of radiomics in predicting early efficacy of chemotherapy in patients with lung cancer.Methods A total of 202 patients with lung cancer were selected which were pathologically confirmed from January 2017 to January 2022,the patients were divided into two groups:partial response(PR)(n=100)and progressive disease(PD)(n=102)according to RECIST 1.1 criteria.The chest CT DICOM files of the enrolled patients were imported into the open source image segmentation software ITK-SNAP for ROI delineation,and then the radiomics features were extracted.According to the ratio of 7∶3,they were randomly assigned to the training group and the test group and establish radiomics models,and finally evaluate the diagnostic and predictive efficacy of model according to the area under the receiver operating characteristic curve(AUC),sensitivity,and specificity.Results The AUC value,specificity and sensitivity of the training group were 0.737(95%CI:0.656-0.819),0.71 and 0.67;The AUC value,specificity and sensitivity of the validation group were 0.722(95%CI:0.594-0.851),0.77 and 0.53.The model achieved good clinical efficacy.Conclusion CT-based radiomics has a high diagnostic value for predicting early efficacy of chemotherapy in patients with lung cancer.
作者
张楚
吴少虹
段绍峰
郭莉莉
ZHANG Chu;WU Shaohong;DUAN Shaofeng(Department of Radiology,Huai'an First Hospital Affiliated to Nanjing Medical University,Huai'an,Jiangxsu Province 223300,P.R.China)
出处
《临床放射学杂志》
北大核心
2023年第12期1895-1899,共5页
Journal of Clinical Radiology
关键词
肺癌
化疗疗效
影像组学
体层摄影术
X线计算机
Lung cancer
Efficacy of chemotherapy
Radiomics Tomography
X-ray computed
