基于治疗前MR-DWI影像组学预测肺癌化疗疗效的初步研究

发布时间：2018-06-24 14:48

本文选题：肺肿瘤 + 扩散加权成像　；参考：《放射学实践》2017年12期

【摘要】：目的:探讨基于磁共振扩散加权成像(DWI)的影像组学对肺癌化疗疗效的预测价值。方法:回顾性搜集连续30例经病理证实的肺癌患者的病例资料,根据第二周期化疗后肿瘤最大径退缩率、按RECIST标准将患者分为治疗有效组(16例)和无效组(14例)。提取所有患者的化疗前ADC图像(b=600、800和1000s/mm2),应用影像组学方法,在每种b值的ADC图像上提取病灶的19985个特征,采用Lasso进行降维和建模。采用受试者工作特征曲线(ROC)计算三种b值模型预测化疗疗效的诊断效能,并采用DeLong检验比较三种曲线的曲线下面积(AUC)。结果:30例中有效组16例、无效组14例。基于b=600s/mm2的ADC图像的影像组学特征所建立的模型,其AUC、诊断敏感度和特异性分别为0.875、0.895和0.750;基于b=800s/mm2的ADC图像,其相应的AUC、诊断敏感度和特异度分别为0.924、0.947和0.938,基于b=1000s/mm2的ADC图像,相应的AUC、诊断敏感度和特异度分别为0.918、0.895和0.875。三种b值的AUC差异无统计学意义(P0.05)。结论:基于MR-DWI的影像组学可在治疗前对肺癌化疗疗效作出准确预测。
[Abstract]:Objective: to evaluate the predictive value of Mr diffusion weighted imaging (DWI) in predicting the chemotherapeutic effect of lung cancer. Methods: the data of 30 consecutive cases of lung cancer confirmed by pathology were retrospectively collected. According to the maximum diameter withdrawal rate of tumor after the second cycle chemotherapy, the patients were divided into effective treatment group (16 cases) and ineffective group (14 cases) according to RECIST standard. The ADC images (bp600800 and 1000s/mm2) of all patients before chemotherapy were extracted. The 19985 features of the lesions were extracted from each B-value ADC image by using the image group method, and the reduction and modeling were performed by Lasso. The diagnostic efficacy of three b value models was calculated by using the operating characteristic curve (ROC), and the area under the curve (AUC) of the three curves was compared by DeLong test. Results there were 16 cases in effective group and 14 cases in ineffective group in 30 cases. The diagnostic sensitivity and specificity of ADCs based on b=600s/mm2 images were 0.895 and 0.750, respectively, and the diagnostic sensitivity and specificity of ADCs images based on b=800s/mm2 were 0.924, 0.947 and 0.938, respectively, and the diagnostic sensitivity and specificity of ADCs based on ADCs were 0.8750.85 and 0.750, respectively, and the diagnostic sensitivity and specificity of ADCs based on ADCs were 0.924 and 0.938, respectively, and the diagnostic sensitivity and specificity of ADCs were 0.924 and 0.938, respectively. The diagnostic sensitivity and specificity of AUC were 0.918 卤0.895 and 0.875 respectively. There was no significant difference in AUC among the three b values (P0.05). Conclusion: MR-DWI imaging can accurately predict the chemotherapeutic effect of lung cancer before treatment.
【作者单位】：昆明医科大学第三附属医院云南省肿瘤医院放射科;
【基金】：云南省应用基础研究(昆医联合专项2017FE467-084)
【分类号】：R445.2;R734.2

【相似文献】