MRI图像纹理特征在评估宫颈癌新辅助化疗中的预测作用  

Application of MRI texture analysis in evaluating the neoadjuvant chemotherapy response in uterine cervix carcinoma

作　　者：张铮铮 刘筱 徐浩 杜文升 陆晓媛 梁婧玲 ZHANG Zhengzheng;LIU Xiao;XU Hao;DU Wensheng;LU Xiaoyuan;LIANG Jingling(Department of Obstetrics and Gynecology,the Affiliated Hospital of Xuzhou Medical University,Xuzhou,Jiangsu 221002,China;Department of Emergency,the Affiliated Hospital of Xuzhou Medical University;Department of Radiology,Jiangsu Province Hospital of Traditional Chinese Medicine,the Affiliated Hospital of Nanjing University of Traditional Chinese Medicine,Nangjing,Jiangsu 210029)

机构地区：[1]徐州医科大学附属医院妇产科,江苏徐州221002 [2]徐州医科大学附属医院急救中心 [3]江苏省中医院南京中医药大学附属医院影像科,江苏南京210029

出　　处：《徐州医科大学学报》2018年第9期592-595,共4页Journal of Xuzhou Medical University

基　　金：国家自然科学基金（81371243）;徐州市科技创新后补助备案项目（KH17045）

摘　　要：目的探讨MRI图像纹理特征在评估宫颈癌新辅助化疗疗效中的预测作用。方法回顾性分析25例ⅠB2-ⅡB早期宫颈癌患者的临床资料。术前均采用新辅助化疗，化疗前行MRI图像纹理分析，获得均数、熵值、偏度、峰度和不均匀度等参数。依据WHO制定的实体瘤客观疗效评定标准，将宫颈癌患者分为完全缓解组（CR）、部分缓解组（PR）和进展组（DR）。采用单因素方差分析对3组患者MRI图像纹理特征参数进行比较，评估上述参数在宫颈癌新辅助化疗疗效中的预测作用。结果25例宫颈癌患者中，CR5例，PR17例，DR3例。3组患者中，均数、熵值和不均匀度参数相比差异有统计学意义（P〈0．05），且PD组中均数、熵值及不均匀度明显高于CR和PR组（P〈0．05）；而3组间偏度、峰度值比较差异无统计学意义（P〉0．05）。结论MRI纹理分析可以有效评估宫颈癌新辅助化疗的预后。Objective To evaluate the feasibility of MRI texture analysis in predicting neoadjuvant chemotherapy re- sponse in patients with uterine cervix carcinoma. Methods A total of 25 uterine cervix carcinoma patients （ stage I B2 - IIA2 earlier） adopted MRI texture analysis and the neoadjuvant chemotherapy preoperatively were retrospectively ana- lyzed. The parameters of mean, entropy, skewness, kurtosis and inhomogeneity were obtained. The patients were divided into three groups： complete response（ CR）, partial response（PR） and deteriorative response（DR） according to the stand- ard of solid tumor diagnosis by WHO. The parameters of MRI texture analysis among the three groups were compared u- sing one - way ANOVA and the predicting effect of the neoadjuvant chemotherapy in uterine cervix carcinoma was evalua- ted. Results There were 5 patients in CR group, 17 patients in PR group and 3 patients in DR group. There were statis- tical differences in mean, entropy, and inhomogeneity among the three groups （P 〈 0.05 ）. At the same time, the value of mean, entropy and inhomogeneity was higher in DR group than in other two groups （ P 〈 0.05 ）. However, there was no statistical difference in skewness and kurtosis among the three groups （ P 〉 0.05 ）. Conclusions MRI texture analysis may have the effect of predicting the prognosis of the neoadjuvant chemotherapy in uterine cervix carcinoma.

关 键 词：宫颈癌 MRI图像纹理分析 新辅助化疗 预后 

分 类 号：R711.74[医药卫生—妇产科学]