摘要
【目的】探讨胸腔积液量与胸腔积液CT横断面上最大深度、矢状面最大纵向累及长度之间的关系,建立多重线性回归模型估算胸腔积液量。【方法】收集151名18岁以上游离性胸腔积液患者的CT检查资料,共166个胸腔积液。测算胸腔积液体积V、胸腔积液CT横断面上最大深度H、矢状面最大纵向累及长度L,随机抽取70%的样本作为训练样本,剩下的为验证样本量。以经对数转换后的胸腔积液体积V为因变量,其余测量指标为自变量,采用多重线性回归方法建立预测模型。【结果】所得线性回归方程为:ln(V)=0.01688H+0.00578L+4.19738,决定系数R2=0.929。由配对设计资料的t检验可得,在验证样本中V测量值与预测值的差值(X=0.203,95%CI(-0.181,0.587),P=0.883)与0没有统计学意义。V测量值与预测值之间的散点图提示两者之间呈高度的线性相关关系(r=0.962,P<0.01)。【结论】在CT图像上,用胸腔积液CT横断面上最大深度H、矢状面最大纵向累及长度L这两个简单、容易测量的指标可以准确地预测胸腔积液量。
[Objective] To explore the relationship among the pleural effusion volume, the maximum depth on CT cross-section and the maximum longitudinal cumulative length of the pleura[ effusion on sagittal plane, and to forecast pleural effusion volume by establishing a multiple linear regression. [Methods] 151 patients, with 166 cases, who were more than 18 years old, and suffered from free pleural effusion examined by CT were collected. The volume of pleural effusion, the maximum depth on cross-section and the maximum longitudinal cumulative length on sagittal plane of pleural effusion were measured. Multiple linear regression was conducted with natural logarithmation of the response variable (V: pleural effusion volume) and two covariates (H: the maximum depth on CT cross-sectional plane; L : the maximum longitudinal cumulative length on sagittal plane). 70% of cases were randomly selected as training set to establish the equation, the rest of cases were selected as validation set. [ Results ] The multiple linear regression : In (I~') = 0.01688H + 0.00578L + 4.19738 ,R^2 = 0.929. The difference(X) between predicted and observed pleural effusion volume was not statistically significant, X-=0.203,95%CI (-0.181,0.587) ,P = 0.883. The scatter diagram between predictive and observed pleural effusion volume showed there was a high linear correlation (r = 0.962, P 〈 0.01 ). [ Conclusion ] The volume of pleural effusion could be accurately predicted and easily measured by using H (the maximum depth on cross-section)and L (the maximum longitudinal cumulative length on sagittal plane of pleural effusion) on CT image.
作者
严超贵
霍梦娟
范淼
王俊丽
杨有优
YAN Chao-gui HUO Meng-juan FAN Miao WANG Jun-li YANG You-you(Department of Radiology, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou 510080, China Department of Radiology, Guangdong Provincial Hospital of Traditional Chinese Medical, Guangzhou 510080, China Department of Radiology, The First Affiliated Hospital ofZhe Jiang University, Hangzhou 310003, China)
出处
《中山大学学报(医学科学版)》
CAS
CSCD
北大核心
2016年第5期781-785,共5页
Journal of Sun Yat-Sen University:Medical Sciences
关键词
CT
胸腔积液
tomography, X-ray computed
pleural effusion