基于两肾CT平扫直方图分析预测血尿素氮价值的研究  

Research on Predicting Blood Urea Nitrogen Value Based on Histogram Analysis of Two Kidney CT Plain Scan

作　　者：韩爱萍 谭伟[1,2] 魏毅 胡守章[1,2] 涂武超 HA Ai-ping;TAN Wei;WE Yi;HU Shou-zhang;Wu-chao(Wuhan University of Science and technology,Wuhan 430070,Hubei Province,China;radiology Department of Geriatric Hospital Affiliated to Wuhan University of Science and technology,Wuhan 430070,Hubei Province,China)

机构地区：[1]武汉科技大学,湖北武汉430070 [2]武汉科技大学附属老年病医院放射科,湖北武汉430070

出　　处：《中国CT和MRI杂志》2025年第2期137-139,143,共4页Chinese Journal of CT and MRI

基　　金：湖北省教育厅科学研究计划青年项目(Q20221114)。

摘　　要：目的评估两肾CT平扫直方图参数预测血尿素氮(BUN)异常的可行性。方法采用FireVoxel软件,对武汉科技大学附属老年病医院69名两肾CT平扫患者(34名BUN异常,35名正常)的图像进行直方图分析。分析包括信号最小值、最大值、平均值、中位数、标准差、变异系数、偏度、峰度、熵及百分位值(1%,5%,10%,25%,50%,75%,90%,95%,99%)。利用接收者操作特征(ROC)曲线评估这些参数在区分BUN异常与正常水平上的诊断价值。结果Kruskal-Wallis检验显示信号标准差、变异系数、偏度、峰度及1%百分位值在正常与异常BUN水平患者间存在显著差异(P<0.05)。ROC曲线分析表明,信号标准差、变异系数及峰度的AUC分别为0.713、0.663、0.667。变异系数联合峰度、标准差联合峰度、标准差联合变异系数及三者联合的AUC值在0.68至0.734之间,表明组合使用能增强预测准确性。结论基于两肾CT平扫的直方图参数对血尿素氮异常具有一定的预测价值。Objective This study assesses the potential of using histogram parameters from non-contrast bilateral kidney CT scans to predict blood urea nitrogen(BUN)abnormalities.Methods We conducted a histogram analysis using FireVoxel software on CT images from 69 patients at the Geriatric Hospital affiliated with Wuhan University of Science and Technology,divided into 34 with abnormal BUN levels and 35 with normal levels.The analysis focused on various signal parameters,including minimum,maximum,average,median,standard deviation,coefficient of variation,skewness,kurtosis,entropy,and percentile values(1%,5%,10%,25%,50%,75%,90%,95%,99%).Receiver Operating Characteristic(ROC)curves evaluated the diagnostic efficacy of these parameters.Results The Kruskal-Wallis test showed significant distinctions in signal standard deviation,coefficient of variation,skewness,kurtosis,and the 1%percentile value between the normal and abnormal BUN groups(P<0.05).ROC analysis revealed AUCs of 0.713,0.663,and 0.667 for signal standard deviation,coefficient of variation,and kurtosis,respectively.Combined AUC values for these parameters ranged from 0.68 to 0.734,indicating improved predictive accuracy when used together.Conclusion Histogram parameters from unenhanced kidney CT scans demonstrate significant predictive value for BUN abnormalities,highlighting their utility as diagnostic tools.

关 键 词：Computed Tomography(CT) 肾脏 直方图分析 血尿素氮 

分 类 号：R334.1[医药卫生—人体生理学]