能谱衰减曲线预测肺癌Ki-67表达:评估最佳keV及能谱曲线斜率

Energy Spectrum Attenuation Curve for Prediction of Ki-67 Expression in Lung Cancer:to Evaluate the Optimal keV Sequence and Slope of Energy Spectrum Curves

目的探讨预测肺癌细胞增殖抗原标记物(Ki-67)表达的能谱CT最佳keV,以及不同能谱曲线斜率对预测效能的影响。方法回顾性分析43例原发性肺癌患者的CT数据,采用Siemens双源CT机型,完成动脉期和静脉期(峰值期)的能谱双期增强扫描。按Ki-67表达将患者分为低风险组(Ki-67≤30%)和高风险组(Ki-67>30%),比较40~140 keV不同单能量图像区分两组患者的效能并获取最佳单能量图像。比较三组不同能谱曲线斜率(λ40~100,λ100~140和λ40~140)对预测Ki-67效能的影响。采用SPSS 22.0软件进行统计分析,P<0.05设为显著性差异值。结果在虚拟单光子能量图像中,动脉期40、50 keV及静脉期40、60、70 keV对应的CT值在Ki-67高、低表达组间有显著差异(P<0.05),其中静脉期40 keV是最佳单能量图像。动脉期三组能谱曲线斜率在Ki-67高、低表达组间均有显著性统计学差异(P<0.01),静脉期差异更显著(P=0.000,AUC分别为0.859、0.856、0.859)。结论能谱CT增强扫描静脉期40 keV单能量序列是预测肺癌Ki-67表达的最佳单能量序列,而能谱曲线斜率的不同未发现对预测结果产生明显偏倚。

Objective To explore the optimal keV for prediction of Ki-67 expression in lung cancer by energy CT,and the influence of the slope of different energy spectrum curve on the prediction efficiency.Methods the CT data of 43 patients with primary lung cancer were included in the retrospective analysis,and the dual-phase enhanced scans of arterial phase and venous phase(peak phase)were performed using Siemens Dual-Energy CT.According to the expression of Ki-67,patients were divided into the low-risk group(Ki-67 median≤30%)and the high-risk group(Ki-67>30%),and the efficacy of the two groups was distinguished by comparing the 40-140 keV single-energy images and the best single-energy images were obtained.The effects of the slopes of three different energy spectrum curves(λHU40-100,λHU100-140 andλHU40-140)on the predicted Ki-67 performance were compared.SPSS 22.0 software was used for statistical analysis,and P<0.05 was considered statistically significant.Results In the virtual single-photon energy images,the CT values corresponding to 40 and 50 keV in the arterial phase and 40,60 and 70 keV in the venous phase were significantly different between the two Ki-67 groups(P<0.05),of which 40 keV in the venous phase was the best single-energy image.There were statistically significant differences in the slopes of energy spectrum curves in the three groups at the arterial phase between the two Ki-67 groups(P<0.01),and even more significant differences at the venous phase(P=0.000,AUC=0.859,0.856,and 0.859,respectively).Conclusion The single energy sequence of 40 keV in venous phase enhanced by energy spectrum CT is the best single energy sequence to predict the expression of Ki-67 in lung cancer.