光子计数探测器CT纯净钙化算法评估冠状动脉钙化积分的最优重建参数研究

Optimizing reconstruction parameters for coronary artery calcium scoring with the purecalcium algorithm on photon-counting detector CT:a preliminary study

目的 基于光子计数探测器CT(PCD-CT)纯净钙化(PureCalcium)虚拟非碘算法在不同的层厚、keV和量子迭代重建(QIR)参数下测得冠状动脉钙化积分(CACS),评估其定量的准确性并探讨最优的重建参数。方法前瞻性纳入行PCD-CT CCTA检查的病人64例。以常规真实平扫(TNC)影像获得的CACS_(TNC)为参考标准,采用不同参数[层厚(3 mm、1.5 mm),keV(55~75)和QIR(1、4)]重建得到_(PureCalcium)影像并计算CACS_(PureCalcium)。使用Wilcoxon符号秩检验比较CACS_(PureCalcium)与CACS_(TNC),采用Spearman相关系数、组内相关系数(ICC)和Bland-Altman检验对CACS_(PureCalcium)与CACS_(TNC)进行相关性分析和一致性检验。结果 采用层厚3 mm,55/60/65 keV+QIR 1/4、70 keV+QIR 1以及采用层厚1.5 mm,55 keV+QIR 1/4和60 keV+QIR 1获得的CACS_(PureCalcium)与CACS_(TNC)相比,差异均无统计学意义(均P>0.05)。以不同的层厚、keV和QIR强度获得的CACS_(PureCalcium)与CACS_(TNC)均呈正相关(r:0.94~0.98,均P<0.05)。采用层厚1.5 mm+55 keV+QIR 1得出的CACS_(PureCalcium)和CACS_(TNC)的偏倚最小(偏倚:4.28,95%一致性上限/下限:167.29/-158.72)。结论 不同的层厚、keV、QIR对CACS的结果均有显著影响,层厚1.5 mm+55 keV+QIR 1参数重建下的CACS_(PureCalcium)更接近CACS_(TNC)。

Objective To evaluate the accuracy of coronary artery calcium scoring(CACS)measured using the _(PureCalcium) virtual non-iodine algorithm on photon-counting detector CT(PCD-CT)under different slice thicknesses,keV,and quantum iterative reconstruction(QIR)parameters,and to identify the optimal reconstruction parameters.Methods A total of 64 patients undergoing PCD-CT coronary CTA(CCTA)were prospectively included.The CACS obtained from true non-contrast enhanced scan(CACS_(TNC))was used as the reference standard._(PureCalcium) images were reconstructed with varying parameters,including slice thickness(3 mm,1.5 mm),keV(55-75),and QIR strength(1,4),to calculate CACS_(PureCalcium).The differences between CACS_(PureCalcium) and CACS_(TNC) were analyzed using the Wilcoxon signed-rank test.Correlation between CACS_(PureCalcium) and CACS_(TNC) was performed using Spearman’s correlation coefficient,and agreement was assessed using the intraclass correlation coefficient(ICC)and Bland-Altman analysis.Results No statistically significant differences between CACS_(PureCalcium) and CACS_(TNC) was observed at 3 mm slice thickness,55/60/65 keV,and QIR 1/4,or at 70 keV and QIR 1,and at 1.5 mm slice thickness,55 keV and QIR 1/4,or at 60 keV and QIR 1(all P>0.05).CACS_(PureCalcium) demonstrated a strong positive correlation with CACS_(TNC) across all tested slice thicknesses,keV levels,and QIR strengths(r=0.94-0.98,all P<0.05).The smallest bias between CACS_(PureCalcium) and CACS_(TNC) was observed at a slice thickness of 1.5 mm,55 keV,and QIR 1(Bias:4.28,95% limits of agreement: 167.29/-158.72). Conclusion Slice thickness, keV level, and QIR strength significantly influence the CACS results. The reconstruction parameters of a 1.5 mm slice thickness, 55 keV, and QIR 1 produce CACS_(PureCalcium) results closest to CACS_(TNC).