【摘要】 目的 探讨双源CT卷积核重建结合迭代去金属伪影技术（iMAR）在脑动脉瘤术后CT血管造影（CTA）中的应用价值。方法 选取2025年1月至2025年9月我院脑动脉瘤术后行CTA检查的30例患者作为研究对象，根据金属植入物类型分为栓塞弹簧圈组（17例）和手术夹组（13例），根据金属植入物位置分为近颅骨组（10例）、颅底组（12例）和大脑中部组（8例）。所有患者均采用双源CT行CTA扫描，生成4组融合图像（MIX）:HV38卷积核（对照组）、 HV38+iMAR（实验组1）、HV45+iMAR（实验组2）、 HV49+iMAR（实验组3）。由2名资深影像科医师采用5分制量表对各组图像的金属伪影程度、血管显示清晰度及诊断可行性进行主观评分；测量金属伪影区CT值标准差（SD）及伪影区与正常脑实质CT值差值（ΔCT），评估伪影抑制效果；记录不同金属植入物类型、位置图像的伪影改善情况。结果 实验组1、 2、 3的金属伪影程度、血管显示清晰度及诊断可行性评分均显著高于对照组，且实验组2的各项图像主观评分最优（P<0.05）。实验组1、 2、 3的伪影区SD值、ΔCT值均显著低于对照组，且实验组2的各项图像客观指标最优（P<0.05）。栓塞弹簧圈组和手术夹组中，实验组2的金属伪影程度评分最高，SD值、ΔCT值最低（P<0.05）。所有组别中均为实验组2的伪影改善效果最佳。同一技术下，颅底组的金属伪影程度评分最低，SD值、ΔCT值均最高，但与近颅骨组、大脑中部组比较，差异不具有统计学意义（P >0.05）。结论 双源CT卷积核重建结合i MAR可有效减少脑动脉瘤术后CTA金属伪影，特别是HV45卷积核联合iMAR技术的伪影抑制效果最优，能显著提升血管显示质量及诊断可行性，且对不同类型及位置的金属植入物均具有良好适应性。更多还原

【Abstract】 Objective To investigate the application value of dual-source CT convolution kernel reconstruction combined with iterative metal artifact reduction(i MAR) in CT angiography(CTA) after cerebral aneurysm surgery. Methods 30 patients who underwent CTA after cerebral aneurysm surgery in our hospital from January 2025 to September 2025 were enrolled. According to the type of metallic implants, the patients were divided into the embolization coil group(17 cases) and the surgical clip group(13 cases). According to the location of metallic implants, they were divided into the paraskull group(10 cases), the skull base group(12 cases), and the middle cerebral group(8 cases). All patients underwent CTA scanning with dual-source CT, and four sets of mixed images(MIX) were generated: HV38 convolution kernel(control group), HV38+iMAR(experimental group 1), HV45+iMAR(experimental group 2), and HV49 + iMAR(experimental group 3). Two senior radiologists subjectively scored the metal artifact severity, vascular clarity, and diagnostic feasibility of each group using a 5-point scale. The standard deviation(SD) of CT values in the metal artifact area and the CT value difference(ΔCT)between the artifact area and normal brain parenchyma were measured to evaluate the artifact suppression effect. The artifact improvement of images with different types and locations of metallic implants was recorded. Results The metal artifact severity, vascular clarity and diagnostic feasibility scores in experimental groups 1, 2, and 3 were significantly higher than those in the control group(P <0.05);Experimental group 2 achieved the best subjective image scores(P <0.05). The SD and ΔCT values in experimental groups 1, 2, and 3 were significantly lower than those in the control group, and experimental group 2 showed the best objective image indicators(P <0.05). In both the embolization coil group and the surgical clip group, experimental group 2 had the highest metal artifact severity scores and the lowest SD and ΔCT values(P <0.05). Among all groups, experimental group 2 exhibited the best artifact improvement effect. Under the same technique, the skull base group had the lowest metal artifact severity score and the highest SD and ΔCT values, but the differences were not statistically significant compared with the paraskull group and the middle cerebral group(P >0.05). Conclusions Dual-source CT convolution kernel reconstruction combined with i MAR can effectively reduce metal artifacts on CTA after cerebral aneurysm surgery. In particular, the HV45 convolution kernel combined with i MAR shows the optimal artifact suppression effect, which can significantly improve vascular visualization quality and diagnostic feasibility, with good adaptability to metallic implants of different types and locations.更多还原