双能CT非线性融合技术提高头颅CTA图像质量的临床应用

Clinical application of head CTA with DSCT non-linear blending technique in improving image quality

目的探讨双源双能量CT非线性融合(NLB)技术提高头颅CTA图像质量的应用价值。方法对线性融合(LB)图像不佳或不能满足诊断要求(主观评分≤3分)的53例双能头颅CTA图像行NLB处理,图像分为80kV组、140kV组、LB组及NLB组。测量并计算4组图像颈内动脉床突段、大脑中动脉、基底动脉及脑干的CT值、标准差(SD)、SNR和CNR;并对图像质量进行主观评分。分别比较4组间CT值、SD值、SNR和CNR及图像质量评分的差异。结果 LB组与NLB组图像主观评分差异有统计学意义[(2.68±0.47)分vs(3.19±0.59)分,P<0.01]。LB组与NLB组可诊断率分别为69.81%(37/53)、83.02%(44/53),图像优良率分别为0(0/53)、37.74%(20/53),差异有统计学意义(P均<0.01)。4组间各部位CT值、SD、SNR及CNR各组间比较差异均有统计学意义(P均<0.05),80kV组CT值最高,NLB组高于LB组(P<0.05);NLB组SNR、CNR均优于其他3组(P均<0.05)。结论 NLB技术可改善和提高头颅CTA图像质量,提高图像的可诊断率和优良率。

Objective To investigate the application value of non-linear blending （NLB） technique in improving the image quality on dual energy head CTA. Methods Totally 58 DSCT head CTA linear blending （LB） data which image quality scores 43 were reconstructed using NLB technique. The row data and reconstructed images were divided into 4 groups, 80 kV group, 140 kV group, LB group and NLB group. The measure ROI were delineated at internal carotid, middle cere- bral artery, basilar artery and brainstem. The CT value, noise （SD）, SNR, CNR of each group were evaluated and com- pared. Results There was significant difference in subjective image score between NLB group and LB group （[2.68± 0.47] vs [3.19±0. 59], P〈0.01）. The diagnostic rate and good rate of images in NLB group and LB group were 83.02%, 37.74% and 69.81%,0, respectively. And the differences were statistical significance （all P〈0.01）. CT val- ue, SD, SNR, CNR had statistical difference among 4 groups （all P〈0.05）. CT value of 80 kV group was highest, NLB group was higher than that of LB group （P〈0.05）; SNR and CNR of NLB group had statistical differences compared with the other three groups （P〈0.05）. Conclusion The NBL takes obvious advantages to improve the image quality.