摘要
目的探讨迭代去金属伪影算法(IMAR)对椎体植入物患者CT图像的影响。方法回顾性分析45例有椎体植入物患者的CT图像,采用常规滤波反投影(FBP)和FBP+IMAR两种算法重建图像,选择两种重建图像中无金属植入物(椎体中份层面)和有金属植入物(椎体中份的螺钉层面)的图像各一幅分析,分别测量4组图像中椎体骨组织的CT值和噪声值,并由两名放射医师分别对两种重建中有金属伪影的图像进行主观图像质量评价。结果金属植入物图像FBP和FBP+IMAR两种算法图像主观评价得分分别为(1.69±0.56)、(3.36±0.53)分,二者比较差异有统计学意义(Z=-6.05,P<0.01)。无金属植入物图像两种算法椎体骨组织的CT值分别为(256.44±32.27)、(257.04±32.31)HU,图像噪声值分别为(31.90±6.66)、(33.17±6.62)HU,二者比较差异有统计学意义(t=-8.27、-31.44,P<0.01);使用IMAR后,无金属植入物图像的CT值和噪声都会增加。金属植入物图像两种算法椎体骨组织的CT值分别为(395.71±106.85)、(197.14±41.77)HU,图像噪声值分别为(64.88±12.84)、(31.13±7.16)HU,差异有统计学意义(t=14.33、19.09,P<0.01),使用IMAR后,其图像的CT值和噪声都会减小。结论 IMAR在无金属植入物的图像中会轻微增加其CT值和噪声值;在有金属植入物的图像中时可改善其CT值和减少噪声值,明显减少金属伪影。
Objective To evaluate the effect of iterative metal artifact reduction(IMAR)algorithm on CT images of patients with vertebral implants.Methods CT images of 45 patients with vertebral implants were analyzed retrospectively.The images were reconstructed with filtered back projection(FBP)or FBP+IMAR.An analysis was performed for one each of the reconstructed images without a metal implant(the median level of the vertebra)and with a metal implant(the screw level of the median level of the vertebra),the CT value and noise value of vertebral bone tissue in the four images were measured respectively.The subjective image quality of the reconstruction images were evaluated by two radiologists respectively.Results In the images with metal implant,the subjective evaluation score of the two algorithms were(1.69±0.56)and(3.36±0.53)respectively,and the difference was statistically significant(Z=-6.05,P<0.01).In the images with no-metal implants,the CT value of the vertebral bone tissue of the two algorithms were(256.44±32.27)HU and(257.04±32.31)HU,the noise value was(31.90±6.66)HU and(33.17±6.62)HU,respectively.The difference was statistically significant(t=-8.27,P<0.01;t=-31.44,P<0.01).With the IMAR algorithm,the CT value and noise value are increased in the image with no-metal implants.In the images with the metal implants,the CT value of the vertebral bone tissue were(395.71±106.85)HU and(197.14±41.77)HU,the noise value were(64.88±12.84)HU and(31.13±7.16)HU,respectively.The difference was statistically significant(t=14.33,P<0.01;t=19.09,P<0.01).With the IMAR algorithm,the CT value and noise value in the image with metal implants are reduced.Conclusion IMAR algorithm can slightly increase the CT value and noise value in the images with no-metal implants.And it can improve the CT value and reduce the noise value,reduce the metal artifacts significantly in the images with metal implants.
作者
郁仁强
吕发金
张志伟
姚开情
彭刚
曾勇明
杨亚
YU Renqiang;LYU Fajin;ZHANG Zhiwei;YAO Kaiqing;PENG Gang;ZENG Yongming;Yang Ya(Department of Radiology,the First Affiliated Hospital of Chongqing MedicalUniversity,Chongqing 400016,China;Department of Radiology,Chengdu Women's andChildren's Central Hospital,Chengdu,Sichuan 610000,China)
出处
《重庆医学》
CAS
2019年第9期1549-1551,1559,共4页
Chongqing medicine
基金
国家临床重点专科建设项目(国卫办医函[2013]544号)