期刊文献+

迭代重建与线性融合联合抑制骨折内固定器所致CT金属伪影

Iterative reconstruction combined with linear blending for suppressing metal artifacts in CT images caused by fracture internal fixation devices
暂未订购
导出
摘要 目的观察迭代重建(IR)与线性融合(LB)联合(IR-LB)抑制骨折内固定器所致CT金属伪影的价值。方法前瞻性纳入100例骨折内固定术后拟接受复查患者,根据手术部位分为A组(颈椎,n=21)、B组(胸腰骶椎,n=37)、C组(骨盆及四肢,n=31)及D组(关节,n=11)。行双能量CT扫描,分别以滤波反投影法(FBP)、虚拟单能谱及IR-LB技术重建图像,并对图像质量进行主、客观评价。结果4组内3种重建图像质量主观评分差异均有统计学意义(P均<0.05),以(5,-0.3)IR-LB重建图像整体质量最佳。4组内3种重建图像伪影指数(AI)差异均有统计学意义(P均<0.05),且虚拟单能谱及IR-LB重建图像AI均显著小于FBP(P均<0.05);其中(5,-0.3)IR-LB重建图像高密度AI最小、(5,-0.7)IR-LB重建图像低密度AI最小。4组内3种重建图像信噪比(SNR)和对比度噪声比(CNR)差异均有统计学意义(P均<0.05),且虚拟单能谱及IR-LB重建图像SNR和CNR均显著大于FBP(P均<0.05),以(5,-0.3)IR-LB重建图像SNR和CNR最高。结论IR-LB能有效抑制骨折内固定器所致CT金属伪影,尤以IR系数5、LB系数-0.3整体抑制效果最佳。 Objective To observe the value of iterative reconstruction(IR)combined with linear blending(LB)(IRLB)for suppressing metal artifacts in CT images caused by fracture internal fixation devices.Methods A total of 100 patients who would undergo postoperative review after internal fixation for fractures were prospectively enrolled and divided into group A(cervical spine,n=21),group B(thoracolumbar sacral spine,n=37),group C(pelvis and extremities,n=31)and group D(joints,n=11)based on surgical sites.Dual-energy CT scanning were performed,and the images were reconstructed with filtered back projection(FBP),virtual single energy spectrum and IR-LB,respectively,and the imaging qualities were subjectively and objectively evaluated.Results Significant differences of subjective scores of 3 kinds reconstructed image quality were found within each group(all P<0.05),and(5,-0.3)IR-LB reconstructed images had the best overall quality.Meanwhile,significant differences of artifact indexes(AI)were among different reconstructed images within each group(all P<0.05).AI of virtual single energy spectrum and IR-LB reconstructed images were significantly lower than that of FBP images(all P<0.05),and(5,-0.3)IR-LB reconstructed images had the smallest high-density AI,while(5,-0.7)IR-LB reconstructed images had the smallest low-density AI.Significant differences of signal to noise ratio(SNR)and contrast to noise ratio(CNR)among different reconstructed images were found within each group(all P<0.05).SNR and CNR of virtual single energy spectrum and IR-LB reconstructed images were significantly higher than those of FBP images(all P<0.05),and(5,-0.3)IR-LB reconstructed images had the highest SNR and CNR.Conclusion IR-LB could effectively suppress CT metal artifacts caused by fracture internal fixation devices,and IR coefficient of 5 and LB coefficient of-0.3 had the best overall suppression effect.
作者 李才林 钱堃 冉春燕 范如雪 LI Cailin;QIAN Kun;RAN Chunyan;FAN Ruxue(Department of Radiology,Affiliated Hospital of Zunyi Medical University,Zunyi 563000,China)
出处 《中国介入影像与治疗学》 北大核心 2026年第1期32-37,共6页 Chinese Journal of Interventional Imaging and Therapy
基金 合肥市科技计划项目[遵市科合HZ字(2023)267号]。
关键词 骨折 骨折固定装置 体层摄影术 X线计算机 深度学习 伪影 图像质量 fractures,bone orthopedic fixation devices tomography,X-ray computed deep learning artifacts image quality
  • 相关文献

参考文献12

二级参考文献44

共引文献59

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部