Tan T J, Lau Kenneth K, Jackson Dana, Ardley Nicholas, Borasu Adina
Department of Diagnostic Imaging, Monash Health, Melbourne, VIC, Australia.
Faculty of Medicine, Nursing and Health Sciences, Monash University, Victoria, Australia.
Emerg Radiol. 2017 Apr;24(2):133-138. doi: 10.1007/s10140-016-1454-6. Epub 2016 Oct 21.
The purpose of this study was to assess the efficacy of model-based iterative reconstruction (MBIR), statistical iterative reconstruction (SIR), and filtered back projection (FBP) image reconstruction algorithms in the delineation of ureters and overall image quality on non-enhanced computed tomography of the renal tracts (NECT-KUB). This was a prospective study of 40 adult patients who underwent NECT-KUB for investigation of ureteric colic. Images were reconstructed using FBP, SIR, and MBIR techniques and individually and randomly assessed by two blinded radiologists. Parameters measured were overall image quality, presence of ureteric calculus, presence of hydronephrosis or hydroureters, image quality of each ureteric segment, total length of ureters unable to be visualized, attenuation values of image noise, and retroperitoneal fat content for each patient. There were no diagnostic discrepancies between image reconstruction modalities for urolithiasis. Overall image qualities and for each ureteric segment were superior using MBIR (67.5 % rated as 'Good to Excellent' vs. 25 % in SIR and 2.5 % in FBP). The lengths of non-visualized ureteric segments were shortest using MBIR (55.0 % measured 'less than 5 cm' vs. ASIR 33.8 % and FBP 10 %). MBIR was able to reduce overall image noise by up to 49.36 % over SIR and 71.02 % over FBP. MBIR technique improves overall image quality and visualization of ureters over FBP and SIR.
本研究的目的是评估基于模型的迭代重建（MBIR）、统计迭代重建（SIR）和滤波反投影（FBP）图像重建算法在泌尿系统非增强计算机断层扫描（NECT-KUB）中输尿管勾画及整体图像质量方面的效果。这是一项对40例因输尿管绞痛接受NECT-KUB检查的成年患者进行的前瞻性研究。图像分别采用FBP、SIR和MBIR技术重建，并由两名盲法放射科医生独立随机评估。测量的参数包括整体图像质量、输尿管结石的存在、肾积水或输尿管积水的存在、每个输尿管节段的图像质量、无法显示的输尿管总长度、图像噪声的衰减值以及每位患者的腹膜后脂肪含量。在尿路结石的图像重建方式之间没有诊断差异。使用MBIR时，整体图像质量以及每个输尿管节段的图像质量更佳（67.5%被评为“良好至优秀”，而SIR为25%，FBP为2.5%）。使用MBIR时，未显示的输尿管节段长度最短（55.0%测量为“小于5厘米”，而SIR为33.8%，FBP为10%）。与SIR相比，MBIR能够将整体图像噪声降低多达49.36%，与FBP相比降低71.02%。与FBP和SIR相比，MBIR技术可提高整体图像质量并改善输尿管的可视化效果。
