不同迭代重组技术对腹部CT图像质量的影响  

Improving image quality in routine dose abdominal CT with model-base iterative reconstruction

作　　者：张兰欣 杨创勃[1] 谷春雨 郭炎兵 师卫华[1] 张兆国[1] 贾永军[1] 李婉茹 ZHANG Lan-xin;YANG Chuang-bo;GU Chun-yu;GUO Yan-bing;SHI Wei-hua;ZHANG Zhao-guo;JIA Yong-jun;LI Wan-ru(Department of Medical Imaging, Affiliated Hospital of Shaanxi University of Chinese Medicine,Shaanxi 712000,China)

机构地区：[1]陕西中医药大学,陕西中医药大学附属医院医学影像科,陕西咸阳712000

出　　处：《影像诊断与介入放射学》2018年第2期103-106,共4页Diagnostic Imaging & Interventional Radiology

摘　　要：目的探讨基于模型的迭代重组(MBIR)技术和自适应统计迭代重组技术(ASi R)在常规剂量下对薄层腹部CT图像质量的影响。方法采用GE Discovery CT 750HD对30例患者进行腹部常规剂量CT扫描。分别采用常规FBP(FBP)重组和迭代重组(ASi R40%和MBIR NR40)技术对原始数据进行0.625 mm薄层重组,应用后处理台对图像进行质量评价分析。客观测量并比较肝脏、脾脏、左侧竖脊肌、皮下脂肪等的噪声值(SD)、信噪比(SNR)、和对比噪声比(CNR)。重组薄层图像的主观评价由2名有经验的影像医师采用盲法进行评分。结果以实质器官肝脏为主要研究对象,MBIR NR40图像噪声最低,且MBIR NR40重组图像的噪声低于常规剂量ASi R40%重组图像;MBIR NR40图像肝脏、脾脏的CNR最高,MBIR NR40高于ASi R40%重组图像。MBIR NR40主观评分最高,且均优于常规剂量ASi R重组图像。结论与常规FBP重组算法比较,在相同剂量条件下,ASi R40%和MBIR NR40能显著降低腹部CT图像噪声,并提高图像质量;其中MBIR NR40重组算法降噪及提高图像质量效果更为显著。Objective To assess the value of a model-based iterative reconstruction（MBIR） in improving image quality of routine dose abdominal CT scans, in comparison with filtered back projection（ FBP） and adaptive statistical iterative reconstruction（ASiR） algorithms. Methods 30 patients who underwent enhanced abdominal CT were enrolled. Raw data were reconstructed with FBP, ASiR40%（40% ASiR and 60% FBP）, and MBIR algorithms at reconstructed section thickness of 0.625 mm. The reconstructed thin-layer images were qualitatively evaluated by two experienced imaging physicians blinded to the reconstruction algorithms. The noise（SD）, signal to noise ratio（SNR）, and contrast to noise ratio（ CNR） of liver, spleen, left erector spinae muscle and subcutaneous fat were compared. Results In the liver, the noise of MBIR NR40 image was the lowest. The highest CNR was found in the liver and spleen of MBIRNR40. The qualitative score of MBIR NR40 was also the highest. Conclusion Compared with the conventional FBP reconstruction algorithm, ASi R40% and MBIR NR40 can significantly reduce the abdominal CT image noise and improve the image quality under the same dose condition. The MBIR NR40 reconstruction algorithm is more effective in reducing noise and improving image quality.

关 键 词：体层摄影术 X线计算机 滤波反投影法 自适应统计迭代重组 基于模型迭代重组 

分 类 号：R816.5[医药卫生—放射医学]