FBP、ASiR和VEO三种重建算法对常规剂量胸部CT图像质量的影响

Impact of reconstruction techniques on routine dose chest CT image quality: Comparison of FBP,ASiR and VEO

目的探讨滤波反投影(FBP)、自适应统计迭代重建技术(ASiR)和基于模型的迭代重建算法(MBIR,商品名"VEO")三种重建技术对常规剂量胸部薄层CT图像质量的影响。方法应用能谱CT对15例成年患者行胸部增强CT扫描,扫描条件:100kVp,自动毫安,噪声指数15,螺距0.984∶1,球管转速0.4秒/圈。分别用FBP、50%ASiR(50%比例ASiR和FBP混合以降低噪声)和VEO三种重建算法对原始数据行0.625mm薄层重建,测量图像噪声及胸主动脉与背部肌肉的对比噪声比(CNR),并对3组图像分别进行质量评分,然后行对比分析。结果FBP、50%ASiR和VEO三组图像的噪声分别为24.30±3.55、17.11±2.55及11.69±1.74,50%ASiR和VEO组图像噪声分别较FBP组降低29.59%和51.89%(P均<0.01);胸主动脉与背部肌肉的CNR FBP、50%ASiR和VEO三组图像分别为10.56±3.05、15.15±3.88及21.69±5.62,50%ASiR和VEO组图像CNR较FBP组分别提高43.47%和105.40%(P均<0.01);图像质量主观评分FBP、50%ASiR和VEO三组图像分别为4.03±0.72、4.63±0.41及5.75±0.25,50%ASiR和VEO组图像较FBP组分别提高14.89%和42.68%(P均<0.01)。结论与FBP重建算法比较,在相同剂量条件下,50%ASiR和VEO能显著降低胸部CT图像噪声并提高图像质量;其中VEO重建算法降噪及提高图像质量效果更为显著。

Objective To investigate the impact of different reconstruction algorithms,including filtered back projection（FBP）,adaptive statistical iterative reconstruction（ASiR） and model-based iterative reconstruction（MBIR,with VEO as its trade name） on image quality of the routine dose chest CT.Methods With institutional review board approval,15 adult patients who underwent enhanced chest CT examination were enrolled.Scanning parameters included a pitch of 0.984∶1,100 kVp（peak）,noise index 15,auto current,40 mm table feed per rotation.Raw data were reconstructed with FBP,50%ASiR（blending of 50% ASiR and 50% FBP for obtaining noise reduction） and VEO algorithm respectively,and the reconstructed section thickness was 0.625 mm.Image noises were measured,and contrast-to-noise ratio（CNR） of thoracic aorta relative to back muscle was assessed.Image quality was assessed using a 6-point scale.Results The image noise of FBP,50%ASIR and VEO was 24.30±3.55,17.11±2.55 and 11.69±1.74,respectively.Compared with FBP,objective image noise reduced by 51.89%（P0.01） and 29.59%（P0.01） in images reconstructed with VEO and 50%ASiR,respectively.The CNR of thoracic aorta to back muscle for FBP,50%ASiR and VEO was 10.56±3.05,15.15±3.88 and 21.69±5.62,respectively.Compared with FBP,CNR of images reconstructed with 50%ASiR and VEO increased by 43.47%（P0.01） and 105.40%（P0.01）,respectively.The mean subjective score of image quality reconstructed with VEO was 5.75±0.25,14.89% and 42.68% higher than that of FBP（4.03±0.72,P0.01） and 50%ASiR（4.63±0.41,P0.01）.Conclusion VEO and ASIR reconstruction techniques have the ability to reduce image noise and improve image quality compared with the current algorithms such as FBP,especially VEO technique.