SPECT图像重建方法对Hoffmann模型图像质量的影响

目的探讨SPECT显像不同断层重建方法对Hoffmann模型图像质量的影响。方法采用放射性线源及Hoffmann模型，进行SPECT配平行孔低能高分辨准直器断层采集。对线源图像和Hoffmann模型图像均用滤波反投影（FBP）法和有序子集最大期望值迭代（OSEM）法进行断层重建。对线源重建图像计算2种重建算法的半高宽（FWHM）值，视觉评价2种方法重建的Hoffmann模型图像，并比较2种重建方法的重建时间和模型特定感兴趣区（ROI）。采用SPSS15．0软件，2种重建方法的脑叶及基底节ROI与小脑ROI计数比值行两独立样本t检验。结果平行孔低能高分辨准直器采集的线源断层图像经FBP法和OSEM法重建，FWHM值分别为18．77mm，12．62mm。OSEM法重建所得Hoffmann模型总的图像质量、基底节区的显示以及核团分辨程度均优于FBP法。FBP法和OSEM法重建时间分别为80s和100s，均在临床允许范围之内。OSEM迭代与FBP法脑叶及基底节ROI与小脑ROI比值差异无统计学意义（t=-0．332，P=0．750）。结论在现有软硬件技术条件下，平行孔低能高分辨准直器配合OSEM法优于FBP法，可获得空间分辨率和图像质量较好的Hoffmann模型重建图像。

The effect of different reconstruction algorithms on the image quality of SPECT Hoffmann phantom

Objective To evaluate the image quality of SPECT Hoffmann phantom using different reconstruction algorithms.Methods Data sampling of line-source and Hoffmann phantom were acquired with parallel-hole low energy and high resolution collimator by a dual-head SPECT system.The images were reconstructed by both filtered backprojection (FBP) and ordered-subsets expectation maximization (OSEM) algorithms.The spatial resolution of line resources was evaluated by the full width at half maximum (FWHM) value.The image quality of Hoffmann phantom was visually accessed by 3 doctors.The reconstruc-tion time was recorded and region of interest (ROI) was indentified on the reconstructed Hoffmann phantom images.Independent samples t-test was used to compare the ratios of counts/pixel of target ROI to cerebellum ROI by different reconstruction algorithms using software SPSS 15.0.Results The FWHM values for FBP and OSEM were 18.77 mm and 12.62 mm, respectively.The visual assessment suggested that OSEM could provide better overall image quality, better edge delineation of candate, putamen and striatum than FBP.The reconstruction time of FBP and OSEM was 80 s and 100 s respectively, which was acceptable in routine clini-cal procedure.The ratios of counts/pixel of target ROI to cerebellum ROI generated by FBP and OSEM showed no significant difference (t=-0.332, P=0.750).Conclusion OSEM algorithm could provide bet-ter spatial resolution and image quality than FBP with an acceptable increase in reconstruction time.