CT高级建模迭代重建算法对图像质量的影响初步研究  

Preliminary study on the effect of advanced modeled iterative reconstruction algorithm for image quality

作　　者：陆亚光[1] 庄静文[1] 吴天棋[1] 白玫[1] 

机构地区：[1]首都医科大学宣武医院医学工程处,北京100053

出　　处：《中国医学装备》2017年第7期68-71,共4页China Medical Equipment

基　　金：国家自然科学基金(81372923)"宽探测器螺旋CT所致辐射剂量评价方法研究"

摘　　要：目的:研究西门子多排螺旋CT的高级建模迭代重建(ADMIRE)算法对重建图像质量的影响。方法:在同一扫描条件下,使用西门子SOMATOM Force CT对Catphan500体模的CTP515和CTP528模块进行螺旋扫描,未用或使用不同强度的ADMIRE算法对图像进行重建,对不同重建条件下的图像进行主观和客观评价。结果:当图像低对比度为0.3%、0.5%和1.0%时,未用ADMIRE对图像进行重建,其噪声分别为4.90、4.50和5.23;使用ADMIRE2和ADMIRE5重建图像时其噪声分别减少18.35%和53.23%、18.82%和52.32%以及20.33%和53.79%。用ADMIRE0、ADMIRE2和ADMIRE5分别对CTP528模块图像进行重建,对其图像进行主观观察,可观察到的最小细节相同,均为7.0 LP/cm。结论:使用ADMIRE对图像进行重建可以在不影响图像清晰度的同时有效降低噪声,提高CT图像的低对比度分辨率,以帮助临床医师更好的观察组织密度相近的病灶,为临床医师诊断病情提供帮助。Objective： To research the effect of advanced modeled iterative reconstruction（ADMIRE） algorithm for images quality of multi-slice spiral CT scanner of SIEMENS. Methods： Under the same scanning condition, the CTP515 and CTP528 of Catphan500 were scanned by using spiral method of SOMATOM Force CT（SIEMENS）. The images were reconstructed by two conditions（didn＇t use ADMIRE algorithm, or used different intensity of ADMIRE algorithm）. And then, these images from different conditions were evaluated by subjective method and objective method, respectively. Results： When the low contrasts were 0.3%, 0.5% or 1.0%, the image noises were 4.90, 4.50 and 5.23, respectively, under didn＇t use ADMIRE algorithm. In the same contrast, the noises were reduced 18,35%, 53.23% and 18.82%, respectively, under ADMIRE2. And the noises were reduced 52.32%, 20.33% and 53.79%, respectively, under ADMIRE5. When ADMIRE0, ADMIRE2 and ADMIRE5 were used to reconstruct the image of CTP528 module, respectively, the smallest details that could be observed were identical that was 7.0 Line Pairs/cm under subjective observation for images. Conclusion： Using ADMIRE to reconstruct images can effectively reduce the noise while the image resolution don＇t be affected. And to enhance the resolution power of low contrast for CT image can more accurately help clinician to observe lesion with similar tissue density and contribute to diagnose disease.

关 键 词：多排螺旋CT 低对比度分辨率 高级建模迭代重建算法 主观评价 图像质量 

分 类 号：R814.42[医药卫生—影像医学与核医学]