肝脏动脉期CT能谱成像的影像优选

Optimal liver CT spectral imaging at arterial phase

目的通过单源双能CT腹部能谱成像,评价单能重建影像的质量并验证以往相关文献的研究结果,研究肝内病灶的最优对比噪声比(CNR)影像所处keV水平,分析单能量影像的CT噪声随单能量水平(keV)的变化特点。方法应用宝石能谱CT,对83例有明确肝内病灶病人进行动脉期能谱成像(GSI扫描)。从原始数据空间重建出一系列单能量影像与混合能量影像;对100个病灶使用Optimal CNR工具获得最佳CNR单能量影像,记录最优CNR影像所在keV水平。测定100个病灶层面腹壁皮下脂肪组织的CT值,记录所得数据中的CT值标准差(SD)代表该部位的CT噪声值(CTN)。筛选最优CNR影像位于70 keV水平附近且大小适中的病灶50个,重建出相同层面的70 keV影像与混合能量影像(140 kV),并以随机展示方式由2个具有一定资历的诊断医师在盲法下评估2种影像探查病灶的能力及显示病灶内外情况的清晰度。结果结果显示最优CNR位于70 keV附近水平明显较高。96%的噪声曲线均在低keV段出现一个低keV段极低噪声,而在70 keV上出现低keV段极低噪声者占63%。在主观影像评价方面,70 keV影像在显示病灶内外细节情况如血管壁的显示优于混合能量影像。结论肝脏病变的动脉期能谱成像的最优CNR影像大多位于70 keV水平附近,低keV段极低噪声亦最多见于70 keV水平,并接近或等于最低噪声。70 keV水平有利于发现大部分肝内病灶,而且噪声小,可清晰显示病灶内外情况,有良好的临床使用价值。

Objective To evaluate the effect of energic level on the image quality of liver disease in spectral imaging. Find out the optimal energic level from the monoenergetic series and mixed-energy mode imaging. Methods Eighty-three patients with liver disease underwent gemstone spectral imaging （GSI） at arterial phase. GSI viewer was used to obtain the optimal contrast to noise ratio （CNR） curve and find out the optimal CNR monochromatic images. Then the optimal energic level （keV） was recorded, and the distribution characteristics of liver diseases on the optimal CNR monoenergetic images were analyzed. The CT values of abdominal fat were measured with fixed size and form regions of interest at the levels of 100 lesions, and the standard deviation of the CT values at each energy level was calculated as CT noise. 50 medium-size lesions on polychromatic images and 70 keV monochromatic images were independently evaluated by two experienced doctors for lesion detectability. Results The optimal CNR monoenergetic images mainly located at 70 keV. The CT noise generally showed dechne with increasing energy level. However, an even lower CT noise was found at low energy level, especially at 70 keV. There was no statistically difference in lesion detetability, but the 70 keV monoenergetic image could better display the tiny structures such as vessels. Conclution The spectral images obtained from gestone spectral CT, the optimal CNR monoenergetic image for detecting liver lesions is near to 70 keV level, and CT noise is lowest also near to 70 keV at low energy. The 70 keV image is better than the polychromatic image in detecting tiny structures. The 70 keV is recommended for spectral imaging.