Assessment of hepatic fatty infiltration using spectral computed tomography imaging: a pilot study.
Xiangpeng Zheng,Yanping Ren,William T Phillips,Ming Li,Meifang Song,Yanqing Hua,Guozhen Zhang
DOI: https://doi.org/10.1097/RCT.0b013e31827ddad3
2013-01-01
Journal of Computer Assisted Tomography
Abstract:Objective: The objective of this study was to investigate the feasibility of using monochromatic spectral computed tomography (CT) imaging to assess fatty infiltration in liver. Materials: With spectral CT imaging, phantoms with known fat concentrations were studied for scanning parameter optimization, then 52 patients enrolled into 4 groups (healthy, mild, moderate, and severe fatty infiltration) received abdominal scanning. Based on reconstructed monochromatic images, hepatic attenuation was analyzed, and dual-energy subtraction imaging (DESI) was created for quantifying fat infiltration. Results: Corresponding to various hepatic fat infiltrations, 4 characteristic CT attenuation curve patterns were described. In DESI images, only fat and fatty components appeared bright. For livers without abnormal fat deposition, isolated bright pixels were visualized (% area = 0.5% +/- 0.3%). With hepatic fat accumulation increasing, more bright pixels appeared in subtraction images with percentages of total liver area involved in 2.5%, 6.7%, and 13.4% of mild, moderate, and severe fat infiltration cases, respectively (P < 0.05). The corresponding CT values were as follows: 1.33, 2.53, 8.69, and 16.4 Hounsfield units (P < 0.01), which correlated with the % DESI area values (r = 0.9811). Conclusions: Spectral CT imaging is a promising method to quantitatively assess hepatic fat content and fatty infiltration with advantages compared with conventional CT imaging.