Bourne Roger, Liang Sisi, Panagiotaki Eleftheria, Bongers Andre, Sved Paul, Watson Geoffrey
Discipline of Medical Radiation Sciences, Faculty of Health Sciences, The University of Sydney, 75 East Street, Lidcomb, New South Wales, Australia.
NMR Biomed. 2017 Oct;30(10). doi: 10.1002/nbm.3751. Epub 2017 Jun 30.
The purpose of this study was to measure and model the diffusion time dependence of apparent diffusion coefficient (ADC) and fractional anisotropy (FA) derived from conventional prostate diffusion-weighted imaging methods as used in recommended multiparametric MRI protocols. Diffusion tensor imaging (DTI) was performed at 9.4 T with three radical prostatectomy specimens, with diffusion times in the range 10-120 ms and b-values 0-3000 s/mm . ADC and FA were calculated from DTI measurements at b-values of 800 and 1600 s/mm . Independently, a two-component model (restricted isotropic plus Gaussian anisotropic) was used to synthesize DTI data, from which ADC and FA were predicted and compared with the measured values. Measured ADC and FA exhibited a diffusion time dependence, which was closely predicted by the two-component model. ADC decreased by about 0.10-0.15 μm /ms as diffusion time increased from 10 to 120 ms. FA increased with diffusion time at b-values of 800 and 1600 s/mm but was predicted to be independent of diffusion time at b = 3000 s/mm . Both ADC and FA exhibited diffusion time dependence that could be modeled as two unmixed water pools - one having isotropic restricted dynamics, and the other unrestricted anisotropic dynamics. These results highlight the importance of considering and reporting diffusion times in conventional ADC and FA calculations and protocol recommendations, and inform the development of improved diffusion methods for prostate cancer imaging.
本研究的目的是测量并模拟在推荐的多参数MRI方案中使用的传统前列腺扩散加权成像方法所衍生的表观扩散系数(ADC)和分数各向异性(FA)的扩散时间依赖性。在9.4T磁场下对三个前列腺癌根治术标本进行扩散张量成像(DTI),扩散时间范围为10 - 120ms,b值为0 - 3000s/mm²。在b值为800和1600s/mm²时,根据DTI测量值计算ADC和FA。独立地,使用双组分模型(受限各向同性加高斯各向异性)合成DTI数据,从中预测ADC和FA并与测量值进行比较。测量得到的ADC和FA表现出扩散时间依赖性,双组分模型能很好地预测这种依赖性。随着扩散时间从10ms增加到120ms,ADC降低约0.10 - 0.15μm²/ms。在b值为800和1600s/mm²时,FA随扩散时间增加,但预计在b = 3000s/mm²时与扩散时间无关。ADC和FA均表现出扩散时间依赖性,可将其模拟为两个未混合的水池——一个具有各向同性受限动力学,另一个具有不受限的各向异性动力学。这些结果突出了在传统ADC和FA计算及方案推荐中考虑和报告扩散时间的重要性,并为改进前列腺癌成像的扩散方法的开发提供了依据。