Correia Teresa, Cruz Gastão, Schneider Torben, Botnar René M, Prieto Claudia
Division of Imaging Sciences and Biomedical Engineering, King's College London, London, UK.
Philips Healthcare, Guildford, Surrey, UK.
Med Phys. 2018 Jan;45(1):214-222. doi: 10.1002/mp.12663. Epub 2017 Dec 12.
To develop an accelerated and nonrigid motion-compensated technique for efficient isotropic 3D whole-heart coronary magnetic resonance angiography (CMRA) with Cartesian acquisition.
Highly efficient whole-heart 3D CMRA was achieved by combining image reconstruction from undersampled data using compressed sensing (CS) with a nonrigid motion compensation framework. Undersampled acquisition was performed using a variable-density Cartesian trajectory with radial order (VD-CAPR). Motion correction was performed in two steps: beat-to-beat 2D translational correction with motion estimated from interleaved image navigators, and bin-to-bin 3D nonrigid correction with motion estimated from respiratory-resolved images reconstructed from undersampled 3D CMRA data using CS. Nonrigid motion fields were incorporated into an undersampled motion-compensated reconstruction, which combines CS with the general matrix description formalism. The proposed approach was tested on 10 healthy subjects and compared against a conventional twofold accelerated 5-mm navigator-gated and tracked acquisition.
The proposed method achieves isotropic 1.2-mm Cartesian whole-heart CMRA in 5 min ± 1 min (~8× acceleration). The proposed approach provides good-quality images of the left and right coronary arteries, comparable to those of a twofold accelerated navigator-gated and tracked acquisition, but scan time was up to about four times faster. For both coronaries, no significant differences (P > 0.05) in vessel sharpness and length were found between the proposed method and reference scan.
The feasibility of a highly efficient motion-compensated reconstruction framework for accelerated 3D CMRA has been demonstrated in healthy subjects. Further investigation is required to assess the clinical value of the method.
开发一种加速且非刚性运动补偿技术,用于采用笛卡尔采集的高效各向同性三维全心冠状动脉磁共振血管造影(CMRA)。
通过将使用压缩感知(CS)从欠采样数据进行图像重建与非刚性运动补偿框架相结合,实现了高效的全心三维CMRA。使用具有径向顺序的可变密度笛卡尔轨迹(VD-CAPR)进行欠采样采集。运动校正分两步进行:使用从交错图像导航器估计的运动进行逐搏二维平移校正,以及使用从使用CS从欠采样三维CMRA数据重建的呼吸分辨图像估计的运动进行逐组三维非刚性校正。将非刚性运动场纳入欠采样运动补偿重建中,该重建将CS与通用矩阵描述形式相结合。在10名健康受试者上对所提出的方法进行了测试,并与传统的两倍加速5毫米导航门控和跟踪采集进行了比较。
所提出的方法在5分钟±1分钟内(约8倍加速)实现了各向同性1.2毫米笛卡尔全心CMRA。所提出的方法提供了左、右冠状动脉的高质量图像,与两倍加速导航门控和跟踪采集的图像相当,但扫描时间快了约四倍。对于两支冠状动脉,所提出的方法与参考扫描之间在血管清晰度和长度方面均未发现显著差异(P>0.05)。
已在健康受试者中证明了用于加速三维CMRA的高效运动补偿重建框架的可行性。需要进一步研究以评估该方法的临床价值。