Haji-Valizadeh Hassan, Guo Rui, Kucukseymen Selcuk, Paskavitz Amanda, Cai Xiaoying, Rodriguez Jennifer, Pierce Patrick, Goddu Beth, Kim Daniel, Manning Warren, Nezafat Reza
Department of Medicine (Cardiovascular Division), Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts, USA.
Siemens Medical Solutions USA, Inc., Boston, Massachusetts, USA.
Magn Reson Med. 2021 Aug;86(2):804-819. doi: 10.1002/mrm.28750. Epub 2021 Mar 15.
To develop and evaluate a real-time phase contrast (PC) MRI protocol via complex-difference deep learning (DL) framework.
DL used two 3D U-nets to separately filter aliasing artifact from radial real-time velocity-compensated and complex-difference images. U-nets were trained with synthetic real-time PC generated from electrocardiograph (ECG) -gated, breath-hold, segmented PC (ECG-gated segmented PC) acquired at the ascending aorta of 510 patients. In 21 patients, free-breathing, ungated real-time (acceleration rate = 28.8) and ECG-gated segmented (acceleration rate = 2) PC were prospectively acquired at the ascending aorta. Hemodynamic parameters (cardiac output [CO], stroke volume [SV], and mean velocity at peak systole [peak mean velocity]) were measured for ECG-gated segmented and DL-filtered synthetic real-time PC and compared using Bland-Altman and linear regression analyses. Additionally, hemodynamic parameters were quantified from DL-filtered, compressed-sensing (CS) -reconstructed, and gridding reconstructed prospective real-time PC and compared to ECG-gated segmented PC.
Synthetic real-time PC with DL showed strong correlation (R > 0.98) and good agreement with ECG-gated segmented PC for quantified hemodynamic parameters (mean-difference: CO = -0.3 L/min, SV = -4.3 mL, peak mean velocity = -2.3 cm/s). On average, DL required 0.39 s/frame to filter prospective real-time PC, which was 4.6-fold faster than CS. Compared to CS, DL showed superior correlation, tighter limits of agreement (LOAs), better bias for peak mean velocity, and worse bias for CO and SV. Compared to gridding, DL showed similar correlation, tighter LOAs for CO and SV, similar bias for CO, and worse bias for SV and peak mean velocity.
The complex-difference DL framework accelerated real-time PC-MRI by nearly 28-fold, enabling rapid free-running real-time assessment of flow hemodynamics.
通过复差分深度学习（DL）框架开发并评估一种实时相位对比（PC）MRI协议。
DL使用两个3D U-Net分别从径向实时速度补偿图像和复差分图像中滤除混叠伪影。U-Net使用从510例患者升主动脉采集的心电图（ECG）门控、屏气、分段PC（ECG门控分段PC）生成的合成实时PC进行训练。对21例患者，在升主动脉前瞻性采集自由呼吸、非门控实时（加速率 = 28.8）和ECG门控分段（加速率 = 2）PC。测量ECG门控分段和DL滤波后的合成实时PC的血流动力学参数（心输出量[CO]、每搏输出量[SV]和收缩期峰值平均速度[峰值平均速度]），并使用Bland-Altman分析和线性回归分析进行比较。此外，从DL滤波、压缩感知（CS）重建和网格化重建的前瞻性实时PC中量化血流动力学参数，并与ECG门控分段PC进行比较。
DL处理后的合成实时PC与ECG门控分段PC在量化血流动力学参数方面显示出强相关性（R > 0.98）和良好一致性（平均差异：CO = -0.3 L/min，SV = -4.3 mL，峰值平均速度 = -2.3 cm/s）。平均而言，DL滤波前瞻性实时PC每帧需要0.39 s，比CS快4.6倍。与CS相比，DL显示出更好的相关性、更窄的一致性界限（LOA）、峰值平均速度的偏差更小、CO和SV的偏差更大。与网格化相比，DL显示出相似的相关性、CO和SV的LOA更窄、CO的偏差相似、SV和峰值平均速度的偏差更大。
复差分DL框架将实时PC-MRI加速了近28倍，能够对血流动力学进行快速自由运行实时评估。
