Qiang Zhang1, Pan Su2, Zhensen Chen3, Ying Liao4, Shuo Chen1, Rui Guo5, Haikun Qi6, Xuesong Li7, Xue Zhang1, Zhangxuan Hu1, Hanzhang Lu2, Huijun Chen1. 1. Center for Biomedical Imaging Research, Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing, China. 2. The Russell H. Morgan, Department of Radiology, Johns Hopkins University School of Medicine, Baltimore, Maryland. 3. Vascular Imaging Laboratory, Department of Radiology, University of Washington, Seattle, Washington. 4. Center for Biomedical Imaging, Department of Radiology, New York University School of Medicine, New York, New York. 5. Department of Medicine (Cardiovascular Division), Beth Israel deaconess Medical Center and Harvard Medical School, Boston, Massachusetts. 6. School of Biomedical Engineering and Imaging Sciences, King's College London, London, London, United Kingdom. 7. School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China.
Abstract
PURPOSE: To develop a reproducible and fast method to reconstruct MR fingerprinting arterial spin labeling (MRF-ASL) perfusion maps using deep learning. METHOD: A fully connected neural network, denoted as DeepMARS, was trained using simulation data and added Gaussian noise. Two MRF-ASL models were used to generate the simulation data, specifically a single-compartment model with 4 unknowns parameters and a two-compartment model with 7 unknown parameters. The DeepMARS method was evaluated using MRF-ASL data from healthy subjects (N = 7) and patients with Moymoya disease (N = 3). Computation time, coefficient of determination (R2 ), and intraclass correlation coefficient (ICC) were compared between DeepMARS and conventional dictionary matching (DM). The relationship between DeepMARS and Look-Locker PASL was evaluated by a linear mixed model. RESULTS: Computation time per voxel was <0.5 ms for DeepMARS and >4 seconds for DM in the single-compartment model. Compared with DM, the DeepMARS showed higher R2 and significantly improved ICC for single-compartment derived bolus arrival time (BAT) and two-compartment derived cerebral blood flow (CBF) and higher or similar R2 /ICC for other parameters. In addition, the DeepMARS was significantly correlated with Look-Locker PASL for BAT (single-compartment) and CBF (two-compartment). Moreover, for Moyamoya patients, the location of diminished CBF and prolonged BAT shown in DeepMARS was consistent with the position of occluded arteries shown in time-of-flight MR angiography. CONCLUSION: Reconstruction of MRF-ASL with DeepMARS is faster and more reproducible than DM.
PURPOSE: To develop a reproducible and fast method to reconstruct MR fingerprinting arterial spin labeling (MRF-ASL) perfusion maps using deep learning. METHOD: A fully connected neural network, denoted as DeepMARS, was trained using simulation data and added Gaussian noise. Two MRF-ASL models were used to generate the simulation data, specifically a single-compartment model with 4 unknowns parameters and a two-compartment model with 7 unknown parameters. The DeepMARS method was evaluated using MRF-ASL data from healthy subjects (N = 7) and patients with Moymoya disease (N = 3). Computation time, coefficient of determination (R2 ), and intraclass correlation coefficient (ICC) were compared between DeepMARS and conventional dictionary matching (DM). The relationship between DeepMARS and Look-Locker PASL was evaluated by a linear mixed model. RESULTS: Computation time per voxel was <0.5 ms for DeepMARS and >4 seconds for DM in the single-compartment model. Compared with DM, the DeepMARS showed higher R2 and significantly improved ICC for single-compartment derived bolus arrival time (BAT) and two-compartment derived cerebral blood flow (CBF) and higher or similar R2 /ICC for other parameters. In addition, the DeepMARS was significantly correlated with Look-Locker PASL for BAT (single-compartment) and CBF (two-compartment). Moreover, for Moyamoya patients, the location of diminished CBF and prolonged BAT shown in DeepMARS was consistent with the position of occluded arteries shown in time-of-flight MR angiography. CONCLUSION: Reconstruction of MRF-ASL with DeepMARS is faster and more reproducible than DM.
Authors: Pan Su; Peiying Liu; Marco C Pinho; Binu P Thomas; Ye Qiao; Judy Huang; Babu G Welch; Hanzhang Lu Journal: Magn Reson Imaging Date: 2022-02-18 Impact factor: 2.546
Authors: Hongli Fan; Pan Su; Doris Da May Lin; Emily B Goldberg; Alexandra Walker; Richard Leigh; Argye E Hillis; Hanzhang Lu Journal: Stroke Date: 2022-03-16 Impact factor: 10.170
Authors: Luis Hernandez-Garcia; Verónica Aramendía-Vidaurreta; Divya S Bolar; Weiying Dai; Maria A Fernández-Seara; Jia Guo; Ananth J Madhuranthakam; Henk Mutsaerts; Jan Petr; Qin Qin; Jonas Schollenberger; Yuriko Suzuki; Manuel Taso; David L Thomas; Matthias J P van Osch; Joseph Woods; Moss Y Zhao; Lirong Yan; Ze Wang; Li Zhao; Thomas W Okell Journal: Magn Reson Med Date: 2022-08-19 Impact factor: 3.737