Multi‐band MR fingerprinting (MRF) ASL imaging using artificial‐neural‐network trained with high‐fidelity experimental data

• Published in: Magnetic resonance in medicine Vol. 85; no. 4; pp. 1974 - 1985
• Main Authors: Fan, Hongli, Su, Pan, Huang, Judy, Liu, Peiying, Lu, Hanzhang
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.04.2021
• Subjects: Abnormalities; Accuracy; Adult; arterial spin labeling; Arteries - diagnostic imaging; artificial neural network; Artificial neural networks; Blood flow; bolus arrival time; Brain Mapping - methods; Cerebral blood flow; Cerebrovascular Circulation; Coefficient of variation; Computer Simulation; Data Compression; Female; Fingerprinting; Gray Matter - diagnostic imaging; Healthy Volunteers; Hemodynamics; Humans; Image reconstruction; Kinetics; Learning; Learning theory; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Male; Moyamoya Disease - diagnostic imaging; MR fingerprinting (MRF); Neural networks; Neural Networks, Computer; Perfusion; Reliability; Reproducibility of Results; Spin labeling; Spin Labels; Training; Young Adult; MR fingerprinting (MRF); artificial neural network; cerebral blood flow; perfusion; bolus arrival time; arterial spin labeling
• ISSN: 0740-3194; 1522-2594; 1522-2594
• DOI: 10.1002/mrm.28560

Purpose We aim to leverage the power of deep‐learning with high‐fidelity training data to improve the reliability and processing speed of hemodynamic mapping with MR fingerprinting (MRF) arterial spin labeling (ASL). Methods A total of 15 healthy subjects were studied on a 3T MRI. Each subject underwent 10 runs of a multi‐band multi‐slice MRF‐ASL sequence for a total scan time of approximately 40 min. MRF‐ASL images were averaged across runs to yield a set of high‐fidelity data. Training of a fully connected artificial neural network (ANN) was then performed using these data. The results from ANN were compared to those of dictionary matching (DM), ANN trained with single‐run experimental data and with simulation data. Initial clinical performance of the technique was also demonstrated in a Moyamoya patient. Results The use of ANN reduced the processing time of MRF‐ASL data to 3.6 s, compared to DM of 3 h 12 min. Parametric values obtained with ANN and DM were strongly correlated (R2 between 0.84 and 0.96). Results obtained from high‐fidelity ANN were substantially more reliable compared to those from DM or single‐run ANN. Voxel‐wise coefficient of variation (CoV) of high‐fidelity ANN, DM, and single‐run ANN was 0.15 ± 0.08, 0.41 ± 0.20, 0.30 ± 0.16, respectively, for cerebral blood flow and 0.11 ± 0.06, 0.20 ± 0.19, 0.15 ± 0.10, respectively, for bolus arrival time. In vivo data trained ANN also outperformed ANN trained with simulation data. The superior performance afforded by ANN allowed more conspicuous depiction of hemodynamic abnormalities in Moyamoya patient. Conclusion Deep‐learning‐based parametric reconstruction improves the reliability of MRF‐ASL hemodynamic maps and reduces processing time.