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

Abstract

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 (R 2 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. K E Y W O R D S arterial spin labeling, artificial neural network, bolus arrival time, cerebral blood flow, MR fingerprinting (MRF), perfusion

Training data distribution significantly impacts the estimation of tissue microstructure with machine learning

“…Recent works that leverage supervised ML for model parameter estimation in qMRI typically employ one of two training data distributions: (1) parameter combinations obtained from traditional model fitting and the corresponding measured qMRI signals, 4,6,9,11,[14][15][16][17] or (2) parameters sampled uniformly from the entire plausible parameter space with simulated qMRI signals. 5,[18][19][20][21][22][23][24] While (1) uses parameter combinations directly estimated from the data so likely quantifies the model parameters with higher accuracy and precision for a given specific dataset, (2) supports choice of training data distribution, which may help improve generalizability and avoid problems arising from imbalance.…”

Automated generation of cerebral blood flow and arterial transit time maps from multiple delay arterial spin‐labeled MRI

“…[9][10][11][12] Deep neural networks have also been advantageous in de-noising Hadamard-encoded ASL images and as parameter estimation models in ASL MR fingerprinting applications. [13][14][15][16] These developments suggest the feasibility of deep neural networks for de-noising and hemodynamic parameter quantification of sequential multi-PLD ASL. Some ongoing challenges of deep learning for ASL processing include 3 dimensional (3D) architectures, uncertainty in network outputs, generalizability to acquisition sequence, and diverse clinical populations.…”

“…Deep learning offers advantages for many stages of conventional image processing pipelines, 8 as is the case for de‐noising and artifact removal in single‐PLD ASL processing 9–12 . Deep neural networks have also been advantageous in de‐noising Hadamard‐encoded ASL images and as parameter estimation models in ASL MR fingerprinting applications 13–16 . These developments suggest the feasibility of deep neural networks for de‐noising and hemodynamic parameter quantification of sequential multi‐PLD ASL.…”

Simultaneous Hemodynamic and Structural Imaging of Ischemic Stroke With Magnetic Resonance Fingerprinting Arterial Spin Labeling

“…14,16 To further improve the accuracy of these estimations, this study used an artificial neural network-based reconstruction method to obtain parametric maps, which has been shown to be superior to the dictionary-matching methods according to the previous work published in the literature. [17][18][19] Compared with gadolinium-based perfusion MRI, MRF-ASL does not require the use of contrast agent, therefore, it is particularly suitable for patients who have allergic reactions, poor renal function, other contraindications, or concerns associated long-term deposition of the gadolinium in the brain. Logistically, MRF-ASL does not require intravenous-line preparation, thus may be advantageous for patients in whom intravenous preparation is not trivial, that is, due to obesity, repeated vein injections or other difficulties in achieving venous access.…”