Enhanced parameter estimation in multiparametric arterial spin labeling using artificial neural networks.

PURPOSE: Multiparametric arterial spin labeling (MP-ASL) can quantify cerebral blood flow (CBF) and arterial cerebral blood volume (CBVa). However, its accuracy is compromised owing to its intrinsically low SNR, necessitating complex and time-consuming parameter estimation. Deep neural networks (DNNs) offer a solution to these limitations. Therefore, we aimed to develop simulation-based DNNs for MP-ASL and compared the performance of a supervised DNN (DNNSup), physics-informed unsupervised DNN (DNNUns), and the conventional lookup table method (LUT) using simulation and in vivo data. METHODS: MP-ASL was performed twice during resting state and once during the breath-holding task. First, the accuracy and noise immunity were evaluated in the first resting state. Second, CBF and CBVa values were statistically compared between the first resting state and the breath-holding task using the Wilcoxon signed-rank test and Cliff's delta. Finally, reproducibility of the two resting states was assessed. RESULTS: Simulation and first resting-state analyses demonstrated that DNNSup had higher accuracy, noise immunity, and a six-fold faster computation time than LUT. Furthermore, all methods detected task-induced CBF and CBVa elevations, with the effect size being larger with the DNNSup (CBF, p = 0.055, Δ = 0.286; CBVa, p = 0.008, Δ = 0.964) and DNNUns (CBF, p = 0.039, Δ = 0.286; CBVa, p = 0.008, Δ = 1.000) than that with LUT (CBF, p = 0.109, Δ = 0.214; CBVa, p = 0.008, Δ = 0.929). Moreover, all the methods exhibited comparable and satisfactory reproducibility. CONCLUSION: DNNSup outperforms DNNUns and LUT with respect to estimation performance and computation time.

Comparison between supervised and physics-informed unsupervised deep neural networks for estimating cerebral perfusion using multi-delay arterial spin labeling MRI.

This study aimed to implement a physics-informed unsupervised deep neural network (DNN) to estimate cerebral blood flow (CBF) and arterial transit time (ATT) from multi-delay arterial spin labeling (ASL), and compare its performance with that of a supervised DNN and the conventional method. Supervised and unsupervised DNNs were trained using simulation data. The accuracy and noise immunity of the three methods were compared using simulations and in vivo data. The simulation study investigated the differences between the predicted and ground-truth values and their variations with the noise level. The in vivo study evaluated the predicted values from the original images and noise-induced variations in the predicted values from the synthesized noisy images by adding Rician noise to the original images. The simulation study showed that CBF estimated using the supervised DNN was not biased by noise, whereas that estimated using other methods had a positive bias. Although the ATT with all methods exhibited a similar behavior with noise increase, the ATT with the supervised DNN was less biased. The in vivo study showed that CBF and ATT with the supervised DNN were the most accurate and that the supervised and unsupervised DNNs had the highest noise immunity in CBF and ATT estimations, respectively. Physics-informed unsupervised learning can estimate CBF and ATT from multi-delay ASL signals, and its performance is superior to that of the conventional method. Although noise immunity in ATT estimation was superior with unsupervised learning, other performances were superior with supervised learning.

Effects of the Training Data Condition on Arterial Spin Labeling Parameter Estimation Using a Simulation-Based Supervised Deep Neural Network.

OBJECTIVE: A simulation-based supervised deep neural network (DNN) can accurately estimate cerebral blood flow (CBF) and arterial transit time (ATT) from multidelay arterial spin labeling signals. However, the performance of deep learning depends on the characteristics of the training data set. We aimed to investigate the effects of the ground truth (GT) ranges of CBF and ATT on the performance of the DNN when training data were prepared using arterial spin labeling signal simulation. METHODS: Deep neural networks were individually trained using 36 patterns of the training data sets. Simulation test data (1,000,000 points), 17 healthy volunteers, and 1 patient with moyamoya disease were included. The simulation test data were used to evaluate accuracy, precision, and noise immunity of the DNN. The best-performing DNN was determined by the normalized mean absolute error (NMAE), normalized root mean squared error (NRMSE), and normalized coefficient of variation over repeated training (CV Net ). Cerebral blood flow and ATT values and their histograms were compared between the GT and predicted values. For the in vivo data, the dependency of the predicted values on the GT ranges was visually evaluated by comparing CBF and ATT maps between the best-performing DNN and the other DNNs. Moreover, using the synthesized noisy images, noise immunity was compared between the best-performing DNN based on the simulation study and a conventional method. RESULTS: The simulation study showed that a network trained by the GT of CBF and ATT in the ranges of 0 to 120 mL/100 g/min and 0 to 4500 milliseconds, respectively, had the highest performance (NMAE CBF , 0.150; NRMSE CBF , 0.231; CV NET CBF , 0.028; NMAE ATT , 0.158; NRMSE ATT , 0.257; and CV NET ATT , 0.028). Although the predicted CBF and ATT varied with the GT range of the training data sets, the appropriate settings preserved the accuracy, precision, and noise immunity of the DNN. In addition, the same results were observed in in vivo studies. CONCLUSIONS: The GT ranges to prepare the training data affected the performance of the simulation-based supervised DNNs. The predicted CBF and ATT values depended on the GT range; inappropriate settings degraded the accuracy, whereas appropriate settings of the GT range provided accurate and precise estimates.