TY - JOUR
T1 - Recurrent inference machines as inverse problem solvers for MR relaxometry
AU - Sabidussi, E. R.
AU - Klein, S.
AU - Caan, M. W. A.
AU - Bazrafkan, S.
AU - den Dekker, A. J.
AU - Sijbers, J.
AU - Niessen, W. J.
AU - Poot, D. H. J.
N1 - Funding Information: This work is part of the project B-QMINDED which has received funding from the European Union's Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 764513. Funding Information: This work is part of the project B-QMINDED which has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 764513. Publisher Copyright: © 2021 The Author(s)
PY - 2021/12/1
Y1 - 2021/12/1
N2 - In this paper, we propose the use of Recurrent Inference Machines (RIMs) to perform T1 and T2 mapping. The RIM is a neural network framework that learns an iterative inference process based on the signal model, similar to conventional statistical methods for quantitative MRI (QMRI), such as the Maximum Likelihood Estimator (MLE). This framework combines the advantages of both data-driven and model-based methods, and, we hypothesize, is a promising tool for QMRI. Previously, RIMs were used to solve linear inverse reconstruction problems. Here, we show that they can also be used to optimize non-linear problems and estimate relaxometry maps with high precision and accuracy. The developed RIM framework is evaluated in terms of accuracy and precision and compared to an MLE method and an implementation of the Residual Neural Network (ResNet). The results show that the RIM improves the quality of estimates compared to the other techniques in Monte Carlo experiments with simulated data, test-retest analysis of a system phantom, and in-vivo scans. Additionally, inference with the RIM is 150 times faster than the MLE, and robustness to (slight) variations of scanning parameters is demonstrated. Hence, the RIM is a promising and flexible method for QMRI. Coupled with an open-source training data generation tool, it presents a compelling alternative to previous methods.
AB - In this paper, we propose the use of Recurrent Inference Machines (RIMs) to perform T1 and T2 mapping. The RIM is a neural network framework that learns an iterative inference process based on the signal model, similar to conventional statistical methods for quantitative MRI (QMRI), such as the Maximum Likelihood Estimator (MLE). This framework combines the advantages of both data-driven and model-based methods, and, we hypothesize, is a promising tool for QMRI. Previously, RIMs were used to solve linear inverse reconstruction problems. Here, we show that they can also be used to optimize non-linear problems and estimate relaxometry maps with high precision and accuracy. The developed RIM framework is evaluated in terms of accuracy and precision and compared to an MLE method and an implementation of the Residual Neural Network (ResNet). The results show that the RIM improves the quality of estimates compared to the other techniques in Monte Carlo experiments with simulated data, test-retest analysis of a system phantom, and in-vivo scans. Additionally, inference with the RIM is 150 times faster than the MLE, and robustness to (slight) variations of scanning parameters is demonstrated. Hence, the RIM is a promising and flexible method for QMRI. Coupled with an open-source training data generation tool, it presents a compelling alternative to previous methods.
KW - Deep learning
KW - Mapping
KW - Quantitative MRI
KW - Recurrent inference machines
KW - Relaxometry
UR - http://www.scopus.com/inward/record.url?scp=85115034262&partnerID=8YFLogxK
U2 - https://doi.org/10.1016/j.media.2021.102220
DO - https://doi.org/10.1016/j.media.2021.102220
M3 - Article
C2 - 34543912
SN - 1361-8415
VL - 74
JO - Medical Image Analysis
JF - Medical Image Analysis
M1 - 102220
ER -