Magnetic Resonance in Medicine · 2020 · 33 citations · 31 references
MedicineComputer ScienceRadiologyBiomedical ImagingDiagnostic ImagingDiagnostic RadiologyBrain ImagingImage ReconstructionMagnetic Resonance Image ReconstructionMagnetic Resonance ImagingMedical ImagingNeuroimaging PhysicsAccuracyB0 Field Inhomogeneity CorrectionComputation SpeedConvolutional Neural NetworksDeep Learning In RadiologyMagnetic Field EffectsMagnetic Field InhomogeneityVoxelwise Quantification
Purpose To introduce a novel deep learning method for Robust and Accelerated Reconstruction (RoAR) of quantitative and B 0‐inhomogeneity‐corrected maps from multi‐gradient recalled echo (mGRE) MRI data. Methods RoAR trains a convolutional neural network (CNN) to generate quantitative maps free from field inhomogeneity artifacts by adopting a self‐supervised learning strategy given (a) mGRE magnitude images, (b) the biophysical model describing mGRE signal decay, and (c) preliminary‐evaluated F‐function accounting for contribution of macroscopic B 0 field inhomogeneities. Importantly, no ground‐truth images are required and F‐function is only needed during RoAR training but not application. Results We show that RoAR preserves all features of maps while offering significant improvements over existing methods in computation speed (seconds vs. hours) and reduced sensitivity to noise. Even for data with SNR = 5 RoAR produced maps with accuracy of 22% while voxel‐wise analysis accuracy was 47%. For SNR = 10 the RoAR accuracy increased to 17% vs. 24% for direct voxel‐wise analysis. Conclusions RoAR is trained to recognize the macroscopic magnetic field inhomogeneities directly from the input magnitude‐only mGRE data and eliminate their effect on measurements. RoAR training is based on the biophysical model and does not require ground‐truth maps. Since RoAR utilizes signal information not just from individual voxels but also accounts for spatial patterns of the signals in the images, it reduces the sensitivity of maps to the noise in the data. These features plus high computational speed provide significant benefits for the potential usage of RoAR in clinical settings.
31
Adam: A Method for Stochastic Optimization
Diederik P. Kingma, Jimmy Ba · UvA-DARE (University of Amsterdam) · 2014 · 84.8K citations · Full text
TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
Martı́n Abadi, Ashish Agarwal, Paul Barham et al. · arXiv (Cornell University) · 2016 · 9.8K citations · Full text
Victor Lempitsky, Andrea Vedaldi, Dmitry Ulyanov · 2018 · 1.9K citations
International Conference on Learning Representations (ICLR)
Maulana Abbas, Jyri Kivinen, Tapani Raiko · 2016 · 1.6K citations