Experimental design for MRI by greedy policy search

In today's clinical practice, magnetic resonance imaging (MRI) is routinely accelerated through subsampling of the associated Fourier domain. Currently, the construction of these subsampling strategies - known as experimental design - relies primarily on heuristics. We propose to learn experimental design strategies for accelerated MRI with policy gradient methods. Unexpectedly, our experiments show that a simple greedy approximation of the objective leads to solutions nearly on-par with the more general non-greedy approach. We offer a partial explanation for this phenomenon rooted in greater variance in the non-greedy objective's gradient estimates, and experimentally verify that this variance hampers non-greedy models in adapting their policies to individual MR images. We empirically show that this adaptivity is key to improving subsampling designs.

[1]  Jin Keun Seo,et al.  Deep learning for undersampled MRI reconstruction , 2017, Physics in medicine and biology.

[2]  Max Welling,et al.  Buy 4 REINFORCE Samples, Get a Baseline for Free! , 2019, DeepRLStructPred@ICLR.

[3]  Jan-Jakob Sonke,et al.  Recurrent inference machines for reconstructing heterogeneous MRI data , 2019, Medical Image Anal..

[4]  Daniel Rueckert,et al.  A Deep Cascade of Convolutional Neural Networks for MR Image Reconstruction , 2017, IPMI.

[5]  Michael Unser,et al.  Self-Supervised Deep Active Accelerated MRI , 2019, ArXiv.

[6]  Iris A.M. Huijben,et al.  Learning Sampling and Model-Based Signal Recovery for Compressed Sensing MRI , 2020, ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[7]  Matthias W. Seeger,et al.  Compressed sensing and Bayesian experimental design , 2008, ICML '08.

[8]  Michael G. Rabbat,et al.  Advancing machine learning for MR image reconstruction with an open competition: Overview of the 2019 fastMRI challenge , 2020, Magnetic resonance in medicine.

[9]  Pascal Vincent,et al.  fastMRI: An Open Dataset and Benchmarks for Accelerated MRI , 2018, ArXiv.

[10]  Carola-Bibiane Schönlieb,et al.  Learning the Sampling Pattern for MRI , 2019, IEEE Transactions on Medical Imaging.

[11]  Patrick Putzky,et al.  Invert to Learn to Invert , 2019, NeurIPS.

[12]  Volkan Cevher,et al.  Scalable Learning-Based Sampling Optimization for Compressive Dynamic MRI , 2019, ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[13]  Thomas Pock,et al.  Learning a variational network for reconstruction of accelerated MRI data , 2017, Magnetic resonance in medicine.

[14]  Leslie Ying,et al.  Accelerating magnetic resonance imaging via deep learning , 2016, 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI).

[15]  Peter L. Bartlett,et al.  Infinite-Horizon Policy-Gradient Estimation , 2001, J. Artif. Intell. Res..

[16]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[17]  Léon Bottou,et al.  Wasserstein Generative Adversarial Networks , 2017, ICML.

[18]  Demis Hassabis,et al.  Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm , 2017, ArXiv.

[19]  Bernhard Schölkopf,et al.  Optimization of k‐space trajectories for compressed sensing by Bayesian experimental design , 2010, Magnetic resonance in medicine.

[20]  Ronald J. Williams,et al.  Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement Learning , 2004, Machine Learning.

[21]  Andreas Krause,et al.  Adaptive Submodularity: Theory and Applications in Active Learning and Stochastic Optimization , 2010, J. Artif. Intell. Res..

[22]  Levi Boyles,et al.  Statistical Tests for Optimization Efficiency , 2011, NIPS.

[23]  Jaganathan Vellagoundar,et al.  A robust adaptive sampling method for faster acquisition of MR images. , 2015, Magnetic resonance imaging.

[24]  Russ Tedrake,et al.  Signal-to-Noise Ratio Analysis of Policy Gradient Algorithms , 2008, NIPS.

[25]  Volkan Cevher,et al.  Learning-Based Compressive MRI , 2018, IEEE Transactions on Medical Imaging.

[26]  Sergey Levine,et al.  High-Dimensional Continuous Control Using Generalized Advantage Estimation , 2015, ICLR.

[27]  Kaito Fujii,et al.  Beyond Adaptive Submodularity: Approximation Guarantees of Greedy Policy with Adaptive Submodularity Ratio , 2019, ICML.

[28]  Pascal Vincent,et al.  Reducing Uncertainty in Undersampled MRI Reconstruction With Active Acquisition , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[29]  Aaron Defazio,et al.  End-to-End Variational Networks for Accelerated MRI Reconstruction , 2020, MICCAI.

[30]  Daeun Kim,et al.  OEDIPUS: An Experiment Design Framework for Sparsity-Constrained MRI , 2018, IEEE Transactions on Medical Imaging.

[31]  Michael Unser,et al.  Deep Convolutional Neural Network for Inverse Problems in Imaging , 2016, IEEE Transactions on Image Processing.

[32]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[33]  Mert R. Sabuncu,et al.  Learning-based Optimization of the Under-sampling Pattern in MRI , 2019, IPMI.

[34]  S. Frick,et al.  Compressed Sensing , 2014, Computer Vision, A Reference Guide.

[35]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[36]  Adriana Romero,et al.  Active MR k-space Sampling with Reinforcement Learning , 2020, MICCAI.

[37]  Morteza Mardani,et al.  Compressed Sensing: From Research to Clinical Practice with Data-Driven Learning , 2019, ArXiv.

[38]  Bernhard Schölkopf,et al.  Bayesian Experimental Design of Magnetic Resonance Imaging Sequences , 2008, NIPS.

[39]  Alex Kendall,et al.  What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision? , 2017, NIPS.

[40]  Volkan Cevher,et al.  Rethinking Sampling in Parallel MRI: A Data-Driven Approach , 2019, 2019 27th European Signal Processing Conference (EUSIPCO).

[41]  Max Welling,et al.  i-RIM applied to the fastMRI challenge , 2019, ArXiv.

[42]  Oren N Jaspan,et al.  Compressed sensing MRI: a review of the clinical literature. , 2015, The British journal of radiology.

[43]  D. Donoho,et al.  Sparse MRI: The application of compressed sensing for rapid MR imaging , 2007, Magnetic resonance in medicine.

[44]  Emmanuel J. Candès,et al.  Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information , 2004, IEEE Transactions on Information Theory.

[45]  Max Welling,et al.  Estimating Gradients for Discrete Random Variables by Sampling without Replacement , 2020, ICLR.

[46]  Volkan Cevher,et al.  Closed loop deep Bayesian inversion: Uncertainty driven acquisition for fast MRI , 2019 .