Proposed (MoDL-SToRM)

Slides:



Advertisements
Similar presentations
Pattern Recognition and Machine Learning
Advertisements

Bayesian Belief Propagation
In Chan Song, Ph.D. Seoul National University Hospital
Edge Preserving Image Restoration using L1 norm
L1 sparse reconstruction of sharp point set surfaces
Principal Component Analysis Based on L1-Norm Maximization Nojun Kwak IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008.
SE263 Video Analytics Course Project Initial Report Presented by M. Aravind Krishnan, SERC, IISc X. Mei and H. Ling, ICCV’09.
Evaluation of Reconstruction Techniques
B.Macukow 1 Lecture 12 Neural Networks. B.Macukow 2 Neural Networks for Matrix Algebra Problems.
More MR Fingerprinting
Master thesis by H.C Achterberg
RBF Neural Networks x x1 Examples inside circles 1 and 2 are of class +, examples outside both circles are of class – What NN does.
Probabilistic video stabilization using Kalman filtering and mosaicking.
Markus Strohmeier Sparse MRI: The Application of
3 Bayes' theorem 2 … Partial discreteness: a new type of prior knowledge for MRI reconstruction Gabriel Ramos-Llordén1, Hilde Segers1, Willem.
Jason P. Stockmann 1 and R. Todd Constable 1,2 Yale University, Department of Biomedical Engineering 1, Department of Diagnostic Radiology 2, New Haven,
Partial Parallel imaging (PPI) in MR for faster imaging IMA Compressed Sensing June, 2007 Acknowledgement: NIH Grants 5RO1CA and 5P41RR008079, Pierre-Francois.
EE369C Final Project: Accelerated Flip Angle Sequences Jan 9, 2012 Jason Su.
Contact: Beyond Low Rank + Sparse: Multi-scale Low Rank Reconstruction for Dynamic Contrast Enhanced Imaging Frank Ong1, Tao Zhang2,
Learning to perceive how hand-written digits were drawn Geoffrey Hinton Canadian Institute for Advanced Research and University of Toronto.
11 1 Backpropagation Multilayer Perceptron R – S 1 – S 2 – S 3 Network.
Rank Minimization for Subspace Tracking from Incomplete Data
Neural Networks Presented by M. Abbasi Course lecturer: Dr.Tohidkhah.
Zuxuan Wu, Xi Wang, Yu-Gang Jiang, Hao Ye, Xiangyang Xue
Declaration of Relevant Financial Interests or Relationships David Atkinson: I have no relevant financial interest or relationship to disclose with regard.
Neural Networks The Elements of Statistical Learning, Chapter 12 Presented by Nick Rizzolo.
Nicole Seiberlich Workshop on Novel Reconstruction Strategies in NMR and MRI 2010 Göttingen, Germany 10 September 2010 Non-Cartesian Parallel Imaging based.
A Fast Algorithm for Structured Low-Rank Matrix Recovery with Applications to Undersampled MRI Reconstruction Greg Ongie*, Mathews Jacob Computational.
Higher Degree Total Variation for 3-D Image Recovery Greg Ongie*, Yue Hu, Mathews Jacob Computational Biomedical Imaging Group (CBIG) University of Iowa.
Super-resolution MRI Using Finite Rate of Innovation Curves Greg Ongie*, Mathews Jacob Computational Biomedical Imaging Group (CBIG) University of Iowa.
Medical Applications of Signal Processing Research Memory Data Joint Image Reconstruction and Field Map Estimation in MRI H.M. Nguyen,
Neural networks and support vector machines
Deep Learning Methods For Automated Discourse CIS 700-7
RNNs: An example applied to the prediction task
Stencil-based Discrete Gradient Transform Using
CS 388: Natural Language Processing: LSTM Recurrent Neural Networks
Artificial Neural Networks
Deep Neural Net Scenery Generation
Deep Predictive Model for Autonomous Driving
Highly Undersampled 0-norm Reconstruction
Multimodal Learning with Deep Boltzmann Machines
Compositional Human Pose Regression
Unrolling: A principled method to develop deep neural networks
Intelligent Information System Lab
Master’s Thesis defense Ming Du Advisor: Dr. Yi Shang
RNNs: Going Beyond the SRN in Language Prediction
Face Recognition with Deep Learning Method
Two-Stream Convolutional Networks for Action Recognition in Videos
Towards Understanding the Invertibility of Convolutional Neural Networks Anna C. Gilbert1, Yi Zhang1, Kibok Lee1, Yuting Zhang1, Honglak Lee1,2 1University.
Collaborative Filtering Matrix Factorization Approach
Incremental Training of Deep Convolutional Neural Networks
Instance Based Learning
Oral presentation for ACM International Conference on Multimedia, 2014
Single Image Rolling Shutter Distortion Correction
Image to Image Translation using GANs
Neural Networks Geoff Hulten.
Biointelligence Laboratory, Seoul National University
Backpropagation.
Volume 11, Issue 2, Pages (April 2016)
CNN-based Action Recognition Using Adaptive Multiscale Depth Motion Maps And Stable Joint Distance Maps Junyou He, Hailun Xia, Chunyan Feng, Yunfei Chu.
Machine Learning based Data Analysis
A Dynamic System Analysis of Simultaneous Recurrent Neural Network
Neural networks (1) Traditional multi-layer perceptrons
Compressive Image Recovery using Recurrent Generative Model
Backpropagation.
Automatic Handwriting Generation
Introduction to Neural Networks
VERY DEEP CONVOLUTIONAL NETWORKS FOR LARGE-SCALE IMAGE RECOGNITION
Unrolling the shutter: CNN to correct motion distortions
Learned Convolutional Sparse Coding
Presentation transcript:

Proposed (MoDL-SToRM) MODEL BASED DEEP LEARNING IN FREE BREATHING, UNGATED, CARDIAC MRI RECOVERY Sampurna Biswas, Hemant Aggarwal, Sunrita Poddar & Mathews Jacob; University of Iowa Contact: sampurna-biswas@uiowa.edu Inverse problems in MRI are ill posed: due to undersampling Model based deep learning: MoDL MoDL: can be trained with less training data Related work -I Related work -II MoDL: CG within network, improved performance CNN : learns the noise Clean data Image of 0s Model based solution A captures physics/ “model” Regularizer needed ‘learnable’ Generative adversarial networks, Mardani et al, 17 Alternating minimization, subproblems: Noise Noise CNN: CG: Learning based recovery: , learned CNN local redundancies Learns local features Captures acquisition Variational networks, Hammernik et al. 17 may not be data consistent!! Cascade networks, Schlemper et al. 17 Unrolling Many gradient descent steps Large network on unrolling GPU memory constraints Conjugate Gradient has faster convergence than GD needs fewer alternating steps Different networks at each iteration: not consistent with model based framework NO weight sharing = increase complexity of networks at each iteration Large network = needs more training data Aggarwal et al.,  MoDL, arXiv, Nov ‘17, Qin et al., CRNN, arXiv, ‘18 Goal: accelerated free breathing, ungated cardiac MR MoDL-SToRM Model based recovery with SToRM Current methods Proposed (MoDL-SToRM) Static; Breath-held or gated free breathing, ungated acceleration of 11x realistic FB accelerations Cartesian. random trajectory mostly Radial, golden angle Most assume single coil acquisition exploits multiple coils Most forward operators with analytical inverse forward operators with non-analytical inverse Mostly, no weight sharing along the iterations enforces weight sharing Mostly, no subject specific priors Can include scan specific priors! Local priors only Non-local priors! State of the art in free breathing, ungated cardiac MR Model based reconstruction : SToRM (SmooThness Regularization on Manifolds) Images in similar phases 5 s dissimilar phases SToRM 𝑨 𝒊 : time varying lines + fixed radial navigators Multi-coil time varying Fourier sampler 45 s/long SToRM +DL SToRM prior: 5 s/short Real time recon W: weight matrix is subject specific, computed using navigators #frames × #frames Implicit “soft” binning: weighting images based on their cardiac & respiratory phases Poddar et al., SToRM, TMI 2016 Long scan time ! Need: 𝑊 45𝑠 → 𝑊 5𝑠 MRI forward model : not invertible analytically Integrating MoDL with SToRM: shorter scan time Proposed iterative model Training: with alternating minimization 3D batch of 512 x 512 x 7 (limited by GPU) Target: SToRM recon from 45s acquisition Input: ZF recon from 5s acquisition W global information! 512 x 512 x 91 CG iterations in DC block 𝜆 1 , 𝜆 2 : trainable Storm: subject dependent, non-local redundancies CNN: local, spatio-temporal redundancies DC captures sampling & coil sensitivities Not invertible analytically Training details Error images & time profiles Conclusion Reconstruction at different cardiac & respiratory phases Dataset Sampling : 6 golden angle + 4 uniform radial navigators Time : 900 frames ~ 45s; Size: 512x512x900 Trained on 4 subjects ; Tested on 3 Network 3 layer (64-64-2), 1st 2 layers, 3x3 last, capacity:1e6, shallow Adam on squared error loss N = 3 repetitions Test time: 5.2s/N to reconstruct 512x512x91 (5s); real time recovery! Denoising on SToRM 5s 3) 3 L CNN 1) SToRM 45s 2) SToRM 5s MoDL-SToRM 4) Proposed 5s Limited training data Shallow recurrent network Parallel-MRI forward model integrated with a CNN ! DC + CNN/local + scan specific /non-local constraints Prospective data with multi-coil acquisition SToRM 45s SToRM 5s SToRM 5s CNN Proposed 5s Respiratory phase 1 5 s SToRM 45 s/long SToRM +DL 5 s/short Real time recon desirable reconstruction from accelerated acquisition in real time Future: navigator-less acquisition! Cardiac phase 1