Learning Macroscopic Brain Connectomes via Group-Sparse Factorization

Mapping structural brain connectomes for living human brains typically requires expert analysis and rule-based models on diffusion-weighted magnetic resonance imaging. A data-driven approach, however, could overcome limitations in such rule-based approaches and improve precision mappings for individuals. In this work, we explore a framework that facilitates applying learning algorithms to automatically extract brain connectomes. Using a tensor encoding, we design an objective with a group-regularizer that prefers biologically plausible fascicle structure. We show that the objective is convex and has unique solutions, ensuring identifiable connectomes for an individual. We develop an efficient optimization strategy for this extremely high-dimensional sparse problem, by reducing the number of parameters using a greedy algorithm designed specifically for the problem. We show that this greedy algorithm significantly improves on a standard greedy algorithm, called Orthogonal Matching Pursuit. We conclude with an analysis of the solutions found by our method, showing we can accurately reconstruct the diffusion information while maintaining contiguous fascicles with smooth direction changes.

[1]  Andrzej Cichocki,et al.  Tensor Decompositions for Signal Processing Applications: From two-way to multiway component analysis , 2014, IEEE Signal Processing Magazine.

[2]  Alan Connelly,et al.  MRtrix: Diffusion tractography in crossing fiber regions , 2012, Int. J. Imaging Syst. Technol..

[3]  O. Sporns,et al.  Rich-Club Organization of the Human Connectome , 2011, The Journal of Neuroscience.

[4]  Y. C. Pati,et al.  Orthogonal matching pursuit: recursive function approximation with applications to wavelet decomposition , 1993, Proceedings of 27th Asilomar Conference on Signals, Systems and Computers.

[5]  Naoki Abe,et al.  Grouped Orthogonal Matching Pursuit for Variable Selection and Prediction , 2009, NIPS.

[6]  Jean-Philippe Thiran,et al.  Reducing false positives in tractography with microstructural and anatomical priors , 2019 .

[7]  Dietrich von Rosen,et al.  The multilinear normal distribution: Introduction and some basic properties , 2013, J. Multivar. Anal..

[8]  Lars Kai Hansen,et al.  Algorithms for Sparse Nonnegative Tucker Decompositions , 2008, Neural Computation.

[9]  Peter F. Neher,et al.  Learn to Track: Deep Learning for Tractography , 2017, bioRxiv.

[10]  Peter F. Neher,et al.  A Machine Learning Based Approach to Fiber Tractography Using Classifier Voting , 2015, MICCAI.

[11]  Olaf Sporns,et al.  Comparative Connectomics , 2016, Trends in Cognitive Sciences.

[12]  Tamara G. Kolda,et al.  Tensor Decompositions and Applications , 2009, SIAM Rev..

[13]  Ariel Rokem,et al.  Evaluating the Accuracy of Diffusion MRI Models in White Matter , 2015, PloS one.

[14]  Rémi Gribonval,et al.  Dynamic Screening: Accelerating First-Order Algorithms for the Lasso and Group-Lasso , 2014, IEEE Transactions on Signal Processing.

[15]  Misha Elena Kilmer,et al.  A tensor-based dictionary learning approach to tomographic image reconstruction , 2015, BIT Numerical Mathematics.

[16]  Michael Elad,et al.  Efficient Implementation of the K-SVD Algorithm using Batch Orthogonal Matching Pursuit , 2008 .

[17]  Rachid Deriche,et al.  Diffusion MRI signal reconstruction with continuity constraint and optimal regularization , 2012, Medical Image Anal..

[18]  Andrzej Cichocki,et al.  Computing Sparse Representations of Multidimensional Signals Using Kronecker Bases , 2013, Neural Computation.

[19]  Timothy Edward John Behrens,et al.  Characterization and propagation of uncertainty in diffusion‐weighted MR imaging , 2003, Magnetic resonance in medicine.

[20]  O. Sporns,et al.  Network neuroscience , 2017, Nature Neuroscience.

[21]  P. Basser,et al.  In vivo fiber tractography using DT‐MRI data , 2000, Magnetic resonance in medicine.

[22]  Genevera I. Allen,et al.  Sparse Higher-Order Principal Components Analysis , 2012, AISTATS.

[23]  Alan Connelly,et al.  Direct estimation of the fiber orientation density function from diffusion-weighted MRI data using spherical deconvolution , 2004, NeuroImage.

[24]  Syed Zubair,et al.  Tensor dictionary learning with sparse TUCKER decomposition , 2013, 2013 18th International Conference on Digital Signal Processing (DSP).

[25]  Andrew J. Saykin,et al.  Unified representation of tractography and diffusion-weighted MRI data using sparse multidimensional arrays , 2017, NIPS.

[26]  Zemin Zhang,et al.  Denoising and Completion of 3D Data via Multidimensional Dictionary Learning , 2015, IJCAI.

[27]  Abhimanyu Das,et al.  Submodular meets Spectral: Greedy Algorithms for Subset Selection, Sparse Approximation and Dictionary Selection , 2011, ICML.

[28]  B. Wandell.  Clarifying Human White Matter. , 2016, Annual review of neuroscience.

[29]  Andrzej Cichocki,et al.  Nonnegative Matrix and Tensor Factorization T , 2007 .

[30]  Wotao Yin,et al.  A Block Coordinate Descent Method for Regularized Multiconvex Optimization with Applications to Nonnegative Tensor Factorization and Completion , 2013, SIAM J. Imaging Sci..

[31]  Tammy Riklin-Raviv,et al.  DeepTract: A Probabilistic Deep Learning Framework for White Matter Fiber Tractography , 2018, MICCAI.

[32]  Hao Xu,et al.  Learning Sparse Representations of High Dimensional Data on Large Scale Dictionaries , 2011, NIPS.