Structured Sparsity Models for Brain Decoding from fMRI Data

Baldassarre, Luca; Mourão-Miranda, Janaı́na; Pontil, Massimiliano

doi:10.1109/prni.2012.31

Cited by 51 publications

(66 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Contrary to [7] and [5] that proposed to use two nested loops of proximal solvers (ISTA/FISTA), we have here a single loop. As the proximal operator for G leads to a linear system with the same operator to invert, the SVD factorization can be precomputed to speed up the computation.…”

Section: Definition 2 (Fenchel Conjugate)mentioning

confidence: 99%

“…For this purpose, we study empirically a large variety of decoding approaches on simulations where the ground truth is known. In addition, we contribute an efficient method that uses a prior specifically-crafted for our purpose, building upon previous work [5].…”

Section: Introductionmentioning

confidence: 99%

“…using sparsity inducing norms such as the ℓ 1 norm [1]. Combining both of these insights leads to consider TV-ℓ 1 penalization [5] that achieves the segmentation of a limited number of predictive brain regions when decoding from full brain data.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Identifying Predictive Regions from fMRI with TV-L1 Prior

Gramfort

Thirion

Varoquaux³

2013

2013 International Workshop on Pattern Recognition in Neuroimaging

View full text Add to dashboard Cite

Abstract-"Decoding", i.e. predicting stimulus related quantities from functional brain images, is a powerful tool to demonstrate differences between brain activity across conditions. However, unlike standard brain mapping, it offers no guaranties on the localization of this information. Here, we consider decoding as a statistical estimation problem and show that injecting a spatial segmentation prior leads to unmatched performance in recovering predictive regions. Specifically, we use ℓ1 penalization to set voxels to zero and Total-Variation (TV) penalization to segment regions. Our contribution is twofold. On the one hand, we show via extensive experiments that, amongst a large selection of decoding and brain-mapping strategies, TV+ℓ1 leads to best region recovery. On the other hand, we consider implementation issues related to this estimator. To tackle efficiently this joint prediction-segmentation problem we introduce a fast optimization algorithm based on a primal-dual approach. We also tackle automatic setting of hyper-parameters and fast computation of image operation on the irregular masks that arise in brain imaging.

show abstract

Section: Definition 2 (Fenchel Conjugate)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Identifying Predictive Regions from fMRI with TV-L1 Prior

Gramfort

Thirion

Varoquaux³

2013

2013 International Workshop on Pattern Recognition in Neuroimaging

View full text Add to dashboard Cite

show abstract

“…Total variation is a popular regularizer used to enforce local smoothness in a signal (Michel et al 2011;Rudin et al 1992;Tibshirani et al 2005). It has successfully been applied in image de-noising and has recently become of particular interest in the neural imaging community where it can be used to reconstruct sparse but locally smooth brain activation (Baldassarre et al 2012b;Michel et al 2011).Two kinds of total variation are commonly considered in the literature, isotropic T V I (w) = ∇w 2,1 and anisotropic T V A (w) = ∇w 1 (Beck and Teboulle 2009). In our theoretical analysis we focus on the anisotropic penalty.…”

Section: Introductionmentioning

confidence: 99%

Convex relaxations of penalties for sparse correlated variables with bounded total variation

2015

View full text Add to dashboard Cite

We study the problem of statistical estimation with a signal known to be sparse, spatially contiguous, and containing many highly correlated variables. We take inspiration from the recently introduced k-support norm, which has been successfully applied to sparse prediction problems with correlated features, but lacks any explicit structural constraints commonly found in machine learning and image processing. We address this problem by incorporating a total variation penalty in the k-support framework. We introduce the (k, s) support total variation norm as the tightest convex relaxation of the intersection of a set of sparsity and total variation constraints. We show that this norm leads to an intractable combinatorial graph optimization problem, which we prove to be NP-hard. We then introduce a tractable relaxation with approximation guarantees that scale well for grid structured graphs. We devise several first-order optimization strategies for statistical parameter estimation with the described penalty. We demonstrate the effectiveness of this penalty on classification in the low-sample regime, classification with M/EEG neuroimaging data, and image recovery with synthetic and real data background subtracted image recovery tasks. We extensively analyse the application of our penalty on the complex task of identifying predictive regions from low-sample high-dimensional fMRI brain data, we show that our method is particularly useful compared to existing methods in terms of accuracy, interpretability, and stability.

show abstract

“…They enforce more structured constraints on the solution, such as the discriminative voxels are grouped together in possibly few clusters [33], [34], [35], [36], [37], [38], [39], [40]. In many cases, the parcellation information is not available beforehand, and therefore, ones can either use the anatomical regions as an approx [41], or use the data driven methods to obtain the grouping information [42], [43].…”

Section: Existing Extensions Of the Plain Sparse Modelmentioning

confidence: 99%