Identifying Predictive Regions from fMRI with TV-L1 Prior

• Published in: 2013 International Workshop on Pattern Recognition in Neuroimaging pp. 17 - 20
• Main Authors: Gramfort, Alexandre, Thirion, Bertrand, Varoquaux, Gael
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2013
• Subjects: Brain; Brain mapping; Decoding; Estimation; fMRI; Predictive models; primal-dual optimization; Signal to noise ratio; sparse; supervised learning; support recovery; total-variation
• DOI: 10.1109/PRNI.2013.14

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 ℓ 1 -penalization to set voxels to zero and Total-Variation (TV) penalization to segment regions. Our contribution is two-fold. On the one hand, we show via extensive experiments that, amongst a large selection of decoding and brain-mapping strategies, TV+ℓ 1 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.