fMRI Deconvolution via Temporal Regularization using a LASSO model and the LARS algorithm - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2018

fMRI Deconvolution via Temporal Regularization using a LASSO model and the LARS algorithm

Abstract

In the context of functional MRI (fMRI), methods based on the deconvolution of the blood oxygenated level dependent (BOLD) signal have been developed to investigate the brain activity, without a need of a priori knowledge about activations occurrence [2]. In this work, we propose a novel temporal regularized deconvolution of the BOLD signal using the Least Absolute Shrinkage and Selection Operator (LASSO) model, solved by means of the Least-Angle Regression (LARS) algorithm. In this way, we were able to recover the underlying neurons activations and their dynamics.
Fichier principal
Vignette du fichier
EMBC18_3334_MS-1.pdf (4.26 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01855467 , version 1 (08-08-2018)

Identifiers

  • HAL Id : hal-01855467 , version 1

Cite

Isa Costantini, Patryk Filipiak, Kostiantyn Maksymenko, Samuel Deslauriers-Gauthier, Rachid Deriche. fMRI Deconvolution via Temporal Regularization using a LASSO model and the LARS algorithm. EMBC'18 - 40th International Engineering in Medicine and Biology Conference, Jul 2018, Honolulu, United States. ⟨hal-01855467⟩
152 View
76 Download

Share

Gmail Facebook X LinkedIn More