Title :
Model selection for hemodynamic brain parcellation in FMRI
Author :
Albughdadi, Mohanad ; Chaari, Lamia ; Forbes, Florence ; Tourneret, Jean-Yves ; Ciuciu, Philippe
Author_Institution :
IRIT, Univ. of Toulouse, Toulouse, France
Abstract :
Brain parcellation into a number of hemodynamically homogeneous regions (parcels) is a challenging issue in fMRI analyses. This task has been recently integrated in the joint detection estimation [1] resulting in the so-called joint parcellation detection estimation (JPDE) model [2]. JPDE automatically estimates the parcels from the fMRI data but requires the desired number of parcels to be fixed. This is potentially critical in that the chosen number of parcels may influence detection-estimation performance. In this paper, we propose a model selection procedure to automatically set the number of parcels from the data. The selection procedure relies on the calculation of the free energy corresponding to each concurrent model, within the variational expectation maximization framework. Experiments on synthetic and real fMRI data demonstrate the ability of the proposed procedure to select the optimal number of parcels.
Keywords :
biomedical MRI; brain; expectation-maximisation algorithm; feature extraction; feature selection; free energy; haemodynamics; medical image processing; neurophysiology; optimisation; parameter estimation; variational techniques; FMRI analysis; JPDE model; automatic parcel estimation; automatic parcel number setting; constant desired parcel number; detection-estimation performance; free energy calculation; hemodynamic brain parcellation; hemodynamically homogeneous regions; joint detection estimation; joint parcellation detection estimation model; model selection; optimal parcel number selection; parcel number selection effect; variational expectation maximization framework; Bayes methods; Brain modeling; Data models; Educational institutions; Estimation; Hemodynamics; Joints; JDE; JPDE; Parcellation; VEM; fMRI;
Conference_Titel :
Signal Processing Conference (EUSIPCO), 2014 Proceedings of the 22nd European
Conference_Location :
Lisbon