A Constrained ICA-EMD Model for Group Level fMRI Analysis

Abstract

Independent component analysis (ICA), as a data driven method, has shown to be a powerful tool for functional magnetic resonance imaging (fMRI) data analysis. One drawback of this multivariate approach is, that it is not compatible to the analysis of group data in general. Therefore various techniques have been proposed in order to overcome this limitation of ICA. In this paper a novel ICA-based work-flow for extracting resting state networks from fMRI group studies is proposed. An empirical mode decomposition (EMD) is used to generate reference signals in a data driven manner, which can be incorporated into a constrained version of ICA (cICA), what helps to eliminate the inherent ambiguities of ICA. The results of the proposed workflow are then compared to those obtained by a widely used group

Functional Network of Neurocognitive Development in Correlations of BOLD Signals

“…The attention to any single task produces a consistent T R (response time) but during task-free and natural viewing the children are able to focus much attention on the movie and adults focus less attention [4].In [5,6] authors have discussed the idea of general linear model (GLM ) approach for the neural activity triggers changes and closely related to hemodynamic response. Several methods from [11] [12] have been proposed to analyze the task-based f M RIsuch as data driven approaches like Independent Component Analysis (ICA), General Linear Model (GLM ) but this model has some drawback, and Multi-voxel pattern analysis (M V P A) it is using the pattern classification techniques [7] [8,9].A Mean Square Prediction Error (M SE) is used to examine the performance of model to predict and measure the (BOLD) signal [13]. Spatial and temporal independent component analysis of f M RIdata are simulated and produced the time course of the signal [15].Multiple kinds of white matter regions are discussed in [23].…”

Brain Connectivity Studies on Structure‐Function Relationships: A Short Survey with an Emphasis on Machine Learning

“…The results revealed robust effects and suggested that the established analysis pipeline could form a useful baseline for investigations of human brain networks. Recently, we proposed a hybrid cICA-EMD approach, where a bidimensional ensemble empirical mode decomposition technique based on Green's functions in tension (GiT-BEEMD) was used to create reference signals for a constrained ICA [ 165 , 166 ]. The idea of this technique is to decompose a signal into its underlying intrinsic frequency compartments [ 56 ], reflecting frequency-specific aspects of the latter.…”