Advances and Applications in Statistics
Volume 53, Issue 1, Pages 29 - 69
(July 2018) http://dx.doi.org/10.17654/AS053010029 |
|
A SYSTEMATIC REVIEW OF SIMULATION PROCEDURES FOR fMRI CONNECTIVITY STUDIES
Maribel Peró-Cebollero, Joan Guà rdia-Olmos and Núria Mancho-Fora
|
Abstract: The strategies used to study functional or effective connectivity in fMRI data are mainly based on the application of correlation studies, structural equation models (SEM), dynamic causal modelling (DCM), or the Granger causality model (GCM), while some contributions focus their attention on simulation studies. Although the tradition is scarce, this increase of the latter studies has become steeper in the last five years. In this work, we present a systematic study and analysis of simulation studies with fMRI data for the analysis of brain connectivity. We conducted a search on the Web of Science (WoS) and PubMed and eventually we reviewed a total of 134 studies. The most remarkable finding is a lack of information on the simulation procedure. For example, 17 works did not specify the model used to generate the signal, 36 did not indicate the model’s white noise addition in the signal generated, and 52 did not detail the design under which the data had been generated. Under these circumstances, it is difficult to compare the different contributions in order to identify the best strategies to simulate data for the study of brain connectivity in fMRI works. However, it is important to note the emergence of the so-called third-generation simulation models, which consider the brain as a complex, dynamical system model. This kind of model to simulate brain activity will change the state of the art in this matter, and it might be a good tool to assess the different analytical procedures to study effective connectivity. |
Keywords and phrases: fMRI, simulation studies, functional connectivity, effective connectivity, brain signal. |
|
Number of Downloads: 452 | Number of Views: 3746 |
|