Model-based estimation of dynamic functional connectivity in resting-state functional magnetic resonance imaging

Abstract

Recently, we have witnessed an increase in scientific interest in understanding the dynamic nature of brain networks by evaluating dynamic functional connectivity (FC) using functional magnetic resonance imaging (fMRI). In this work, we introduce two multivariate volatility models, standardized dynamic conditional correlation, and standardized exponentially weighted moving average, both of which are built upon the framework of dynamic conditional correlation and exponentially weighted moving average models, respectively. In these two models, we use standardized residuals with the goal of determining whether the use of standardized residuals reduces the mean square rate error. Moreover, in traditional simulation studies, time series were considered with zero conditional expectation and static conditional variance which do not capture the nature of the real data. This is because of hemodynamic response function in the brain and dynamic functional connectivity of each brain region with itself during the experiment time, respectively. That is why, next, some new simulation studies are introduced which are more similar to blood-oxygen-level-dependent time series of brain regions. Then, methods' proficiency is analyzed using previous and new simulation studies. Results show that, in both former and latter simulations, the new methods work better. Finally, the best model is utilized to model FC in an Iranian resting-state fMRI data.

Modified models and simulations for estimating dynamic functional connectivity in resting state functional magnetic resonance imaging

“…Three simulations are used in this study. Simulations 1 and 2 are similar to those used by Lindquist et al and Behboudi et al, respectively. Simulation 3 is proposed in this study, in which the conditional mean is not equal to zero, conditional variance is not assumed to be a constant number, and data are generated by a bivariate nonnormal distribution.…”

“…Results of using KDCC for the estimation of dFC on Iranian data reveal that dFC is not the same in all regions and participants. These changes are also observed in the use of DCC and SDCC for the estimation of dFC because we had no task during rs‐fMRI experiments.…”

“…Lindquist et al showed that the DCC model has the best performance, and its performance is as good as that of OSW method. The SDCC model has better performance than DCC. Based on the simulations presented here, KDCC outperforms all other methods for all types of the data (bivariate normal and nonnormal).…”

“…Then, using the resulting values for , the covariance matrix ∑ t was produced as follows: where, similar to simulations utilized by Lindquist et al and Behboudi et al, σ 12, t was assumed in the following nine forms, which were divided into three groups: σ 12, t = 0 for t = 1, …, 600 …”

“…For instance, Choe et al showed that the reliability of DCC model is better than that of nonparametric models such as sliding window (SW) and tapered sliding window (TSW). Moreover, Behboudi et al corrected the simulation studies employed in the Lindquist et al and proposed two models of standardized DCC (SDCC) and standardized EWMA (SEWMA) based on DCC and EWMA. Their analyses show that the level of mean square error (MSE) is lower in the proposed model than in the two noted models.…”