Limits...
An empirical Bayes normalization method for connectivity metrics in resting state fMRI.

Chen S, Kang J, Wang G - Front Neurosci (2015)

Bottom Line: Moreover, the normalization function maps the original connectivity metrics to values between zero and one, which is well-suited for the graph theory based network analysis and avoids the information loss due to the (negative value) hard thresholding step.We apply the normalization method to a simulation study and the simulation results show that our normalization method effectively improves the robustness and reliability of the quantification of brain functional connectivity and provides more powerful group difference (biomarkers) detection.We illustrate our method on an analysis of a rs-fMRI dataset from the Autism Brain Imaging Data Exchange (ABIDE) study.

View Article: PubMed Central - PubMed

Affiliation: Department of Epidemiology and Biostatistics, University of Maryland College Park, MD, USA.

ABSTRACT
Functional connectivity analysis using resting-state functional magnetic resonance imaging (rs-fMRI) has emerged as a powerful technique for investigating functional brain networks. The functional connectivity is often quantified by statistical metrics (e.g., Pearson correlation coefficient), which may be affected by many image acquisition and preprocessing steps such as the head motion correction and the global signal regression. The appropriate quantification of the connectivity metrics is essential for meaningful and reproducible scientific findings. We propose a novel empirical Bayes method to normalize the functional brain connectivity metrics on a posterior probability scale. Moreover, the normalization function maps the original connectivity metrics to values between zero and one, which is well-suited for the graph theory based network analysis and avoids the information loss due to the (negative value) hard thresholding step. We apply the normalization method to a simulation study and the simulation results show that our normalization method effectively improves the robustness and reliability of the quantification of brain functional connectivity and provides more powerful group difference (biomarkers) detection. We illustrate our method on an analysis of a rs-fMRI dataset from the Autism Brain Imaging Data Exchange (ABIDE) study.

No MeSH data available.


Related in: MedlinePlus

(A,B) Heatmaps of the simulated correlations of the control and case groups; (C,D) Histograms of the simulated correlations of the control and case groups.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4584951&req=5

Figure 2: (A,B) Heatmaps of the simulated correlations of the control and case groups; (C,D) Histograms of the simulated correlations of the control and case groups.

Mentions: The beta distribution is employed to simulate correlation coefficients because it is more flexible and better resembles the real distribution of correlation coefficients from rs-fMRI data than other distributions (e.g., Gaussian distribution) (Ji et al., 2005; Jantschi and Sorana, 2011). We generate z1 from the non- distribution by a transformed Beta distribution: x1 ~ Beta (α1 = 3, β1 = 3) and z1 = 1.55x1 − 0.55 for correlation coefficients with higher connectivity expression levels; and z0 from the distribution by x0 ~ Beta (α0 = 18, β0 = 18) and then z0 = x0 ∗ 2 − 1. z1 represent 435 highly expressed correlation coefficients between the first 30 nodes for each subject in the control group, and z0 represents the rest of correlations for subjects in control group and all correlations for subjects in the case group. In this way, all simulated correlations range from [−1, 1], and Figure 2 demonstrates the simulated data for case and control group. Additionally, we use different set of parameters to represent various patterns of correlation distribution (e.g., Murphy et al., 2009) including: (i) more dispersed component x0 ~ Beta (α0 = 9, β0 = 9) (1); (ii) right skewed connected component x1 ~ Beta (α1 = 2, β1 = 3) (2); (iii) left skewed connected component x1 ~ Beta (α1 = 3, β1 = 2) (3).


An empirical Bayes normalization method for connectivity metrics in resting state fMRI.

Chen S, Kang J, Wang G - Front Neurosci (2015)

(A,B) Heatmaps of the simulated correlations of the control and case groups; (C,D) Histograms of the simulated correlations of the control and case groups.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4584951&req=5

Figure 2: (A,B) Heatmaps of the simulated correlations of the control and case groups; (C,D) Histograms of the simulated correlations of the control and case groups.
Mentions: The beta distribution is employed to simulate correlation coefficients because it is more flexible and better resembles the real distribution of correlation coefficients from rs-fMRI data than other distributions (e.g., Gaussian distribution) (Ji et al., 2005; Jantschi and Sorana, 2011). We generate z1 from the non- distribution by a transformed Beta distribution: x1 ~ Beta (α1 = 3, β1 = 3) and z1 = 1.55x1 − 0.55 for correlation coefficients with higher connectivity expression levels; and z0 from the distribution by x0 ~ Beta (α0 = 18, β0 = 18) and then z0 = x0 ∗ 2 − 1. z1 represent 435 highly expressed correlation coefficients between the first 30 nodes for each subject in the control group, and z0 represents the rest of correlations for subjects in control group and all correlations for subjects in the case group. In this way, all simulated correlations range from [−1, 1], and Figure 2 demonstrates the simulated data for case and control group. Additionally, we use different set of parameters to represent various patterns of correlation distribution (e.g., Murphy et al., 2009) including: (i) more dispersed component x0 ~ Beta (α0 = 9, β0 = 9) (1); (ii) right skewed connected component x1 ~ Beta (α1 = 2, β1 = 3) (2); (iii) left skewed connected component x1 ~ Beta (α1 = 3, β1 = 2) (3).

Bottom Line: Moreover, the normalization function maps the original connectivity metrics to values between zero and one, which is well-suited for the graph theory based network analysis and avoids the information loss due to the (negative value) hard thresholding step.We apply the normalization method to a simulation study and the simulation results show that our normalization method effectively improves the robustness and reliability of the quantification of brain functional connectivity and provides more powerful group difference (biomarkers) detection.We illustrate our method on an analysis of a rs-fMRI dataset from the Autism Brain Imaging Data Exchange (ABIDE) study.

View Article: PubMed Central - PubMed

Affiliation: Department of Epidemiology and Biostatistics, University of Maryland College Park, MD, USA.

ABSTRACT
Functional connectivity analysis using resting-state functional magnetic resonance imaging (rs-fMRI) has emerged as a powerful technique for investigating functional brain networks. The functional connectivity is often quantified by statistical metrics (e.g., Pearson correlation coefficient), which may be affected by many image acquisition and preprocessing steps such as the head motion correction and the global signal regression. The appropriate quantification of the connectivity metrics is essential for meaningful and reproducible scientific findings. We propose a novel empirical Bayes method to normalize the functional brain connectivity metrics on a posterior probability scale. Moreover, the normalization function maps the original connectivity metrics to values between zero and one, which is well-suited for the graph theory based network analysis and avoids the information loss due to the (negative value) hard thresholding step. We apply the normalization method to a simulation study and the simulation results show that our normalization method effectively improves the robustness and reliability of the quantification of brain functional connectivity and provides more powerful group difference (biomarkers) detection. We illustrate our method on an analysis of a rs-fMRI dataset from the Autism Brain Imaging Data Exchange (ABIDE) study.

No MeSH data available.


Related in: MedlinePlus