All CBIS software and introduction can be found at our Github page and Youtube channel.

singR: An R package for Simultaneous non-Gaussian Component Analysis for data integration

                                       by Liangkang Wang, Irina Gaynanova, and Benjamin Risk 

This R package implements Simultaneous non-Gaussian Component Analysis for data integration. SING uses a non-Gaussian measure of information to extract feature loadings and scores (latent variables) that are shared across multiple datasets. We describe and implement functions through two examples. The first example is a toy example working with images. The second example is a simulated study integrating functional connectivity estimates from a resting- state functional magnetic resonance imaging dataset and task activation maps from a working memory functional magnetic resonance imaging dataset. The SING model can produce joint components that accurately reflect information shared by multiple datasets, particularly for datasets with non-Gaussian features such as neuroimaging.

CRAN website:  https://cran.r-project.org/web/packages/singR/index.html 

HINT: Hierarchical Independent Component Analysis Toolbox

Shi, R. and Guo, Y. (2017). Investigating differences in brain functional networks using hierarchical covariate-adjusted independent component analysis. Annals of Applied Statistics. 10(4): 1930-1957.An earlier version of the paper was selected for the First-Place winner of the 2015 Student Paper Competition, American Statistical Association (ASA) Statistics in Imaging Section.

This paper proposes the hc-ICA model that provides a formal statistical framework for estimating covariate effects and testing differences between brain functional networks. hc-ICA can provide a more reliable and powerful statistical tool for evaluating group differences in brain functional networks while appropriately controlling for potential confounding factors. HINT is a matlab toolbox implementing the hc-ICA model. To download the Matlab GUI-based hc-ICA Toolbox, please click here. For a video tutorial on how to use the toolbox, please click here. The manuscript for this toolbox is available here.

Difference Degree Test Toolbox

Higgins, IA, Kundu, S., Choi, KS., Mayberg, H., Guo, Y. (2019), A Differential Degree Test for Comparing Brain Networks. Human Brain Mapping, 1-19.

The Difference Degree Test (DDT) is a two stage method to detect regions incident to a statistically significant number of differentially weighted edges (DWEs). In the phase, we select a data-adaptive threshold to identify the DWEs followed by a statistical test for the number of DWEs incident to each brain region. The key to our procedure the Hirscheberger-Qi-Steuer (Hirschberger et al., 2007) algorithm, which is a computationally efficient algorithm for generating random null networks that replicate statistical properties of the observed difference network. To download the Toolbox, please click here.

Structurally informed Gaussian Graphical Model package (siGGM)

Higgins, I., Kundu, S. and Guo, Y. (2018). Integrative Bayesian analysis of brain functional networks incorporating anatomical knowledge. NeuroImage. Volume 181, Pages 263-278, ISSN 1053-8119.

The structurally informed Gaussian Graphical Model (siGGM), is a R package providing a method for integrating structural connectivity (SC) information into the estimation of functional connectivity (FC). The package currently supports functional connectivity estimation as described in Higgins et al. (2018). The R code can be download from the CBIS Github Page.

Connectivity Change Point Detection

Kundu, S., Ming, J., Pierce, J., McDowell, J., and Guo, Y. (2018). Estimating Dynamic Brain Functional Networks Using Multi-subject fMRI Data. NeuroImage. to appear

Connectivity Change Point Detection (CCPD) is a fully automated two-stage approach which pools information across multiple subjects to estimate change points in functional connectivity and subsequently estimates the brain networks within each state phase lying between consecutive change points. The number and positioning of the change points are unknown and learned from the data in the first stage by modeling a time-dependent connectivity metric under a fused lasso approach. In the second stage, the brain functional network for each state phase is inferred via sparse inverse covariance matrices. The Matlab code can be found on the CBIS Github Page.

DensParcorr: Dens-based Partial Correlation Estimation method

Wang Y, Kang J, Kemmer PB and Guo Y (2016) An Efficient and Reliable Statistical Method for Estimating Functional Connectivity in Large Scale Brain Networks Using Partial Correlation. Frontiers in Neuroscience, 10:123. doi: 10.3389/fnins.2016.00123.

This paper presents an efficient and reliable statistical method for estimating partial correlation matrices for investigating direct functional connectvity in large-scale brain networks. To download the R code, please click here.

Bayesian spatial model for activation and connectivity (BSMAC)

MATLAB toolbox that implements methods and graphics from Bowman, F. D., Caffo, B. A, Bassett, S., and Kilts, C., 2008, NeuroImage paper, and Zhang, L., Agravat, S. Derado, D., Chen, S. McIntosh, B.J. Bowman, F.D., 2011 Neuroscience Methods paper. To download the BSMAc Package, please click here