Vector Auto-Regressive Deep Neural Network: A Data-Driven Deep Learning-Based Directed Functional Connectivity Estimation Toolbox

Abstract

An important goal in neuroscience is to elucidate the causal relationships between the brain’s different regions. This can help reveal the brain’s functional circuitry and diagnose lesions. Currently there are a lack of approaches to functional connectome estimation that leverage the state-of-the-art in deep learning architectures and training methodologies. Therefore, we propose a new framework based on a vector auto-regressive deep neural network (VARDNN) architecture. Our approach consists of a set of nodes, each with a deep neural network structure. These nodes can be mapped to any spatial sub-division based on the data to be analyzed, such as anatomical brain regions from which representative neural signals can be obtained. VARDNN learns to reproduce experimental time series data using modern deep learning training techniques. Based on this, we developed two novel directed functional connectivity (dFC) measures, namely VARDNN-DI and VARDNN-GC. We evaluated our measures against a number of existing functional connectome estimation measures, such as partial correlation and multivariate Granger causality combined with large dimensionality counter-measure techniques. Our measures outperformed them across various types of ground truth data, especially as the number of nodes increased. We applied VARDNN to fMRI data to compare the dFC between 41 healthy control vs. 32 Alzheimer’s disease subjects. Our VARDNN-DI measure detected lesioned regions consistent with previous studies and separated the two groups well in a subject-wise evaluation framework. Summarily, the VARDNN framework has powerful capabilities for whole brain dFC estimation. We have implemented VARDNN as an open-source toolbox that can be freely downloaded for researchers who wish to carry out functional connectome analysis on their own data.

A reappraisal of the default mode and frontoparietal networks in the common marmoset brain

“…Multiseed-based connectivity analysis of marmoset resting-state fMRI data Multiseed-based connectivity analysis was done by calculating the correlation coefficients between seed voxels and all other voxels. MATLAB scripts for this analysis were developed inhouse and worked together with the VARDNN toolbox (Okuno and Woodward, 2021). To investigate the functional connectivity between the frontal pole and PCC regions, the seed voxels of the marmoset mPFC and PCC regions were manually edited in ITK-SNAP (Yushkevich et al, 2006).…”

Major Problems in Clinical Psychological Science and How to Address them. Introducing a Multimodal Dynamical Network Approach

“…To analyze longitudinal networks, our approach aims to utilize various statistical methods, including autoregressive models (e.g., Bringmann et al, 2013;Epskamp, 2020), Group Iterative Multiple Model Estimation (GIMME; Gates & Molenaar, 2012), multivariate Granger causality (Granger, 1969), and Bayesian networks (e.g., McNally et al, 2017;Pearl & Mackenzie, 2019), and possibly advanced machine learning techniques such as deep neural networks (e.g., Goodfellow et al, 2016;Okuno & Woodward, 2021). These methods facilitate the identification of stable network connections (i.e., strength of connections) and contribute to improving treatment outcomes (McElroy et al, 2019).…”

A reappraisal of the default mode and frontoparietal networks in the common marmoset brain

“…Multiseed-based connectivity analysis was done by calculating the correlation coefficients between seed voxels and all other voxels. MATLAB scripts for this analysis were developed in-house and worked together with the VARDNN toolbox [54]. The seed voxels of the marmoset mPFC and PCC regions were manually edited in ITK-SNAP [55].…”