Quantitative Biology > Neurons and Cognition
[Submitted on 21 Feb 2023]
Title:Establishing group-level brain structural connectivity incorporating anatomical knowledge under latent space modeling
View PDFAbstract:Brain structural connectivity, capturing the white matter fiber tracts among brain regions inferred by diffusion MRI (dMRI), provides a unique characterization of brain anatomical organization. One fundamental question to address with structural connectivity is how to properly summarize and perform statistical inference for a group-level connectivity architecture, for instance, under different sex groups, or disease cohorts. Existing analyses commonly summarize group-level brain connectivity by a simple entry-wise sample mean or median across individual brain connectivity matrices. However, such a heuristic approach fully ignores the associations among structural connections and the topological properties of brain networks. In this project, we propose a latent space-based generative network model to estimate group-level brain connectivity. We name our method the attributes-informed brain connectivity (ABC) model, which compared with existing group-level connectivity estimations, (1) offers an interpretable latent space representation of the group-level connectivity, (2) incorporates the anatomical knowledge of nodes and tests its co-varying relationship with connectivity and (3) quantifies the uncertainty and evaluates the likelihood of the estimated group-level effects against chance. We devise a novel Bayesian MCMC algorithm to estimate the model. By applying the ABC model to study brain structural connectivity stratified by sex among Alzheimer's Disease (AD) subjects and healthy controls incorporating the anatomical attributes (volume, thickness and area) on nodes, our method shows superior predictive power on out-of-sample structural connectivity and identifies meaningful sex-specific network neuromarkers for AD.
Submission history
From: Selena (Shuo) Wang [view email][v1] Tue, 21 Feb 2023 21:20:56 UTC (17,451 KB)
Current browse context:
stat.ME
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.