Bayesian Models of Functional Connectomics and Behavior

TL;DR

This paper introduces Bayesian models for joint analysis of functional connectomics and behavior, addressing challenges of limited data in clinical rs-fMRI studies through Bayesian representation learning and prediction.

Contribution

It proposes a fully Bayesian framework for joint representation learning and behavior prediction from connectomics data, improving reliability with small sample sizes.

Findings

Preliminary results on Autism Spectrum Disorder data show promise.

Bayesian approaches outperform classical methods in limited data scenarios.

Demonstrates the feasibility of Bayesian models for clinical connectomics analysis.

Abstract

The problem of jointly analysing functional connectomics and behavioral data is extremely challenging owing to the complex interactions between the two domains. In addition, clinical rs-fMRI studies often have to contend with limited samples, especially in the case of rare disorders. This data-starved regimen can severely restrict the reliability of classical machine learning or deep learning designed to predict behavior from connectivity data. In this work, we approach this problem from the lens of representation learning and bayesian modeling. To model the distributional characteristics of the domains, we first examine the ability of approaches such as Bayesian Linear Regression, Stochastic Search Variable Selection after performing a classical covariance decomposition. Finally, we present a fully bayesian formulation for joint representation learning and prediction. We present preliminary results on a subset of a publicly available clinical rs-fMRI study on patients with Autism Spectrum Disorder.