arXiv subjects

Jin Ming

Publications and source records attributed to Jin Ming.

3 recordsLinked to original sources

arXiv2022

Flexible Bayesian Support Vector Machines for Brain Network-based Classification

Objective: Brain networks have gained increasing recognition as potential biomarkers in mental health studies, but there are limited approaches that can leverage complex brain networks for accurate classification. Our goal is to develop a novel Bayesian Support Vector Machine (SVM) approach that incorporates high-dimensional networks as covariates and is able to overcome limitations of existing penalized methods. Methods: We develop a novel Dirichlet process mixture of double exponential priors on the coefficients in the Bayesian SVM model that is able to perform feature selection and uncertainty quantification, by pooling information across edges to determine differential sparsity levels in an unsupervised manner. We develop different versions of the model that incorporates static and dynamic connectivity features, as well as an integrative analysis that jointly includes features from multiple scanning sessions. We perform classification of intelligence levels using resting state fMRI data from the Human Connectome Project (HCP), and a second Attention Deficiency Hyperactivity Disorder (ADHD) classification task. Results: Our results clearly reveal the considerable greater classification accuracy under the proposed approach over state-of-the-art methods. The multi-session analysis results in the highest classification accuracy in the HCP data analysis. Conclusion: We provide concrete evidence that the novel Bayesian SVMs provides an unsupervised and automated approach for network-based classification, that results in considerable improvements over penalized methods and parametric Bayesian approaches. Significance: Our work is one of the first to conclusively demonstrate the advantages of a Bayesian SVM in network-based classification of mental health outcomes, and the importance of multi-session network analysis.

stat.ME↗

arXiv2021

Integrative Learning for Population of Dynamic Networks with Covariates

Although there is a rapidly growing literature on dynamic connectivity methods, the primary focus has been on separate network estimation for each individual, which fails to leverage common patterns of information. We propose novel graph-theoretic approaches for estimating a population of dynamic networks that are able to borrow information across multiple heterogeneous samples in an unsupervised manner and guided by covariate information. Specifically, we develop a Bayesian product mixture model that imposes independent mixture priors at each time scan and uses covariates to model the mixture weights, which results in time-varying clusters of samples designed to pool information. The computation is carried out using an efficient Expectation-Maximization algorithm. Extensive simulation studies illustrate sharp gains in recovering the true dynamic network over existing dynamic connectivity methods. An analysis of fMRI block task data with behavioral interventions reveal sub-groups of individuals having similar dynamic connectivity, and identifies intervention-related dynamic network changes that are concentrated in biologically interpretable brain regions. In contrast, existing dynamic connectivity approaches are able to detect minimal or no changes in connectivity over time, which seems biologically unrealistic and highlights the challenges resulting from the inability to systematically borrow information across samples.

stat.ME↗

arXiv2019

Dynamic Brain Functional Networks Guided By Anatomical Knowledge

Recently, the potential of dynamic brain networks as a neuroimaging biomarkers for mental illnesses is being increasingly recognized. However, there are several unmet challenges in developing such biomarkers, including the need for methods to model rapidly changing network states. In one of the first such efforts, we develop a novel approach for computing dynamic brain functional connectivity (FC), that is guided by brain structural connectivity (SC) computed from diffusion tensor imaging (DTI) data. The proposed approach involving dynamic Gaussian graphical models decomposes the time course into non-overlapping state phases determined by change points, each having a distinct network. We develop an optimization algorithm to implement the method such that the estimation of both the change points and the state-phase specific networks are fully data driven and unsupervised, and guided by SC information. The approach is scalable to large dimensions and extensive simulations illustrate its clear advantages over existing methods in terms of network estimation accuracy and detecting dynamic network changes. An application of the method to a posttraumatic stress disorder (PTSD) study reveals important dynamic resting state connections in regions of the brain previously implicated in PTSD. We also illustrate that the dynamic networks computed under the proposed method are able to better predict psychological resilience among trauma exposed individuals compared to existing dynamic and stationary connectivity approaches, which highlights its potential as a neuroimaging biomarker.

q-bio.NC↗