Parsing altered brain connectivity in neurodevelopmental disorders by integrating graph-based normative modeling and deep generative networks

Rui Sherry Shen,Yusuf Osmanlıoğlu,Drew Parker,Darien Aunapu,Benjamin E. Yerys,Birkan Tunç,Ragini Verma

2024-10-15

Abstract:Many neurodevelopmental disorders can be understood as divergent patterns of neural interactions during brain development. Advances in neuroimaging have illuminated these patterns by modeling the brain as a network structure using diffution MRI tractography. However, characterizing and quantifying individual heterogeneity in neurodevelopmental disorders within these highly complex brain networks remains a significant challenge. In this paper, we present for the first time, a framework that integrates deep generative models with graph-based normative modeling to characterize brain network development in the neurotypical population, which can then be used to quantify the individual-level neurodivergence associated with disorders. Our deep generative model incorporates bio-inspired wiring constraints to effectively capture the developmental trajectories of neurotypical brain networks. Neurodivergence is quantified by comparing individuals to this neurotypical trajectory, enabling the creation of region-wise divergence maps that reveal latent developmental differences at each brain regions, along with overall neurodivergence scores based on predicted brain age gaps. We demonstrate the clinical utility of this framework by applying it to a large sample of children with autism spectrum disorders, showing that the individualized region-wise maps help parse the heterogeneity in autism, and the neurodivergence scores correlate with clinical assessments. Together, we provide powerful tools for quantifying neurodevelopmental divergence in brain networks, paying the way for developing imaging markers that will support disorder stratification, monitor progression, and evaluate therapeutic effectiveness.

Neurons and Cognition,Artificial Intelligence,Quantitative Methods

What problem does this paper attempt to address?

The paper attempts to address the issue of heterogeneity analysis in brain connectivity patterns in neurodevelopmental disorders (such as Autism Spectrum Disorder). Specifically, the study proposes a new framework called BRIDGE (Brain Representation via Individualized Deep Generative Embedding), which for the first time combines deep generative models with graph-based normative modeling methods to characterize brain network development in typical populations and quantify neural differentiation at the individual level. The main objectives include: 1. **Characterizing brain network development in typical populations**: By integrating biologically inspired wiring constraints, BRIDGE can effectively capture the developmental trajectory of typical brain networks. 2. **Quantifying neural differentiation at the individual level**: By comparing individuals to the typical developmental trajectory, BRIDGE can create regional differentiation maps, revealing potential developmental differences in each brain region and providing an overall neural differentiation score. 3. **Clinical application**: By applying the framework to a large sample of children with autism, the study demonstrates its utility in clinical diagnosis, showing that individualized regional maps help to parse the heterogeneity of autism and that neural differentiation scores correlate with clinical assessments. These tools provide strong support for quantifying structural network differentiation in neurodevelopmental disorders and lay the foundation for further development of imaging markers that can be used for disease stratification, monitoring progress, and evaluating treatment efficacy.

Parsing altered brain connectivity in neurodevelopmental disorders by integrating graph-based normative modeling and deep generative networks

Aberrant Development of Cross-Frequency Multiplex Functional Connectome in First-Episode, Drug-Naive Major Depressive Disorder and Schizophrenia.

Age-atypical brain functional networks in autism spectrum disorder: a normative modeling approach

Mapping the Heterogeneous Brain Structural Phenotype of Autism Spectrum Disorder Using the Normative Model

Topological and Graph Theoretical Analysis of Dynamic Functional Connectivity for Autism Spectrum Disorder

Comprehensive exploration of multi-modal and multi-branch imaging markers for autism diagnosis and interpretation: insights from an advanced deep learning model

Network Dependency Index Stratified Subnetwork Analysis of Functional Connectomes: An application to autism

A Deep-Generative Hybrid Model to Integrate Multimodal and Dynamic Connectivity for Predicting Spectrum-Level Deficits in Autism

Delineating Transdiagnostic Subtypes in Neurodevelopmental Disorders via Contrastive Graph Machine Learning of Brain Connectivity Patterns

Connectome-scale Group-Wise Consistent Resting-State Network Analysis in Autism Spectrum Disorder.

Interpretable Graph Neural Networks for Connectome-Based Brain Disorder Analysis

Brain functional activity-based classification of autism spectrum disorder using an attention-based graph neural network combined with gene expression

Personalized functional network mapping for autism spectrum disorder and attention-deficit/hyperactivity disorder

Toward Developmental Connectomics of the Human Brain.

SSR-GPCsT: Deep Learning Models Based on Functional Connectivity Maps in Autism Research.

Capturing the individual deviations from normative models of brain structure for depressive diagnosis and treatment

Functional hypergraph uncovers novel covariant structures over neurodevelopment

Modeling the dynamic brain network representation for autism spectrum disorder diagnosis

Graph Theoretical Modeling of Baby Brain Networks

Understanding the heterogeneity of dynamic functional connectivity patterns in first-episode drug naive depression using normative models

Predicting Transdiagnostic Social Impairments in Childhood using Connectome-based Predictive Modeling