ConnectomeDiffuser: Generative AI Enables Brain Network Construction from Diffusion Tensor Imaging

TL;DR

This paper addresses the challenge of robustly constructing individualized brain networks from diffusion tensor imaging (DTI) for neurodegenerative and neurodevelopmental disorders. It introduces ConnectomeDiffuser, a three-component pipeline comprising a geometry-aware Template Network, a latent diffusion generator, and a Graph Convolutional Network classifier to jointly learn and utilize disease-relevant connectivity patterns. The approach yields higher classification accuracy and more faithful topology than state-of-the-art baselines (e.g., PANDA) on the ADNI and ABIDE datasets, with strong ablation evidence for the necessity of each component and efficient inference. The work demonstrates disease progression-related connectivity changes—predominant reductions in Alzheimer’s disease and a balanced reconfiguration in autism—highlighting implications for biomarker discovery, clinical diagnosis, and therapeutic monitoring.

Abstract

Brain network analysis plays a crucial role in diagnosing and monitoring neurodegenerative disorders such as Alzheimer's disease (AD). Existing approaches for constructing structural brain networks from diffusion tensor imaging (DTI) often rely on specialized toolkits that suffer from inherent limitations: operator subjectivity, labor-intensive workflows, and restricted capacity to capture complex topological features and disease-specific biomarkers. To overcome these challenges and advance computational neuroimaging instrumentation, ConnectomeDiffuser is proposed as a novel diffusion-based framework for automated end-to-end brain network construction from DTI. The proposed model combines three key components: (1) a Template Network that extracts topological features from 3D DTI scans using Riemannian geometric principles, (2) a diffusion model that generates comprehensive brain networks with enhanced topological fidelity, and (3) a Graph Convolutional Network classifier that incorporates disease-specific markers to improve diagnostic accuracy. ConnectomeDiffuser demonstrates superior performance by capturing a broader range of structural connectivity and pathology-related information, enabling more sensitive analysis of individual variations in brain networks. Experimental validation on datasets representing two distinct neurodegenerative conditions demonstrates significant performance improvements over other brain network methods. This work contributes to the advancement of instrumentation in the context of neurological disorders, providing clinicians and researchers with a robust, generalizable measurement framework that facilitates more accurate diagnosis, deeper mechanistic understanding, and improved therapeutic monitoring of neurodegenerative diseases such as AD.

Figures (7)

• Figure 1: The overall architecture of the proposed ConnectcomeDiffuser comprises several steps. Initially, the Template Network is applied to DTI images to extract topological features of the brain network. These features are then passed to the diffusion model, resulting in the construction of a comprehensive brain network. Subsequently, the GCN classifier is employed, incorporating classification knowledge to further enhance the feature extraction and diffusion model.
• Figure 2: The architecture of the Template Network involves processing a tensor format of a 3D DTI image and an AAL template as input to generate the brain feature matrix as output.
• Figure 3: Comparative analysis of classification performance across various metrics is conducted between the proposed model and PANDA, as well as the ablation study.
• Figure 4: T-SNE analysis between ConnectomeDiffuser and PANDA.
• Figure 5: Chord diagram of the connectivity difference in the ADNI dataset. The top row represents the increased connectivity, from left to right: EMCI to NC, LMCI to NC and AD to NC. The bottom row represents the decreased connectivity, from left to right: EMCI to NC, LMCI to NC and AD to NC.