DCA: Graph-Guided Deep Embedding Clustering for Brain Atlases

2025-09-01

AI Summary

Existing brain atlases are predominantly predefined, population-level templates, lacking individual specificity and voxel-wise resolution. To address this, we propose Deep Cluster Atlas (DCA), the first framework to integrate graph-guided deep embedding clustering into brain atlas construction. DCA combines a pretrained autoencoder with a spatially regularized clustering loss to generate functionally homogeneous and spatially contiguous voxel-level parcellations directly from individual fMRI data. It enables flexible adjustment of parcellation granularity and anatomical coverage, and introduces a standardized evaluation platform. Across multiple datasets, DCA achieves a 98.8% improvement in functional homogeneity and a 29% increase in silhouette coefficient over state-of-the-art atlases. Moreover, it significantly enhances performance in downstream tasks—including autism classification and cognitive decoding—demonstrating superior generalizability and utility for precision neuroscience.

Technology Category

• Machine Learning: Clustering
• Humans and AI: Brain-Sensing and Analysis
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Brain atlases are essential for reducing the dimensionality of neuroimaging data and enabling interpretable analysis. However, most existing atlases are predefined, group-level templates with limited flexibility and resolution. We present Deep Cluster Atlas (DCA), a graph-guided deep embedding clustering framework for generating individualized, voxel-wise brain parcellations. DCA combines a pretrained autoencoder with spatially regularized deep clustering to produce functionally coherent and spatially contiguous regions. Our method supports flexible control over resolution and anatomical scope, and generalizes to arbitrary brain structures. We further introduce a standardized benchmarking platform for atlas evaluation, using multiple large-scale fMRI datasets. Across multiple datasets and scales, DCA outperforms state-of-the-art atlases, improving functional homogeneity by 98.8% and silhouette coefficient by 29%, and achieves superior performance in downstream tasks such as autism diagnosis and cognitive decoding. Codes and models will be released soon.

Problem

• Generating individualized brain atlases with flexible resolution control
• Overcoming limited flexibility in predefined group-level brain templates
• Ensuring functional coherence and spatial contiguity in brain parcellations

Innovation

• Graph-guided deep embedding clustering framework
• Combines autoencoder with spatially regularized clustering
• Produces individualized voxel-wise brain parcellations