NeuroGAN-3D: Enhancing Intrinsic Functional Brain Networks via High-Fidelity 3D Generative Super-Resolution

2026-05-08

AI Summary

Low spatial resolution limits the precise localization and reliable parcellation of functional brain networks, as well as the detection of subtle neural changes. To address this challenge, this work proposes a generative super-resolution model specifically designed for three-dimensional resting-state fMRI data, introducing—for the first time—the application of 3D generative adversarial networks (GANs) to enhance the spatial resolution of functional brain maps. The method achieves high-fidelity detail enhancement at the voxel level, significantly outperforming conventional super-resolution baselines. While preserving the intrinsic functional topology, it effectively recovers fine-grained patterns of neural activity, thereby offering a novel and powerful tool for high-resolution functional brain network research.

Abstract

Recent advances in neuroimaging have deepened our understanding of the brain's complex functional and structural organization. Among these, functional Magnetic Resonance Imaging (fMRI) - particularly resting-state fMRI (rs-fMRI) - has emerged as a tool for identifying biomarkers of intrinsic brain connectivity and delineating large-scale neural networks. These networks are typically represented as volumetric spatial maps that capture functionally coherent brain regions and reflect individual differences in brain activity and structure. The spatial resolution of these maps plays an important role, as it determines the ability to localize functional units with precision, perform reliable brain parcellation, and detect subtle, spatially specific neurobiological alterations associated with development, aging, or disease. Therefore, improving the effective resolution of neuroimaging-derived maps holds significant promise for enabling more detailed insights into brain architecture and its relationship to behavior and pathology. To address this need, we propose NeuroGAN-3D, a novel 3D generative super-resolution model tailored to the computational demands of volumetric neuroimaging. Our model leverages a generative adversarial network architecture to enhance the spatial resolution of rs-fMRI spatial maps, significantly outperforming a conventional baseline.

Problem

• super-resolution
• resting-state fMRI
• functional brain networks
• spatial resolution
• neuroimaging

Innovation

• NeuroGAN-3D
• 3D generative super-resolution
• resting-state fMRI
• functional brain networks
• generative adversarial network