🎉 Latest News: Our paper on "Vision-Language Transformers in 3D MRI Synthesis" has been accepted at MICCAI 2026!
Home > Projects > NeuroSynth-3D: Synthetic Brain MRI Gener...
NeuroSynth-3D: Synthetic Brain MRI Generation
3D Computer Vision Model Development
NIH Sponsored ($2.4M)

NeuroSynth-3D: Synthetic Brain MRI Generation

Lead Investigator: Prof. Michael Zhang

Executive Summary

Generative 3D diffusion models capable of producing privacy-preserving synthetic dataset extensions for rare neurodegenerative disease research.

Technology Stack & Domain Frameworks

Diffusion Models 3D Volumetric MRI Synthetic Data

Methodology & Clinical Validation

Project Overview: NeuroSynth-3D uses latent diffusion architectures tailored for 3D isotropic brain MRI volumes (T1w, T2w, FLAIR), eliminating data sharing bottlenecks in rare neurodegenerative disease studies. Clinical Impact: Provides high-fidelity, anatomically realistic brain scans with controlled ventricular enlargement and white matter hyperintensity lesions.

Related Research Projects