NV-Generate-MR-Brain

3D Brain MRI Synthesis

Commercial

NV-Generate-MR-Brain is a 3D latent diffusion model for synthetic brain MRI in contrasts including T1, T2, FLAIR, and SWI, selected with an integer label. Volumes reach 512×512×256 at 0.45×0.45×0.7 mm, finer than full-body MR. It is intended for synthetic training data and augmentation; the model card lists cross-modality synthesis (for example T1 to FLAIR) as coming soon.

MRI HuggingFace GitHub Paper Demo

Agent skills

Run it with an agent

Install the skill, then just ask your agent in plain language. medagent.nv_generate_mr_brain

install skill

npx skills add NVIDIA-Medtech/medical-AI-skills --skill nv-generate-mr-brain

Try saying

“Prepare a reproducible NV-Generate MR-Brain v1 command for a synthetic T1 brain MRI request at /data/brain_request.json. This is a command-shape review only; do not install packages, download weights, or start GPU inference in this environment.”

“Can this skill synthesize a skull-stripped FLAIR brain MRI?”

“Prepare an NV-Generate MR-Brain v1 command for an axial MRA request at /data/mra_request.json.”

“Produce a FLAIR brain MRI sample using nv-generate-mr-brain and report the output geometry.”

Skill on GitHub → Validation: Benchmark·Evals·Tests·Validators How agent skills work

Finetune it with an agent

Install the skill, then just ask your agent in plain language. medagent.nv_generate_mr_brain_finetune

install skill

npx skills add NVIDIA-Medtech/medical-AI-skills --skill nv-generate-mr-brain-finetune

Try saying

“Prepare a reproducible NV-Generate MR-Brain v1 finetune command for /data/mrbrain/datalist.json, with relative training images rooted at /data/mrbrain and outputs under runs/nv_generate_mr_brain_finetune. This is a command-shape review only; do not install packages, download weights, or start GPU training in this environment.”

“Prepare the MR-Brain v1 preflight command for /data/mrbrain/datalist.json, with relative training images rooted at /data/mrbrain and outputs under runs/nv_generate_mr_brain_finetune_preflight. This is a command-shape review only; do not install packages or start GPU training in this environment.”

“Validate my MONAI-style datalist and stage the configs for an MR-brain finetune, but don't launch training yet.”

Skill on GitHub → Validation: Benchmark·Evals·Tests·Validators How agent skills work

Engineering & research use only — not a medical device, not for diagnosis or clinical decision-making. Skills follow the open Agent Skills spec.