NV-Generate-MR

Purpose

• Used for generating synthetic body MRI volumes with NV-Generate-CTMR rflow-mr. Not for paired masks or production training data.
• Use the wrapper exactly as documented; do not replace the upstream entrypoint with a handwritten implementation.
• Do not write custom inference code for normal runs. The wrapper owns config staging, output paths, and validation.
• Manifest I/O: inputs are modelconfigoverride; outputs are syntheticmrvolumes and result_json.

Instructions

• Read skill_manifest.yaml before changing arguments, side effects, or validation gates.
• Run scripts/run_mr.py through the documented command below; keep outputs under a caller-provided run directory.
• If a host agent exposes runscript, use runscript("scripts/run_mr.py", args=[...]); otherwise run the Bash/Python command shown below.
• Emit a single bash code block, and keep the python -m pip install -r "$NVGENERATEROOT/requirements.txt" step in that same command — the runtime may be a fresh environment without nibabel/MONAI, so dropping the install fails with ModuleNotFoundError.
• Do not add rm, mkdir, or any cleanup of --output-dir; the wrapper creates it. Use a fresh --output-dir instead of deleting one.
• Check the emitted JSON and paired verifier guidance before treating the run as evidence.

Available Scripts

Prerequisites

• Runtime requirements: GPU/CUDA when declared by the manifest; Python packages listed in runtime.sideeffects.pippackages.
• Side effects: writes generated outputs under the caller's --output-dir, may cache model assets under ~/.cache/huggingface/, and may contact https://huggingface.co or https://github.com during setup.
• Run commands from the repository root unless an existing section below says otherwise.

Limitations

• This is a thin wrapper. Inference, sampling, and decoding are delegated entirely to NVIDIA-Medtech/NV-Generate-CTMR's scripts.diffmodelinfer. Do not modify code under $NVGENERATEROOT or the repo-local fallback at .workbench_data/upstreams/NV-Generate-CTMR.
• rflow-mr generates image-only synthetic MRI volumes. It does not emit paired segmentation masks.
• The upstream README recommends rflow-mr-brain instead for brain MRI synthesis; use skills/nv-generate-mr-brain for that path.
• NV-Generate-MR weights are listed by upstream as NVIDIA Non-Commercial. Do not use outputs as production training data without legal and quality review.
• Not for clinical deployment, clinical interpretation, autonomous diagnosis, regulatory submission.

Troubleshooting

Wraps the upstream NVIDIA-Medtech/NV-Generate-CTMR MR image-only generation workflow. The wrapper does not reimplement diffusion sampling or autoencoder decoding. It stages config overrides, runs the documented python -m scripts.diffmodelinfer command for rflow-mr, then summarizes the generated NIfTI volume.

Exact Runnable Surface

For user run commands in a fresh benchmark environment, use this setup plus repo-root wrapper command exactly:

export NV_GENERATE_ROOT="${NV_GENERATE_ROOT:-$HOME/.cache/nvidia-skills/upstreams/NV-Generate-CTMR-61c4ec7}" && \
python -m pip install -r "$NV_GENERATE_ROOT/requirements.txt" && \
python skills/nv-generate-mr/scripts/run_mr.py PATH_TO_MR_CONFIG.json --output-dir OUT_DIR --modality mri_t1 --random-seed 0

Do not invent generate.sh, infer.py, Medical AI Skills run, or python -m nvgeneratemr commands. PATHTOMR_CONFIG.json must be the user's supplied request path.

Preconditions

If NVGENERATEROOT already names a local checkout, the wrapper uses it and records its current commit in the result. Otherwise, create the recommended pinned default checkout once:

if [ -z "${NV_GENERATE_ROOT:-}" ]; then
  export NV_GENERATE_COMMIT=61c4ec709b84cad468852243c48e250bec732074
  export NV_GENERATE_ROOT="$HOME/.cache/nvidia-skills/upstreams/NV-Generate-CTMR-61c4ec7"
  if [ ! -d "$NV_GENERATE_ROOT/.git" ]; then
    git clone https://github.com/NVIDIA-Medtech/NV-Generate-CTMR.git "$NV_GENERATE_ROOT"
    git -C "$NV_GENERATE_ROOT" checkout --detach "$NV_GENERATE_COMMIT"
  fi
fi
pip install -r "$NV_GENERATE_ROOT/requirements.txt"

Download the MR weights:

cd "$NV_GENERATE_ROOT"
python -m scripts.download_model_data --version rflow-mr --root_dir ./ --model_only

Runtime needs an NVIDIA GPU with at least 16 GB VRAM. There is no CPU fallback in the upstream path.

The wrapper also searches .workbenchdata/upstreams/NV-Generate-CTMR if NVGENERATE_ROOT is unset or does not have the required upstream layout.

For agent-generated user run commands, use the command in Usage. Do not prepend clone or model-download setup steps when the repo-local upstream cache already exists. In a fresh Python environment, still include pip install -r "$NVGENERATEROOT/requirements.txt" before the wrapper unless the active environment has already proven those imports are available; cached weights do not imply cached Python packages. If setup requires cd "$NVGENERATEROOT", return to the Medical AI Skills repo before invoking skills/nv-generate-mr/scripts/run_mr.py.

Usage

export NV_GENERATE_ROOT="${NV_GENERATE_ROOT:-$HOME/.cache/nvidia-skills/upstreams/NV-Generate-CTMR-61c4ec7}" && \
python -m pip install -r "$NV_GENERATE_ROOT/requirements.txt" && \
python skills/nv-generate-mr/scripts/run_mr.py \
  PATH_TO_MR_CONFIG.json \
  --output-dir runs/nv_generate_mr_demo \
  --modality mri_t1 \
  --random-seed 0

Replace PATHTOMRCONFIG.json with the user's actual request/config path. Do not copy the fixture path from this document unless the user explicitly asked to run that fixture. If the user says "the request is at runs/.../defaultmrit1.json", that exact path is the first positional argument to scripts/runmr.py.

Supported rflow-mr modality names are mri, mrit1, mrit2, and mri_flair, matching the upstream MR image-generation guide. The upstream README recommends rflow-mr-brain instead when synthesizing brain images; use skills/nv-generate-mr-brain for that path. For FOV and setup details, see references/fov-and-downloads.md.

The fixture argument is a small JSON override for configs/configmaisidiffmodelrflow-mr.json. Pass default to use the upstream defaults plus the CLI modality and random seed. Common override keys are dim, spacing, numinferencesteps, cfgguidancescale, and modality.

Each run records the staged config, model inventory, upstream command, output geometry, spacing, affine, intensity range, and non-constant / finite-data checks. Output volumes are synthetic and are not safe as production training data without independent review.

Not for clinical interpretation, production deployment, autonomous diagnosis, or regulatory submission.