skills.rest
NVIDIA
NVIDIAOFFICIAL
Explore all skills by this author →
WebsiteGitHub

i4h-catheter-navigation-smoke

Runs CPU-only fluorosim smoke tests for catheter navigation workflows in CI environments.

3.2k|370|Updated Feb 25, 2026
Search Tags:#ci#smoke test#unittest#cpu-only#catheter navigation#isaac for healthcare#fluorosim
One-click install
npx skills add https://github.com/NVIDIA/skills --skill i4h-catheter-navigation-smoke
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: i4h-catheter-navigation-smoke
Source: https://github.com/NVIDIA/skills/tree/main/skills/i4h-catheter-navigation-smoke
Command: npx skills add https://github.com/NVIDIA/skills --skill i4h-catheter-navigation-smoke

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Validating the Isaac for Healthcare catheter navigation workflow normally requires GPU hardware, which slows down CI pipelines and blocks early sanity checks. This Skill runs CPU-only smoke tests that verify imports, CT-to-mu preprocessing, and CLI parsers without any GPU.

Core Features & Use Cases

  • CPU-Only Smoke Testing: Executes the fluorosim unittest suite covering module imports, CT-to-mu preprocessing, synthetic phantom building, and CLI parser sanity checks.
  • CI-Friendly Logging: Creates timestamped run directories with persistent smoke logs and a .latest symlink for easy artifact retrieval.
  • Result Verification: Greps logs for Ran N tests and OK markers, correctly treating expected parser stderr from negative CLI tests as non-failures.
  • Use Case: Before provisioning GPU resources, a developer asks the agent to smoke-test the catheter navigation workflow in CI; the agent clones or locates the i4h-workflows repo, runs the unittest module, and reports pass/fail with the log path.

Quick Start

Run the catheter navigation smoke tests in CI mode without a GPU and save a reusable log.

Frequently Asked Questions about i4h-catheter-navigation-smoke

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run catheter navigation smoke tests without a GPU?▼

Run the CPU-only fluorosim smoke suite with python3 -m unittest workflows/catheter_navigation/tests/test_fluorosim_smoke.py from the i4h-workflows root. It exercises imports, CT-to-mu preprocessing, and CLI parsers without requiring nvidia-smi, slangpy, or a display.

How to smoke-test Isaac for Healthcare workflows in CI?▼

Use the CPU-only smoke test suite, which is safe for CI runners without GPUs. It creates a timestamped run directory, captures output in smoke.log, and verifies success by checking for the Ran N tests and OK lines.

Why does the smoke test log show parser errors but still pass?▼

Parser error lines on stderr come from negative CLI test cases and are expected behavior, not failures. The run passes when the unittest output ends with a Ran N tests line followed by OK.

Does the catheter navigation smoke test validate GPU rendering?▼

No, the smoke suite is CPU-only and does not exercise GPU Slang rendering or the interactive viewport. For GPU validation and synthetic DRR rendering, use the separate i4h-catheter-navigation-render-drr workflow.

What are the prerequisites for running fluorosim smoke tests?▼

You need Python 3 with the catheter workflow dependencies installed, typically via workflows/catheter_navigation/requirements.txt. No GPU is required, and the i4h-workflows repository is cloned automatically if not present locally.

Related Skills

Explore More Software Engineering Skills→
edlovesjavaedlovesjava

python-venv

Create and activate Python venv environments and install dependencies from requirements.txt.

Community
Intermediate
castform-aicastform-ai
49

launch-run

Launch Castform training runs after validation, upload, and argument checks.

Official
Intermediate
vince-winkintelvince-winkintel
43

glab-ci

Manage GitLab CI/CD pipelines, jobs, and artifacts.

Community
Intermediate
Tyler-R-KendrickTyler-R-Kendrick
1

huggingface-community-evals

Executes Hugging Face Hub model evaluations locally using inspect-ai and light frameworks with vLLM, Transformers, or accelerate backends.

Community
Intermediate