Skip to main content
Version: Research-v26.0

AI

Automatically generate anatomical segments using AI models.

  • The AI tool allows users to automatically segment anatomical structures from CT images using deep learning models based on nnU-Net v2.
  • It supports multiple AI models to identify specific anatomical regions and create segments automatically.

Steps:​

  • Choose the AI tool from the Segment Editor effects panel.
  • Select the desired AI model (OMF AI or Knee AI).
  • Review the information about the selected model.
  • Click Apply to start AI-based segmentation.
  • Wait for the process to complete. The generated segments will be added automatically.

Additional Information​

Model Selection​

  • OMF AI performs segmentation of skull, mandible, teeth, and mandibular canal from CT images.
  • Knee AI performs segmentation of knee structures including femur, tibia, and fibula from CT images.
  • Hip AI is currently disabled and will be available in a future release.

First-Time Setup​

  • During the first use of AI segmentation, the system may download required AI components or models.

  • Possible downloads include:

    • AI components (~2GB)
    • AI model (~230MB)

A confirmation dialog will appear before the download starts.

Progress and Processing​

  • A single progress dialog is shown during the entire workflow.
  • The progress includes component installation, model download, and AI inference.
  • Users can cancel the process while it is running.

AI