Skip to main content

Dynamic Region Grow

The Dynamic Region Grow tool segments the connected regions in the active volume object based on either similar gray value statistics or a specific range of gray values. You can choose between two region growing methods: Connectivity-based (analyzing statistical properties around seed points) or Threshold-based (segmenting within a specified intensity range).

Accessing the Tool​

  1. Navigate to the Segmentation tab in the ribbon.
  2. Click Dynamic Region Grow in the Create Mask section.

Parameters​

Region Growing Method​

Choose the region growing method from the dropdown:

Connectivity​

Identifies a connected set of pixels with pixel intensities matching the statistical properties of a seed point.

Parameters:

  • Multiplier: 1.0 to 100.0 (default: 1.0)
    • The multiplier to define the confidence interval for intensity similarity.
  • Number of iterations: Minimum 0 (default: 1)
    • The number of times the filter will run.
  • Initial neighborhood radius (pixels): Minimum 1 (default: 1)
    • The size of the initial neighborhood around the seed.

Threshold​

Identifies pixels connected to a seed point with grayscale values within a specified range.

Parameters:

  • Threshold adjustment: Use the histogram to set minimum and maximum threshold values.
  • Connectivity mode:
    • 6-connectivity: Only the neighboring faces of the selected voxel are considered (default).
    • 26-connectivity: The neighboring faces, nodes, and vertices of the selected voxel are considered.

Options​

  • Fill cavities: If checked, internal gaps in the mask will be filled. If unchecked, no gap filling will be applied.
  • Multiple layer: If checked, the operation will be applied to all slices of the dataset. If unchecked, it will only be applied to the current slice (default: checked).

Result​

  • Target mask: Select the mask object where the segmentation result will be stored.
  • Replace: Replace the selected mask with the segmentation calculated by this operation (default).
  • Merge: Merge the segmentation calculated by this operation into the selected mask object.

Workflow​

  1. Load your volume data and ensure it is the active volume object.
  2. Open the Dynamic Region Grow tool from the Segmentation tab.
  3. Select the Region growing method from the dropdown:
    • Choose Connectivity for intensity-based statistical region growing
    • Choose Threshold for range-based region growing
  4. Configure Method Parameters:
    • For Connectivity method: Adjust Multiplier (default: 1.0), Number of iterations (default: 1), and Initial neighborhood radius (default: 1 pixel).
    • For Threshold method: Use the histogram to set minimum and maximum threshold values, and choose 6-connectivity or 26-connectivity.
  5. Set Options:
    • Enable Fill cavities if you want internal gaps filled.
    • Enable/disable Multiple layer based on whether you want to segment all slices (default: checked).
  6. Configure Result settings:
    • Select the Target mask where results will be stored.
    • Choose Replace or Merge mode.
  7. Place seed point: Left-click on the object of interest in either the 2D or 3D view.
    • Use Ctrl+drag in the 3D view to rotate while selecting.
  8. The region grows automatically from the seed point based on the selected method.
  9. Review the result and adjust parameters if needed, then place new seed points to refine.

How It Works​

Dynamic region growing combines spatial connectivity with intensity analysis:

For Connectivity method:

  1. The algorithm starts at the seed point and records intensity statistics.
  2. Neighboring voxels are examined based on the multiplier and initial neighborhood.
  3. Voxels with intensities matching the statistical properties are added to the region.
  4. The process iterates the specified number of times, growing outward.
  5. The result is saved to the target mask.

For Threshold method:

  1. The algorithm starts at the seed point.
  2. Neighboring voxels are examined based on the connectivity setting.
  3. If a neighbor's intensity falls within the threshold range, it is added to the region.
  4. The process continues iteratively until no more qualifying voxels are found.
  5. The result is saved to the target mask.
Threshold: Include voxel v if: connected to region AND Tmin≤I(v)≤Tmax\text{Threshold: Include voxel } v \text{ if: connected to region AND } T_{min} \leq I(v) \leq T_{max}

Use Cases​

  • Organ Segmentation: Segment organs with relatively uniform intensity.
  • Lesion Detection: Isolate lesions that differ in intensity from surrounding tissue.
  • Vessel Extraction: Trace connected vascular structures.

Tips​

  • Connectivity Method: Best for regions with similar statistical properties; adjust the Multiplier to control sensitivity to intensity variations.
  • Threshold Method: Best when you know the specific intensity range of your target region.
  • Fill Cavities: Enable this to automatically fill holes in the segmented region.
  • Multiple Layer: Enable for 3D segmentation across all slices; disable for slice-by-slice segmentation.
  • Place the seed point in a representative area of the target structure.
  • If the region leaks into adjacent structures (Threshold method), narrow the threshold range.
  • If the region stops growing too early (Threshold method), widen the threshold range.

See Also​