Quantification for Mask or Model Objects¶

Overview¶

This dialog provides access to quantify properties for shapes and intensities of model and mask 2D or 3D objects.

Workflow in a nutshell¶

• Select material or mask in the Material section
• Define 2D or 3D shape of the objects to detect in the Object shape section
• Choose to detect object shapes or intensities in the Detection type section
• Select the property to quantify (Area)

• Hit the Run button to start quantification

Material¶

Use the combo boxes to select the object source and the dataset extent:

• Object source (Material): choose Mask, Exterior, or any model material. For models with more than 255 materials a Model option is also listed, quantifying all materials at once.
• Dataset extent (DatasetType): 2D, Slice (current slice), 3D, Stack (current Z-stack), or 4D, Dataset (whole volume).

Object shape¶

Select the dimensionality of the objects to detect:

• Shape2D — detects 2D objects on each slice independently.
• Shape3D — detects 3D connected objects across the stack.

Detection type¶

Choose what to measure:

• Object — statistics based on shape properties of the objects (see 3D properties and 2D properties below).
• Intensity — statistics based on image intensities inside the objects (see intensity properties below).

Properties¶

• Property: select the property to measure (see sections below for the full list).

• Connectivity: defines how pixels/voxels are grouped into objects:
  • 4/6 connectivity — combines pixels touching at North, West, South, East sides (2D) and also Top, Bottom (3D).
  • 8/26 connectivity — same as 4/6, plus diagonal (45°) neighbours.
• ColorChannel1: color channel for intensity analysis.
• ColorChannel2: second color channel for correlation analysis.
• Units: pixels or physical units.

• Multiple: enables measuring several properties in one run; click Define properties to choose the set.

• See below for details of quantification properties (2D shapes, 3D shapes, intensities)
• Quantify: starts quantification and populates the table.

Quantification table¶

The Quantification table displays the calculated values. Sort results with Sorting, or a row for the context menu.

• New selection / Add to selection / Remove from selection: create or modify the Selection layer from the highlighted rows.
• Copy column(s) to clipboard: copies the selected column(s) to the system clipboard.

• New annotations: places MIB annotations at the centroids of selected objects. A settings dialog lets you compose the label from a custom text string, the material name, and/or the object ID. Accessible afterwards via Ribbon → Model → Annotations → List of annotations or Segmentation panel → Segmentation tools → Annotations.
• Add to annotations: appends selected objects to the existing annotation list (same settings dialog).
• Remove from annotations: removes annotations at the selected object centroids.
• Calculate Mean / Sum / Min / Max: computes the statistic for column 2 of the highlighted rows and copies the result to the clipboard.
• Crop to a file/MATLAB: opens the Crop Objects dialog to export image patches cut around each detected object.

• Objects to a new model: generates a new model assigning each selected object a unique material index. Demo

• Plot histogram: opens a MATLAB figure with a bar histogram of the selected values (prompts for number of bins).

• Export: exports the full results table.

Histogram and object selection¶

The Histogram shows the distribution of values in the table, with linear or logarithmic Y-axis (via Log scale).

Use / on the histogram to set the minimum / maximum of the Highlight range.

• Auto highlight on a click: instantly highlights the clicked objects in the Image View panel. Ctrl + deselects an object.
• Sorting: changes table sort order.
• Highlight range + Do: select all objects whose value falls within the specified min–max range.
• The Details panel: defines the selection action — add, remove, or replace the Selection layer.

Quantification properties of 3D objects¶

• Volume: total voxels (~volume) within 3D objects.
• ConvexVolume: total voxels of the smallest convex hull enclosing the 3D object.
• EndpointsLength: distance between endpoints of 1-pixel-wide lines (use Membrane ClickTracker with Straight line, Width=1). Calculated in image units.
• EquatorialEccentricity: eccentricity of the cross-section through the second-longest and shortest axes (0 = sphere, 1 = line).
• EquivDiameter: diameter of a sphere with the same volume as the object.
• Extent: ratio of object voxels to bounding-box voxels.
• FilledArea: total voxels within the filled (hole-free) 3D object.
• HolesArea: total voxels of internal cavities within 3D objects.
• MajorAxisLength: length of the major principal axis in pixels (see the note below the list).
• MeridionalEccentricity: eccentricity of the cross-section through the longest and shortest axes.
• SecondAxisLength: length of the second principal axis in pixels (see the note below the list).
• Solidity: proportion of convex-hull voxels occupied by the object (Volume / ConvexVolume).
• SurfaceArea: surface area of the 3D object (requires isotropic voxels for physical units).
• ThirdAxisLength: length of the minor (third) principal axis in pixels (see the note below the list).

Quantification properties of 2D objects¶

• Area: total pixels within 2D objects.
• ConvexArea: total pixels of the smallest convex polygon containing the object.
• CurveLength: length of thinned curves (closed or open) in image units. Requires thinning (Ribbon → Selection → Morphological 2D/3D operations) and branch-point removal.
• Eccentricity: ratio of foci distance to major-axis length of the equivalent ellipse (0 = circle, 1 = line).
• EndpointsLength: distance between endpoints of 1-pixel-wide lines (use Membrane ClickTracker with Straight line, Width=1; thin with Brush size 1 or Ribbon → Selection → Morphological 2D/3D operations → Thin). Calculated in image units.
• EquivDiameter: diameter of a circle with the same area as the object (sqrt(4*Area/pi)).
• EulerNumber: number of objects minus number of holes (e.g., 1 hole → 0, 2 holes → -1).
• Extent: ratio of object pixels to bounding-box pixels (Area / bounding-box area).
• FilledArea: total pixels of the filled (hole-free) object.
• FirstAxisLength: actual length of the major 2D axis, measured on the object (see the note below the list).
• HolesArea: total pixels within holes.
• MajorAxisLength: length of the major axis of the equivalent-second-moment ellipse (pixels) (see the note below the list).
• MinorAxisLength: length of the minor axis of the equivalent-second-moment ellipse (pixels) (see the note below the list).
• Orientation: angle (−90° to +90°) between the x-axis and the major axis of the equivalent ellipse.
• Perimeter: boundary length (sum of distances between adjacent border pixels).
• SecondAxisLength: actual length of the minor 2D axis, measured on the object (see the note below the list).
• Solidity: proportion of convex-hull pixels occupied by the object (Area / ConvexArea).

Quantification properties of intensities of 2D/3D objects¶

Calculated when Intensity is selected in the Detection type section.

• MinIntensity: minimal image intensity within 2D/3D objects.
• MaxIntensity: maximal image intensity within 2D/3D objects.
• MeanIntensity: mean image intensity within 2D/3D objects.
• StdIntensity: standard deviation of image intensity within 2D/3D objects.
• SumIntensity: sum of image intensities within 2D/3D objects.
• Correlation: correlation between intensities of two selected color channels (see MATLAB's corr2).

Back to MIB | User interface | Ribbon | Mask (Model)