Quantification

The Quantification tab extracts measurements from segmented images and exports them to spreadsheet files.

Interface overview

Main interface controls

• Add feature (button): Create a new quantification feature.
• Features list (panel): Shows all configured features with name, type, and delete button.
• Output folder (browse field): Select folder where results will be saved.
• Save name (text field): Name for the output subfolder (optional).
• Format (dropdown): Choose xlsx or csv for export format.
• Process and save (button): Execute quantification and save results.

Feature configuration dialog

When you click Add feature, a numbered feature appears in the Features list:

Common fields:

• Name: Custom name for this feature (e.g., IF markers, IF spots).
• Feature type (dropdown): Select "Markers" or "Spots".
• Delete (button): Remove this feature.

Feature types

The quantification tab organizes exports by Features. A feature defines what to quantify and how. In the current version, two feature types are supported: Markers and Spots. This is based on common data types in senescence research:

• Markers: Measure intensity-based markers (e.g., IF markers) within nuclear/cytoplasmic masks.
• Spots: Count spots and analyze colocalization within cell masks.

Markers feature

Measures channel intensity and morphological properties within segmentation labels.

Add channel(s) popup:

Segmentation section:

• Add segmentation: Add a segmentation layer.
• Labels (dropdown per row): Select a labels layer (nuclear or cytoplasmic masks).
• Delete (per row): Remove this segmentation.

Each segmentation added here will be applied to all channels in this feature. SenoQuant will export one cell x marker table per segmentation.

Channels section:

• Add channel: Add an image channel.
• Auto populate channel(s): Add one row per image layer and fill channel names automatically from reader metadata when available.
• Channel (dropdown per row): Select an image layer to measure intensities from.

• Set threshold (checkbox per row): Enable intensity thresholding.

  • Threshold (slider): Threshold level (enabled when checkbox checked). The values set here are linked to the napari layer contrast limits for easy visualization. Mean intensities outside and within the thresholded region will marked in the output files.

  • Method (dropdown): Various thresholding methods:

    • Manual: Set threshold using slider.
    • Otsu: Automatic Otsu thresholding.
    • Yen: Automatic Yen thresholding.
    • Li: Automatic Li thresholding.
    • Isodata: Automatic Isodata thresholding.
    • Triangle: Automatic Triangle thresholding.

    Click the "Auto threshold" button to compute the threshold using the selected automatic method. For more details on these methods, refer to the skimage documentation.

  • Delete (per row): Remove this channel.

Closing the popup or clicking Save will save the settings.

The Auto populate channel(s) button is enabled only after at least one channel or one segmentation is configured in the popup.

ROI section:

Enabled by checking the ROIs checkbox in the main feature configuration box.

• Add ROI: Add a region of interest filter.
• Layer (dropdown): Select a Shapes layer.
• Type (dropdown): "Include" (mark overlapping as 1 in output) or "Exclude" (mark overlapping as 0 in output).
• Delete: Remove this ROI.

Exported columns (Markers)

Morphological metrics (2D images):

• morph_area - Area in pixels.
• morph_area_um2 - Area in µm² (if pixel sizes available).
• morph_perimeter - Perimeter in pixels.
• morph_perimeter_crofton - Crofton perimeter estimate.
• morph_circularity - 4π·area/perimeter² (1.0 = perfect circle).
• morph_eccentricity - 0 (circular) to 1 (elongated).
• morph_solidity - area / convex hull area.
• morph_extent - area / bounding box area.
• morph_feret_diameter_max - Maximum Feret diameter.
• morph_major_axis_length - Major axis of fitted ellipse.
• morph_minor_axis_length - Minor axis of fitted ellipse.
• morph_aspect_ratio - major axis / minor axis.
• morph_orientation - Angle in radians.

Morphological metrics (3D images):

• morph_volume - Volume in pixels.
• morph_volume_um3 - Volume in µm³ (if pixel sizes available).
• Limited shape descriptors (regionprops 3D limitation).

Centroid coordinates:

• 2D: centroid_y, centroid_x (pixels and µm if available).
• 3D: centroid_z, centroid_y, centroid_x (pixels and µm if available).

Intensity metrics (per channel):

• <channel>_mean_intensity - Mean pixel intensity.
• <channel>_integrated_intensity - mean × area × pixel_volume.
• <channel>_raw_integrated_intensity - Sum of pixel values.

Thresholded intensity (when enabled):

• <channel>_mean_intensity_thresholded
• <channel>_integrated_intensity_thresholded
• <channel>_raw_integrated_intensity_thresholded

Thresholding will zero the _thresholded fields of cells with mean intensities outside the thresholding bounds.

Reference columns:

• label_id - Unique ID of the segmentation instance (a nucleus or a cytoplasm).
• file_path - Full source path.
• segmentation_type - "nuclear" or "cytoplasmic".
• overlaps_with - Semicolon-separated list of overlapping labels from other segmentations in this feature, if there are multiple segmentations configured.

ROI columns:

• excluded_from_roi_<name> - 0/1 flag indicating if the cell is excluded by the named ROI.
• included_in_roi_<name> - 0/1 flag indicating if the cell is included by the named ROI.

Spots feature

Measures spot counts and spot-level properties from configured spot labels, with optional cell-segmentation context.

Add channel(s) popup:

Segmentation section:

• Add segmentation: Add a nuclear/cytoplasmic labels layer (optional).
• Segmentation (dropdown per row): Select a cell segmentation layer.
• Delete (per row): Remove this segmentation.

When one or more valid segmentations are configured, SenoQuant exports one pair of files (*_cells and *_spots) per segmentation.

If no valid segmentation is configured, SenoQuant still exports all spots in a single all_spots file (no cells table).

Channels section:

• Add channel: Add a spot channel row.
• Auto populate channel(s): Add one row per image layer, fill channel names from metadata when available, and auto-select matching spot-label layers when they can be resolved.
• Name (text): Custom channel label for output columns and colocalization labels.
• Channel (dropdown per row): Select the image layer used for spot intensity.
• Spots segmentation (dropdown per row): Select the labels layer containing spot instances for this channel.
• Delete (per row): Remove this channel.

A channel row is exported only when both Channel and Spots segmentation are selected.

If a selected layer is missing at runtime, the channel is skipped.

When segmentations are configured, a channel is skipped for a segmentation if the spot-label shape does not match that segmentation shape.

When no segmentations are configured, spot channels must share the same spot-label shape to be exported together.

Spot labels are assigned to cells by centroid position when a segmentation is present. Spots outside segmentation are still exported and marked with within_segmentation = 0.

Closing the popup or clicking Save will save the settings.

The Auto populate channel(s) button is enabled only after at least one channel or one segmentation is configured in the popup.

ROI section:

Enabled by checking the ROIs checkbox in the main feature configuration box.

ROI controls are the same as in the Markers feature.

Colocalization:

• Export colocalization (checkbox): Include colocalization analysis in output.

Colocalization is computed from overlap between spot label masks across channels (not intensity correlation). This is only computed when there are at least 2 valid spot channels for a given feature.

Exported tables (Spots)

Cells table (one per segmentation, only when a segmentation is configured):

• label_id - Cell label ID from the selected segmentation.
• centroid_<axis>_pixels - Cell centroid in pixels (y/x for 2D, z/y/x for 3D).
• centroid_<axis>_um - Cell centroid in micrometers when physical pixel sizes are available.
• Morphology columns (same naming as Markers): e.g., morph_area, morph_volume, morph_*.
• overlaps_with - Semicolon-separated overlapping labels from other configured segmentations in the same feature.
• ROI columns: included_in_roi_<name> and/or excluded_from_roi_<name>.
• <channel>_spot_count - Number of spots assigned to that cell for each configured channel.
• <channel>_spot_mean_intensity - Mean of per-spot mean intensities for spots assigned to that cell.
• colocalization_event_count (when enabled) - Count of unique overlapping spot pairs within the same cell.

<channel> is derived from the channel Name (or the image layer name if Name is empty), sanitized to lowercase with underscores.

Spots table (one per segmentation, or a single all_spots table when no segmentation is configured):

• spot_id - Unique identifier within channel.
• cell_id - Parent cell label ID (0 when not assigned to a segmented cell).
• within_segmentation - 1 when assigned to a segmented cell, 0 otherwise (present only when a segmentation is configured).
• channel - Channel label (custom Name if provided, otherwise the image layer name).
• centroid_<axis>_pixels - Spot centroid in pixels.
• centroid_<axis>_um - Spot centroid in micrometers when physical pixel sizes are available.
• spot_area_pixels (2D) or spot_volume_pixels (3D) - Spot size in pixel units.
• spot_area_um2 (2D) or spot_volume_um3 (3D) - Spot size in physical units when pixel sizes are available.
• spot_mean_intensity - Mean intensity of the spot region in the selected image channel.
• ROI columns: included_in_roi_<name> and/or excluded_from_roi_<name>, evaluated at the spot centroid.
• colocalizes_with (when enabled) - Semicolon-separated list of overlapping spots as <channel_label>:<spot_id>.

Output structure

Results are saved to: <output_folder>/<output_name>/<feature_name>/

Each feature creates its own subfolder containing: - One file per segmentation (Markers). - Spots: - With segmentation(s): two files per segmentation (*_cells and *_spots). - Without segmentation: one file (all_spots). - feature_settings.json (feature configuration snapshot and run metadata).

File naming: - Markers: <segmentation_label>.<format>. - Spots (with segmentation): <segmentation_label>_cells.<format> and <segmentation_label>_spots.<format>. - Spots (without segmentation): all_spots.<format>.