For Gene- and Protein Expression you have to specify settings for the nuclei segmentation as well as for quantification of the gene expression in the respective step.

Parameter

Description

Segm. Method

Only visible in the Nuclei step.
Selects the method for segmentation.

–

Automatic

Uses threshold values that are determined automatically from the histogram based on the Otsu method. For all possible threshold values, the Otsu method calculates the variance of intensities on each side of the respective threshold. It minimizes the sum of the variances for the background and the foreground.

–

Manual

Sets the threshold manually by clicking on the regions in the image that you want to segment, or by using the Threshold control displayed below.

–

Semantic

Uses semantic segmentation based on machine learning to automatically segment (fluorescently labeled) cell nuclei. This requires an installation of the 3rd party Python Tools during the installation of ZEN.

–

Instance

Uses an AI model for instance segmentation to automatically segment (fluorescently labeled) cell nuclei. The AI models for instance segmentation need the software Docker Desktop to run.

AI Model

Only visible if Semantic or Instance is selected.
Selects the model for segmentation from a dropdown which includes your own trained and imported networks as well as a default neural network for the segmentation of (fluorescently labeled) nuclei. Note that you can only use models trained on a single channel.

AI Model Version

Only visible if Instance is selected.
Selects the model version you want to use.

Model Class

Only visible if Semantic or Instance is selected.
Selects the model class used for segmentation.

Min. Confidence

Only visible if Semantic or Instance is selected.
Sets the minimum confidence in % for the prediction of every object.

Threshold

Only visible in the Nuclei step and if Manual is selected.
Sets the threshold for the pixel intensity used by the segmentation. For an RGB image you can set the threshold for each color channel individually.

–

Low

Defines the lowest pixel intensity considered for the segmentation.

–

High

Defines the highest pixel intensity considered for segmentation.

–

Undoes the last change.

Pick to Segment

Only visible in the Nuclei step and if Manual is selected.

–

+

Enables you to expand the currently segmented regions by the gray values/colors of the objects subsequently clicked on.

–

-

Enables you to reduce the currently segmented regions by the gray values/colors of the objects subsequently clicked on.

Histogram

Only visible in the Nuclei step and if Manual is selected.
In the histogram you can change the lower and upper threshold value by dragging the lower or upper adjustment handle or shift the entire highlighted area between the lower and upper threshold value.

BG Subtraction

–

On

Applies a rolling ball background subtraction to the image.

–

Off

Applies no background subtraction to the image.

Area

Only available in the Nuclei step.
Sets the lowest and highest value for the area of the objects. For more information, see the list of Measurement Features.

Circularity

Only available in the Nuclei step.
Sets the lowest and highest value for the roundness of the objects. For more information, see the list of Measurement Features.

Pick to Include
+

Enables you to expand the values employed for the region filters (area and circularity) and the mean intensity in the second step by clicking on objects in the image.

Ring Distance

Only available in the Nuclei step.
Sets the distance between the inner border of the ring and the border of the segmented nuclei.

Ring Width

Only available in the Nuclei step.
Sets the width of the ring where the gene expression is measured.

Intensity Mean

Only available in the Gene- and Protein Expression step.
Sets the lowest and highest value for the mean intensity measurement of the selected channel. For more information, see the list of Measurement Features. If the mean intensity of a cell falls into that defined range, it is considered as a positive cell.