Full Detection Processor - LeonieZ/ACCEPT GitHub Wiki

In case you want to analyze all events present in the images, you have to use the Full Detection processor. In this case all objects in the images of the samples that are processed will be identified using the algorithms described in detail elsewhere (L. Zeune et al., Multiscale Segmentation via Bregman Distances and Nonlinear Spectral Analysis, SIAM Journal on Imaging Sciences 2017 10:1, 111-146).

After detecting all objects, the following measurements of the detected signal are extracted in each fluorescent channel:

• Size (in um^2),
• Eccentricity (circularity measurement between 0 (perfect circle) and 1 (line)),
• Perimeter,
• Mean Intensity,
• Max Intensity,
• Median Intensity,
• Standard Deviation of the Intensity,
• Mass (Sum of the Intensity),
• Perimeter2Area (circularity measurement, 1 for a perfect circle),
• Relative overlay with the signal in the nucleus channel.

To analyze the results and use gates, open the sample in the Sample Visualizer.