I've been working through the 'CellMap' pipeline with datasets I've obtained using multiphoton and lightsheet microscopy. My goal is to count the number of amyloid plaques within cleared mouse brain hemispheres, and measure the size of each detected plaque. However, I'm having trouble with the 'Cell Detection' step.
Regardless of the cell detection parameters I set, and whether I use the higher resolution multiphoton data or lower resolution lightsheet data, I have the same issue: multiple maxima coordinates are detected for each plaque. Often the coordinates are identical, but with different corresponding 'source' and 'size' values (see image attached). Filtering the results by applying different 'source' and 'size' thresholds has not been successful because the 'source' and 'size' values for each duplicated coordinate can be quite similar and any threshold applied seems arbitrary.
I thought I understood how the cell detection parameters should work in theory, but this doesn't seem to have translated in practice, so I have been using a trial-and-error approach to find suitable parameters.
Troubleshooting attempts:
I've set all the cell_detection parameters to 'none', except for:
intensity_detection, shape_detection, and maxima_detection
I have set the intensity_detection settings as follows:
'measure' is set to 'source', 'method' is set to 'max', and shape is set to 'none'
I do not change these settings.
For shape_detection:
I've tried various 'threshold' values
For maxima_detection:
'h_max' is set to 'none' (I tried to manipulate this but it doesn't seem to change anything - this problem was reported in another issue within the tracker);
I've tried various 'shape' values;
I've also tried different 'intensity' values but found that it doesn't help my problem so have left it set to 'none'.
My questions are:
Why might duplicate coordinates be detected, and how could I fix this?
What exactly is the 'source' value and how is it measured?_
E.g. Using p3d.plot([[ws.filename('stitched'), ws.filename('cells', postfix='shape')]]) I can view the output from using the shape_detection parameter. I see that the 'source' value corresponds to intensity values assigned the 'shapes' that are overlaid on my data. However, I'm not sure how this is calculated and why it is different to the original intensity values of my data.
I would be grateful for any guidance to solve this problem.
_Note - I am using the ClearMapstable.yml environment file because of issues with the most recent version
I've been working through the 'CellMap' pipeline with datasets I've obtained using multiphoton and lightsheet microscopy. My goal is to count the number of amyloid plaques within cleared mouse brain hemispheres, and measure the size of each detected plaque. However, I'm having trouble with the 'Cell Detection' step.
Regardless of the cell detection parameters I set, and whether I use the higher resolution multiphoton data or lower resolution lightsheet data, I have the same issue: multiple maxima coordinates are detected for each plaque. Often the coordinates are identical, but with different corresponding 'source' and 'size' values (see image attached). Filtering the results by applying different 'source' and 'size' thresholds has not been successful because the 'source' and 'size' values for each duplicated coordinate can be quite similar and any threshold applied seems arbitrary.
I thought I understood how the cell detection parameters should work in theory, but this doesn't seem to have translated in practice, so I have been using a trial-and-error approach to find suitable parameters.
Troubleshooting attempts:
My questions are:
I would be grateful for any guidance to solve this problem.
_Note - I am using the ClearMapstable.yml environment file because of issues with the most recent version