sima.ROI.Nonbooleanmask: unable to convert a non-boolean mask to polygons

[wcy20...@gmail.com]

Hi,

https://drive.google.com/file/d/0B6HcXHKONk3taHJKSVJBdk1wRTg/view?usp=sharing

Attached is the data I used which is a series of TIF files, which only contains one channel 2. I also attached the code here.

The first question I encountered is after I created sequences, corrected motion, did segmentation, I ran roi_buddy.py to see how the segmentation worked, but there was no ROIs circuled with different colors, and in cmd, I got this 'sima.ROI.Nonbooleanmask: unable to convert a non-boolean mask to polygons'. (these are done by question1.py)

Then I manually circled one ROI, and plot the signal for that ROI using question2.py. The second question is that even if I used show(), I cannot see the figure, I think this may be due to some issue with the matplotlib rather than sima package. So I save the signals arrary and load it in python3 which I can plot the signals, it worked.

My third question is that what segmentation method do you recommend for my case?

Thank you so much.

Eric

[Patrick Kaifosh]

Hi,

https://drive.google.com/file/d/0B6HcXHKONk3taHJKSVJBdk1wRTg/view?usp=sharing

Attached is the data I used which is a series of TIF files, which only contains one channel 2. I also attached the code here.

The first question I encountered is after I created sequences, corrected motion, did segmentation, I ran roi_buddy.py to see how the segmentation worked, but there was no ROIs circuled with different colors, and in cmd, I got this 'sima.ROI.Nonbooleanmask: unable to convert a non-boolean mask to polygons'. (these are done by question1.py)

Is your issue the same as this one? (see the bottom three posts at this link: https://github.com/losonczylab/sima/issues/168#issuecomment-107594567)
 

Then I manually circled one ROI, and plot the signal for that ROI using question2.py. The second question is that even if I used show(), I cannot see the figure, I think this may be due to some issue with the matplotlib rather than sima package. So I save the signals arrary and load it in python3 which I can plot the signals, it worked.

Yes, probably some issue with how you have configured matplotlib. Sorry, I don't think I can help you there.
 

My third question is that what segmentation method do you recommend for my case?

What are you imaging? How densely packed are the cells? STICA is a more generally applicable method, whereas the CA1PC approach is designed for very densely packed CA1 pyramidal cells and may not be too generally applicable in its current form.