You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
An option to remove stacks with a ROI signal below a certain threshold
Motivation
When segmenting ROI such as cells or nuclei, cellpose appears to normalize the signal in each image before segmenting, particularly when doing 2D analysis. This leads to an over-extension of the cellpose mask in the z-directions, where diffuse, out-of-focus, signal is interpreted as an ROI. This over-extension leads to challenges in proper parameter selection, and segmentation.
Pitch
An additional flag to allow for "pruning" of analyzed z-frames based on the max ROI signal intensity of the z-frame.
Alternatives
This could feasibly exist either as a part of the setup_ROI_segmentation and run_ROI_segmentation notebooks, or exist as a standalone pre-processing utility script/notebook. The second option would depend on puncta analysis notebooks to be able to batch analyze stacks of different sizes. I do not know that the notebooks as-is currently allow for different sized z-stacks.
Additional context
The text was updated successfully, but these errors were encountered:
🚀 Feature
An option to remove stacks with a ROI signal below a certain threshold
Motivation
When segmenting ROI such as cells or nuclei, cellpose appears to normalize the signal in each image before segmenting, particularly when doing 2D analysis. This leads to an over-extension of the cellpose mask in the z-directions, where diffuse, out-of-focus, signal is interpreted as an ROI. This over-extension leads to challenges in proper parameter selection, and segmentation.
Pitch
An additional flag to allow for "pruning" of analyzed z-frames based on the max ROI signal intensity of the z-frame.
Alternatives
This could feasibly exist either as a part of the setup_ROI_segmentation and run_ROI_segmentation notebooks, or exist as a standalone pre-processing utility script/notebook. The second option would depend on puncta analysis notebooks to be able to batch analyze stacks of different sizes. I do not know that the notebooks as-is currently allow for different sized z-stacks.
Additional context
The text was updated successfully, but these errors were encountered: