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
I love suite2p's registration - it is fast and accurate even on low SNR Drosophila calcium imaging data. We do most of our calcium imaging as volumetric multiplane imaging with planes spaced relatively close together. This is so that ROIs can span appropriate regions of a 3d brain structure and because it allows for correcting 3d drift and motion which can be prevalent in preps where it is hard to mechanically stabilize the brain.
I am wondering if there are plans to add 3d registration for volumetric multiplane imaging?
I could imagine a minimal soultion would be rigid "best plane" matching that did not incorporate any interpolation between planes or non-rigid 3d motion, instead simply calculating a integer valued 3d shift in z and applying that. I imagine this would work reasonably well for recordings where planes are spaced very close together.
A more complex solution would be 3d nonrigid registration as is done in caiman, for example.
Thank you for making such a valulable tool. It is safe to say that suite2p has been a game changer and has saved me hundreds of hours over the years!
Attempted alternative approaches:
I have dabbled with caiman 3d registration. haven't found it works as well as Suite2p's plane-by-plane registration in low SNR cases, although I have not tried it out extensively.
Additional Context
No response
The text was updated successfully, but these errors were encountered:
Feature you'd like to see:
Hello,
I love suite2p's registration - it is fast and accurate even on low SNR Drosophila calcium imaging data. We do most of our calcium imaging as volumetric multiplane imaging with planes spaced relatively close together. This is so that ROIs can span appropriate regions of a 3d brain structure and because it allows for correcting 3d drift and motion which can be prevalent in preps where it is hard to mechanically stabilize the brain.
I am wondering if there are plans to add 3d registration for volumetric multiplane imaging?
I could imagine a minimal soultion would be rigid "best plane" matching that did not incorporate any interpolation between planes or non-rigid 3d motion, instead simply calculating a integer valued 3d shift in z and applying that. I imagine this would work reasonably well for recordings where planes are spaced very close together.
A more complex solution would be 3d nonrigid registration as is done in caiman, for example.
Thank you for making such a valulable tool. It is safe to say that suite2p has been a game changer and has saved me hundreds of hours over the years!
Attempted alternative approaches:
I have dabbled with caiman 3d registration. haven't found it works as well as Suite2p's plane-by-plane registration in low SNR cases, although I have not tried it out extensively.
Additional Context
No response
The text was updated successfully, but these errors were encountered: