Are MiniAn results robust with these issues? (sample video included)

[Yifang-Yuan]

Hi,
We have beed recording using one-photon miniscope V4 for several days, I think MiniAn can give relatively reliable results in most cases if the recordings are stable. But I have some animals' FOV or miniscope attachment are not that good.
Here's two sample videos:
https://drive.google.com/drive/folders/1CIqKpI7KYtP340Cak6N9lu2ikozZ6EX-?usp=sharing

For example1, there’s some motion and flash (can be seen clearer on the right panel subtraction mode), I processed by MiniAn, the spatial filter seems fine(but cells are too dense, might be some "false positive"). I know MiniAn have glow removing process, but I’m still wondering if such large amount of flash will affect the transient of the temporal results?

For example2, there’re also motions, I think MiniAn is fine to deal with the motion. But the Field of View is "too good" that we can even see signals from dendrites, and some units are firing together and connected at times. I’m not sure whether MiniAn could separate these units and if the single unit's fluorescence traces could be affected.

It would be the best if the developer can try these videos and share some thoughts about parameter choosing on unstable recordings like these.