Implement QC process for looking at raw/MC movies

[mattjdavis]

[skeleton of issue to be fleshed out in WC working group meeting]

There is a desire to implement a new part of the QC process to look at raw data to catch motion correction and line artifact issues. We want to develop a processed with new data outputs that can facilitate this without overburdening operators.

TODO

• brainstorm data outputs (e.g. parts of raw movies) that can serve as sufficient check, or determine convenient way to load and scrub though movies
• meet with pika/scientists to develop process
• Example: ophys_etl_pipelines icket --> X case is failing MC, need correction redo

[matchings]

@samiamseid Do operators look at the raw & motion corrected movies as part of the QC process currently?

[samiamseid]

not for every experiment, just when the QC report has some abnormality (abnormalities in motion correction plot, fluorescence intensity plot, or epilepsy metric normally). The current QC has frequent links to the network filepaths so we just copy paste that into the file browser in order to watch the physio. In cases where an abnormality is found we usually just scrub the video to the point in time indicated by the QC plot