Open peiyaoli opened 7 years ago
It's not trivial - we haven't identified a method yet but I imagine it would be similar to MIMIC - examining the data, looking to see if there are trends in the documentation (e.g. it occurs every 12 hours), then defining a rule for endtime (e.g. no documentation after 16 hours).
I agree that this is not trivial - different treatments would require different end times. For example CRRT calculations may be required as frequently as 100 to 160 times in a 24 hour period and nurse are documenting hourly so an end time of 16 hours without documentation is not meaningful.
Could you please to give us a demo? @alistairewj
Not sure I'll have time to look at this in the next few weeks. @tompollard will be releasing an intake output table soon and that might have some data elements to look at. Also we should probably move this discussion to eicu-code as it's not about documentation of the database.
On Nov 11, 2017 9:39 AM, "peiyaoli" notifications@github.com wrote:
Could you please to give us a demo? @alistairewj https://github.com/alistairewj
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/MIT-LCP/eicu-website/issues/36#issuecomment-343669677, or mute the thread https://github.com/notifications/unsubscribe-auth/ABOSdLT6VGovQdxW2tm-wl4AlyO4oIg-ks5s1bGsgaJpZM4P9d4n .
It seems that endtime of treatment is not available here? So how could we find the treatment duration? Such as duration of CRRT? @tompollard @alistairewj