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
Our association model only accepts binary concepts, which for now is just the presence or absence of a diagnoses, prescription, or lab event. The lab events in particular could benefit from analyzing the actual values of the lab tests.
We could design a clustering model that would be run over each lab event to cluster events into categories (value intervals). We wouldn't be able to say that one category was normal or abnormal, but we could say there exists various groups of typical results and atypical ones.
Note that this will greatly increase the size of our itemset for the association model, so that model will have to be reconfigured to be runnable again.
The text was updated successfully, but these errors were encountered:
Our association model only accepts binary concepts, which for now is just the presence or absence of a diagnoses, prescription, or lab event. The lab events in particular could benefit from analyzing the actual values of the lab tests.
We could design a clustering model that would be run over each lab event to cluster events into categories (value intervals). We wouldn't be able to say that one category was normal or abnormal, but we could say there exists various groups of typical results and atypical ones.
Note that this will greatly increase the size of our itemset for the association model, so that model will have to be reconfigured to be runnable again.
The text was updated successfully, but these errors were encountered: