diff --git a/ML/0-data_prep.R b/ML/0-data_prep.R index ad4a7d2..d88d892 100644 --- a/ML/0-data_prep.R +++ b/ML/0-data_prep.R @@ -72,6 +72,13 @@ patients <- dplyr$tbl(db, "patients") %>% dplyr$select(-anchor_year, -anchor_year_group, -dod) %>% dplyr$collect() +admissions <- dplyr$tbl(db, "admissions") %>% + dplyr$select(subject_id, race, hadm_id) %>% + dplyr$collect() + + +admissions <- admissions %>% dplyr$mutate(dplyr$across(race, factor)) + # usable data set is using chart time as it appears # LIS uses different id's for groups of tests # BMP and CBC Results together