kyle.belanger/DHSC-Capstone
1
0
Fork 0
Code Issues 1 Pull requests Projects Releases Packages Wiki Activity Actions

Update 0-data_prep.R

Browse source
This commit is contained in:
Kyle Belanger 2023-01-06 14:48:53 -05:00
parent 19918b04a3
commit 8a348759cc
1 changed files with 49 additions and 34 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

83
ML/0-data_prep.R
View file

@ -23,7 +23,7 @@ db <- dbConnect(
#item list shows two different numbers for a few tests, second set of items do not have
# any results that are on the same samples as TSH and Free T4
test_list <- c(
test_list_cmp <- c(
50862 #Albumin
,50863 #Alkaline Phosphatase
,50861 #Alanine Aminotransferase (ALT)
@ -42,39 +42,43 @@ test_list <- c(
,50995 #Thyroxine (T4), Free
)
test_list_bmp <- c(
51006 #Urea Nitrogen
,50893 #Calcium, Total
,50882 #Bicarbonate
,50902 #Chloride
,50912 #Creatinine
,50931 #Glucose
,50971 #Potassium
,50983 #Sodium
,50993 #Thyroid Stimulating Hormone
,50995 #Thyroxine (T4), Free
)
# TSH Ref Range from File 0.27 - 4.2 uIU/mL
# Free T4 Ref Range from File 0.93 - 1.7 ng/dL
# load data ---------------------------------------------------------------
ds <- dplyr$tbl(db, "labevents") %>%
dplyr$filter(itemid %in% test_list) %>%
dplyr$select(-charttime,-storetime) %>%
tidyr$pivot_wider(
id_cols = c(subject_id,specimen_id)
,names_from = itemid
,values_from = valuenum
) %>%
dplyr$filter(!is.na(`50993`) & !is.na(`50995`)) %>%
dplyr$collect()
# most likely will not use this as there are not as many complete rows. However
# gathering it just in case.
# first is using specimen id, usable data set is using chart time as it appears
# LIS uses different id's for groups of tests
#
# ds_cmp <- dplyr$tbl(db, "labevents") %>%
# dplyr$filter(itemid %in% test_list_cmp) %>%
# dplyr$select(-charttime,-storetime) %>%
# tidyr$pivot_wider(
# id_cols = c(subject_id,specimen_id)
# ,names_from = itemid
# ,values_from = valuenum
# ) %>%
# dplyr$filter(!is.na(`50993`) & !is.na(`50995`)) %>%
# dplyr$filter(dplyr$across(where(is.numeric), ~!is.na(.x))) %>%
# dplyr$collect()
ds %>% dplyr$filter(dplyr$across(where(is.numeric), ~!is.na(.x)))
count <- data.frame(colSums(is.na(ds))) %>% tibble::rownames_to_column()
testds <- readr::read_csv(
here("ML","data-unshared", "labevents.csv")
,col_types = "_d_ddTT_d______"
,n_max = 100
)
#using chart time instead of spceimen id results in less NA values.
# total protien still have very low resulting
ds1 <- dplyr$tbl(db, "labevents") %>%
dplyr$filter(itemid %in% test_list) %>%
ds_cmp <- dplyr$tbl(db, "labevents") %>%
dplyr$filter(itemid %in% test_list_cmp) %>%
dplyr$select(-storetime) %>%
tidyr$pivot_wider(
id_cols = c(subject_id,charttime)
@ -84,13 +88,24 @@ ds1 <- dplyr$tbl(db, "labevents") %>%
dplyr$filter(!is.na(`50993`) & !is.na(`50995`)) %>%
dplyr$collect()
count2 <- data.frame(colSums(is.na(ds1))) %>% tibble::rownames_to_column()
#this keeps failing if run as part of the above query. Moving here to keep going
# keeps only rows that have values for all columns
ds_cmp <- ds_cmp %>% dplyr$filter(dplyr$if_all(.fns = ~!is.na(.)))
counts <- count %>%
dplyr$left_join(count2)
# using charttime, total of 5,424 rows with all values filled in
ds1 %>% dplyr$filter(dplyr$across(where(is.numeric), ~!is.na(.x)))
ds_bmp <- dplyr$tbl(db, "labevents") %>%
dplyr$filter(itemid %in% test_list_bmp) %>%
dplyr$select(-storetime) %>%
tidyr$pivot_wider(
id_cols = c(subject_id,charttime)
,names_from = itemid
,values_from = valuenum
) %>%
dplyr$filter(!is.na(`50993`) & !is.na(`50995`)) %>%
dplyr$collect()
ds_bmp <- ds_bmp %>% dplyr$filter(dplyr$if_all(.fns = ~!is.na(.)))
# close database ----------------------------------------------------------