In hdshea/MIMIC4db: Package for building, loading and referencing the Google BigQuery verion of the MIMIC-IV

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

MIMIC4db

This work references data from the Medical Information Mart for Intensive Care MIMIC-IV database. MIMIC-IV is a large, freely-available database comprising de-identified health-related data from patients who were admitted to the critical care units of the Beth Israel Deaconess Medical Center from 2008-2019. Detailed information can be obtained on the MIMIC-IV website.

MIMIC4db provides a tightly bound set of routines to reference the Google BigQuery version of the MIMIC-IV v1.0 database using base access routines from the DBI R package with an appropriate bigrquery DBIConnection.

Installation

You can install MIMIC4db from GitHub with:

# install.packages("devtools")
devtools::install_github("hdshea/MIMIC4db")

Basic Usage

These are basic examples which show you how to solve a common problem.

Connecting to the database:

library(MIMIC4db)

con <- bigrquery::dbConnect(
    bigrquery::bigquery(),
    project = bigrquery::bq_test_project(),
    quiet = TRUE
)

library(MIMIC4db)
# To connect to the database, you must provide an appropriate project (in the `your-project` 
# slot below) which has billing set up and to which you have write access.  This project 
# must be approved to access the PhysioNet BigQuery MIMIC schemas.

con <- bigrquery::dbConnect(
    bigrquery::bigquery(),
    project = "your-project",
    quiet = TRUE
)

Using the meta data functions:

core <- m4_meta_data()
core <- core[core$module == "mimic_core", ]
for(i in 1:length(core)) {
  cat(stringr::str_c("select * from physionet-data.", 
                     core[i,"module"], ".", core[i,"table"], "\n"))
}

neuro <- m4_service_decsriptions()
neuro <- neuro[startsWith(neuro$short_description, "Neurologic"), ]
neuro[,c("service", "short_description")]

Using the two base functions:

d_items <- m4_select_data(
  con, 
  "select * from physionet-data.mimic_icu.d_items order by itemid limit 5"
)
d_items

d_items <- m4_get_from_table(con, "d_items", where = "where itemid <= 220048")
d_items

Using the tables specific data access functions:

tab <- m4_chartevents_items(con)
tab[c(1:5), ]

pat <- m4_patients(con, cohort = 16904137)
pat

svs <- m4_services(con, cohort = 16904137)
svs

evt <- m4_procedureevents(con, cohort = 16904137)
evt[c(1:5), ]

Using the common pattern data access functions:

# Combined and preprocessed patient and admission data
patadm <- m4_patient_admissions(con, cohort = 16904137)
patadm[, c(1:6)]
patadm[, c(7:10)]
patadm[, c(11:14)]
patadm[, c(15:19)]

# Combined and preprocessed patient, admission and icustay data
icudet <- m4_patient_icustays(con, cohort = 16904137)
icudet[, c(1:6)]
icudet[, c(7:10)]
icudet[, c(11:14)]
icudet[, c(15:19)]
icudet[, c(20:22)]
icudet[, c(23:25)]

Disconnecting from the database:

bigrquery::dbDisconnect(con)

Citations for MIMIC-IV publications:

Johnson, A., Bulgarelli, L., Pollard, T., Horng, S., Celi, L. A., & Mark, R. (2021). MIMIC-IV (version 1.0). PhysioNet. https://doi.org/10.13026/s6n6-xd98.

PhysioNet: Goldberger, A., Amaral, L., Glass, L., Hausdorff, J., Ivanov, P. C., Mark, R., ... & Stanley, H. E. (2000). PhysioBank, PhysioToolkit, and PhysioNet: Components of a new research resource for complex physiologic signals. Circulation [Online]. 101 (23), pp. e215–e220.

hdshea/MIMIC4db documentation built on Dec. 20, 2021, 3:45 p.m.