Welcome to healthcare.ai
This package will get you started with healthcare machine learning in R.
Find our code at our Github Repo
What can you do with it?
- Create and compare models based on your data.
- Save and deploy a model.
- Perform risk-adjusted comparisons.
- Do trend analysis following Nelson rules.
- Improve sparse data via longitudinal imputation.
How is it specific to healthcare?
- Longitudinal machine learning via mixed models
- Longitudinal imputation
- Risk-adjusted comparisons
For those starting out
Note: if you're setting up R on an ETL server, don't download RStudio--simply open up RGui
Install the latest release on Windows
Open RStudio and work in the console
How to install the latest version on macOS
Note: If using macOS with healthcare.ai, you'll have to use csv files. We're working on adding MySQL connections. We'd love to hear which other databases your connecting to, so we can provide native support!
- Open the Mac Terminal
- Install Xcode compilers via
- Accept the Xcode license via
sudo xcodebuild -license
- Install Homebrew (the macOS package manager) with
- Install ODBC driver via
brew update && brew install unixODBC
- Open R Studio
- In the console, install RODBC from source with
install.packages('RODBC',type = "source")
- In the console, install other R healthcare.ai prerequisites via
install.packages(c('caret','data.table','doParallel','e1071','grpreg','lme4','lubridate','pROC','R6','ranger','ROCR'),repos = "https://cran.cnr.berkeley.edu/")
- Install healthcare.ai
How to install latest version on Ubuntu (Linux)
- Follow steps 1 and 2 here to install R
sudo apt-get install libiodbc2-dev
sudo apt-get install r-cran-rodbc
- After typing
Install the bleeding edge version (for folks providing contributions)
- Grab prerequisites via the console of RGui or (preferably) RStudio
install.packages(c('caret','data.table','devtools','doParallel','e1071','grpreg','lme4','lubridate','pROC','R6','ranger','ROCR','RODBC'),repos = "https://cran.cnr.berkeley.edu/") library(devtools) devtools::install_github(repo='HealthCatalystSLC/healthcareai-r')
- Load the package you just installed and read the built-in docs
Since the CRAN 0.1.11 release, the following has been added to the bleeding edge version
- Output nightly predictions to a dataframe for use with MySQL, Oracle, etc. See the csv example at ?RandomForestDeployment
- Evaluate in-the-wild performance via AU_ROC and AU_PR scores. See more at ?generateAUC
- Double check that the code follows the examples in the built-in docs
Make sure you've thoroughly read the descriptions found here
If you're still seeing an error, file an issue on Stack Overflow using the healthcare-ai tag. Please provide
- Details on your environment (OS, database type, R vs Py)
- Goals (ie, what are you trying to accomplish)
- Crystal clear steps for reproducing the error
How to help
Check out our github repo.