Skip to content

Random Forest Patient Level Prediction of Alzheimer's Disease 10 years in advance using EHR data. Includes a permuted feature importance (PFI) analysis knee point thresholding.

Notifications You must be signed in to change notification settings

dbmi-pitt/ad_prediction_PLP

Repository files navigation

AD Prediction using the OHDSI Patient Level Prediction library

Random Forest Patient Level Prediction of Alzheimer's Disease 10 years in advance using EHR data. Includes a permuted feature importance (PFI) analysis.

To replicate:

  1. Instantiate the two cohort definitions in the cohort-definitions folder

  2. Install the R PatientLevelPrediction package using the code ADRandomForestJune2022-Atlas (use RStudio to help)

  3. Configure the code in [ADRandomForestJune2022-Atlas/extras/CodeToRun.R] to connect to your database

  4. Run the code and review the output

About

Random Forest Patient Level Prediction of Alzheimer's Disease 10 years in advance using EHR data. Includes a permuted feature importance (PFI) analysis knee point thresholding.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published