Click on a row to explore the results for that model. When you wish to explore a different model, then select the new result row and the tabs will be updated.Demo Video
OHDSI PatientLevelPrediction Model Repository
This shiny application is an interface for viewing, uploading and downloading healthcare prediction models.
Our aim is to improve the reproducibility and transparency of prediction models. Reproducibiliy is improved by the ability to access code that can be run against any OMOP common data model data to implement and evaluate any model in this repository. Transparency is improved by providing all the details for the model development and performance within the app. This will help us decide which models to trust by examining extensive external validation results.
How to use?
Click on the 'Library' option in the left-hand menu to explore the models in the repository. You can select a model to explore further by selecting the corresponding row in the library table. This will then populate the settings/performance/model views.
How to download a model?
Click on the 'download development' to download an R package that can be executed on any OMOP common data model data to replicate the model development process in your data. Click the 'download validation' to download an R package that can be executed on any OMOP common data model data to validate the choosen model in your data. You can then upload the new model or validation results to the repo.
How to upload results?
Click the 'upload' left-hand menu option, fill out the form and click submit. The results will be reviewed and you will recieve and email once the results are in the repository.
Please send any questions to: …@…com
The following databases were used in this study:
|OptumDoD||Optum® De-Identified Clinformatic Data Mart Database – Date of Death (DOD)||USA||Claims||2000-2019|
|CCAE||IBM MarketScan® Commercial Database||USA||Claims||2000-2019|
|MDCD||IBM MarketScan® Multi-State Medicaid Database||USA||Claims||2006-2019|
|MDCR||IBM MarketScan® Medicare Supplemental Database||USA||Claims||2000-2019|
|JMDC||Japan Medical Data Center||Japan||Claims||2000-2019|
|optumEhr||Optum® de-identified Electronic Health Record Dataset||USA||EHR||2006-2019|
All databases obtained IRB approval or used deidentified data that was considered exempt from IRB approval.
Net Benefit Plot
Development R Package
Click here to download an R package that contains all the settings requires to replicate the model development using any OMOP CDM database.
Validation R Package
Click here to download an R package that contains all the settings requires to validate the existing model using any OMOP CDM database.