Scientific best practices are only useful if the scientific community can actually use them. Within the OHDSI community, we believe one of the best ways to ensure adoption and consistent application of scientific best practices is by embedding them in open-source software and making the software freely available for all to use.
See for yourself, try any of these demos of open source analytics tools:
ATLAS – a web-based integrated platform for database exploration, standardized vocabulary browing, cohort definition, and population-level analysis:
ACHILLES – a standardized database profiling tool for database characterization and data quality assessment:
Observational Data Management – tools and processes to standardize the structure and content of healthcare data in preparation for observational analyses, including:
- ATHENA standardized vocabularies
- Common data model and standardized vocabularies specifications
- Extract, transform, and load (ETL) design, development, and testing
- Database profiling and data quality assessment
Clinical Characterization – descriptive analyses to support disease natural history and quality improvement, including:
- Cohort definition and phenotype evaluation
- Patient record profiling
- Study feasibility assessment
- Population summarization and comparison
Population-Level Estimation – epidemiologic designs for estimating average treatment effects for medical product safety surveillance and comparative effectiveness, including:
- Comparative cohort analysis
- Self-controlled case series
- Self-controlled cohort
Patient-level prediction – machine learning methods for precision medicine and disease interception, including:
- Regularized regression
- Random forest
- k-nearest neighbors
OHDSI’s open-source software is made freely available on our GitHub repository. Multiple Java based client applications have been developed to provide support for ETL activities. Additional tools for research activities have been developed with HTML5, web-based client applications and a Java based web service layer. Statistical analysis packages have been developed using R.
So, anyone anywhere in the world can build their own environment that can store patient-level observational health data, convert their data to OHDSI’s open community data standards (includng the OMOP Common Data Model), run open-source analytics using the OHDSI toolkit, and collaborate in OHDSI research studies that advance our shared mission toward reliable evidence generation. Join the journey!