Decentralized Geomarker Assessment for Multi-Site Studies
Questions? Problems? Please file an issue on GitHub and/or see our troubleshooting guide.
DeGAUSS is a decentralized method for geocoding and deriving community and individual level environmental characteristics while maintaining the privacy of protected health information. It is a standalone and versatile software application based on containerization. This means that geomarker assessment is reproducible, standardized, and can be computed on at scale. Importantly, DeGAUSS is executable on a local machine -- it does not require extensive computational resources and PHI is never exposed to a third party or the internet, making it ideal for geomarker assessment in a multi-site study. Please see detailed documentation within the wiki:
- A background on geocoding, geomarkers, and private health information
- Installing Docker
- Reproducibility and Versioning
- Geocoding with DeGAUSS
- Geomarker Assessment with DeGAUSS
- Troubleshooting
- Memoisation
| image | description | version | source | build |
|---|---|---|---|---|
degauss/geocoder |
batch geocoding |  |
 |
 |
degauss/census_tracts |
census tract FIPS ID |  |
 |
 |
degauss/dist_to_major_roadway |
distance to TIGER/Line 2015 S1100 line in meters | |
|
 |
degauss/acs_income |
census tract median household income from 2015 5-year ACS | |
|
 |
degauss/geocoder_slim |
return geocoded text string as JSON |  |
 |
 |
degauss/cchmc_batch_geocoder |
geocoding, census tract, deprivation index |  |
 |
 |
degauss/crew_census |
historical census tracts and data built for CREW |  |
 |
Please see CONTRIBUTING.md for guidelines on how to contribute to the development of DeGAUSS. Pull requests for new containers using the DeGAUSS framework are always welcome but early visibility is highly encouraged; we often create containers that are not yet publicly available and would love to still work together. Automated tests for containers are conducted through Travis, see .travis.yml for details.
If you use this software in a scientific publication, please consider citing one of our publications:
- Cole Brokamp, Chris Wolfe, Todd Lingren, John Harley, Patrick Ryan. Decentralized and Reproducible Geocoding and Characterization of Community and Environmental Exposures for Multi-Site Studies. Journal of American Medical Informatics Association. 25(3). 309-314. 2018. Download.
- Cole Brokamp. DeGAUSS: Decentralized Geomarker Assessment for Multi-Site Studies. Journal of Open Source Software. 2018. Download.