Understanding how diseases and infections spread through a population requires knowing when and where cases develop and in which patients. Contemporary public health data is aggregated and doesn't include the metadata essential to effective prediction of future outbreaks.
Public health information should be open, free and available to everyone who wants it, without compromising patient privacy.
New data sources can facilitate analysis and prediction, enable what-if scenarios, and enhance dynamic visualizations. All in real time.
OpenDIGIT is a grand vision for free and open public health and epidimiological data and the infrastructure that supports it. We imagine three facets...
Develop and deploy mobile and desktop clients that allow health professionals to record confirmed and suspected cases of infectious diseases. Work with global, national, regional, and local public health providers to tap into existing data streams while ensuring patient confidentiality. Build services that are easy and convenient to use and can work in near-real time and in regions with little or no internet connectivity.
Create web clients that leverage the data services to develop new insights into disease spread, epidemic prediction, and temporal dynamics of emergent diseases. Implement novel and advanced visualizations to make the most of the new data. Provide built-in pattern recognition and machine learning algorithms to enable what-if scenarios by the public and by health care providers.
Creating maps of disease spread in real time
Filtering by age and report veracity
Comparing spread trajectories of multiple diseases
Health records are spread out over the internet, often in inaccessible forms. Existing applications use search term trends, social media data, or or user reports to understand disease dynamics. The World Health Organization, along with national and regional public health officials collect and distribute data about infections, but often at spatially aggregated and temporally low resolution. Moreover, few existing data sources provide programmatic access to raw data in digestable formats suitable for integrating into interactive and engaing client applications.A new approach for data collection, management, and distribution is the future.
Anyone interested in opening up public health data should get in touch! There's lots to be done in many different areas.