Efforts to model and forecast infectious disease transmission are critical both to the science of understanding pathogens and in planning public health responses to improve health outcomes and save lives. The barriers to developing these models are high — due to required scientific expertise, the cost to develop, calibrate, and run bespoke software solutions, and myriad combinations of potential critical factors. This has kept modeling out of reach for many local governments where so many public health decisions are made.
The NIH-funded EpiMoRPH project envisions a platform to lower these barriers and place trustworthy predictions within reach. epymorph is the computational core of this developing platform. Our key value: by reducing the time from concept to results, we maximize experimentation and discovery. epymorph's modular design enables rapid construction of spatial epidemiological models while maintaining flexibility. Data-wrangling utilities for common sources let you focus on the parts of your model that count. And built-in features for fitting model parameters to observational data (e.g., disease surveillance data) and producing forecasts with uncertain quantification bring state-of-the-art mathematical methods right to your virtual environment. See how we are leveraging Python's object-oriented paradigm and mature scientific computing ecosystem to tackle some of the hardest problems in public health.
https://github.com/NAU-CCL/epymorph