EpiWatch is an early-stage research project exploring how surveillance, climate, and vulnerability signals can be translated into weekly health-zone risk briefs to support public-health decision-making.
Cholera remains a leading cause of preventable death in Sub-Saharan Africa. Public-health teams receive surveillance, climate, WASH, and field signals, but few tools help translate them into routine, health-zone-level prioritization.
EpiWatch is exploring whether machine-learning methods can help teams decide where to verify, where to prepare, and when to escalate, starting with cholera in the DRC as a first use case.
We are developing a weekly decision-support brief with health-zone-level risk scoring, combining onset probability estimation and persistence forecasting. The goal is to complement, not replace, existing surveillance workflows and human expertise.
Signals under investigation:
A weekly brief designed to support, not replace, human decision-making.
Each health zone receives a risk level based on available signals, helping teams prioritize where to investigate further or prepare resources.
Outputs are mapped to existing administrative boundaries and surveillance workflows, including DHIS2-compatible structures.
Risk briefs are designed to integrate into existing decision workflows: no new dashboards or software required.
Early-stage research prototype, illustrative output from ongoing model development.
We welcome conversations with researchers, practitioners, and institutions interested in cholera surveillance and early-warning approaches.
EpiWatch is an early-stage research and venture-building project developed by an independent team participating in the Harvard HSIL Venture Building Program. This page is intended for academic, networking, and partnership exploration purposes in the context of the Boston ecosystem delegation, June 2026. It does not constitute a commercial offering, nor an official Harvard or HSIL product.
