Data collected in humanitarian responses is rarely shared in real-time and is often never shared at all for a range of valid reasons related to privacy, ethics and practicalities. Yet shared data could be extremely valuable to guide the optimization of humanitarian action, evaluate its impact, and improve transparency and accountability. DISCO-DHRIVE seeks to build and validate a privacy-preserving DIStributed COllaborative learning platform to meet the ICRC’s needs and enable continuous learning for data-driven humanitarian responses. The implementation strategy covers: Building a suite of AI models adapted to ICRC needs without sharing any data; adapting the DISCO platform to the specific challenges of humanitarian contexts; and creating a structured dialogue within the ICRC data protection unit to evaluate the technology’s use in routine practice.
Expected impact: Enabling the ICRC to be and to remain a trusted manager of sensitive data and information
EPFL Research Group: Dr. Mary-Anne Hartley (Intelligent Global Health (iGH) Group), Prof. Martin Jaggi (Tenure Track Assistant Professor, Machine Learning and Optimization Laboratory).
ICRC: Dr. Javier Elkin (Digital Health Coordinator at the Health Department).