Close
    Search for HOT projects, news, people and jobs.

Leveraging ML-generated building data and motorcycle mapping in the DRC and Uganda to enhance the effectiveness of humanitarian response

In combining machine-learning-based algorithms with a motorcycle mapping approach, HOT will generate comprehensive base map data of North Kivu and Ituri provinces in the Democratic Republic of Congo and the bordering districts in western Uganda to better guide and inform local actors, such as health care workers, government representatives and national/international organizations, of where humanitarian assistance should be delivered.

Over the next year, HOT is working to scale-up Missing Maps across current areas of instability and population movement in the African Great Lakes region. Intensified by ongoing conflict in the region, the Ebola outbreak in DRC is now the second largest in history after the 2014 West Africa outbreak. Cases of Ebola virus disease, after expanding into 19 health zones at the height of the outbreak, have been reduced to just 1 health zone as of May 2020 but continue to significantly impact the public health and stability in the region. In western Uganda, districts bordering the affected areas in the DRC face immediate threat due to the porous border and insufficient screening points to evaluate the health status of those passing through points of entry, whether for business, seeking asylum, or personal travel. Availability of common operational data sets (e.g. comprehensive place/village names and location of buildings) are still lacking throughout many provinces of DRC, inhibiting the ability to respond optimally to present outbreaks and build preparedness for future humanitarian action. Addressing data gaps isn’t only a critical challenge for contact tracing but for the Ebola response and COVID-19 outbreaks which are already diverting resources from weakened healthcare systems in these locations.

With funding from Grand Challenges Canada, HOT is working to leverage ML-generated building footprints to generate base map data that will be further built upon through motorcycle mapping, creating maps that will increase the effectiveness of humanitarian health assistance. Geographic scope for this project will include North Kivu and Ituri provinces in the DRC and bordering districts in western Uganda. Through the provision of Maxar’s (formerly DigitalGlobe) satellite imagery and the building footprints generated by Ecopia, HOT will leverage these comprehensive datasets to create an updated base map in OpenStreetMap to guide field mapping in the AOI. In detail, HOT will engage local communities to carry out motorcycle mapping with the aim of closing data gaps such as village names, health facilities and other related points of interest to assist the response to the health crises facing the region.

The resulting project will focus on the most critical humanitarian need in the DRC, better targeting overall project scope to address information needs for the ongoing complex emergency in DRC and refugee crisis in Uganda. The regional project team, based out of DRC and Uganda will train local mappers in producing common humanitarian map data sets that follow national standards. HOT will be working with a wide range of individuals across the geographic scope, including health ministry and government officials, health care workers, humanitarian responders, national and international NGOs, and local community leaders and members, to build institutional knowledge and capacity to use map data to appropriately inform decision-making and respond to local public health needs, both in times of crises and long-term development planning.


Copy-of-Untitled-3.png

This Project is supported by Grand Challenges Canada. Grand Challenges Canada is funded by the Government of Canada and is dedicated to supporting Bold Ideas with Big Impact®.