The UCSF DiSARM project commemorates four years of impact

For the past 4 years, the DiSARM project has been focused on building spatial analysis tools to support malaria programs in southern Africa. Initially, this included building a user facing mobile and desktop app to support the planning, implementation and monitoring of Indoor Residual Spray campaigns. As of 2018, the MEI has partnered with Akros to integrate the functionality of the DiSARM app into a new app called Reveal. With support from the Bill and Melinda Gates Foundation, and led by Akros, Reveal is being rolled out to support IRS and mass drug administration for malaria and NTDs in several countries across Africa. See for more details.

In parallel, the DiSARM team has been building algorithms to solve important problems facing programs. For example, the team developed a machine learning algorithm to distinguish residential from non-residential buildings which has been used across a number of countries including Swaziland, Botswana, Zambia and Mozambique. The GRID3 project is now leveraging this algorithm to improve population estimates across a number of countries including DRC and Nigeria. Similarly, with funding from the Taskforce for Global Health, the DiSARM team has developed a hotspot mapping algorithm which predicts the locations of disease hotspots on the basis of climatic and environmental variables. The algorithm also provides recommendations on where to sample next in order to find hotspots. Results show that compared to traditional random sampling, when trying to identify hotspot communities, it is possible to reduce sampling effort by up to 93% without losing accuracy. The algorithm can be used to more efficiently find hotspots of infection, or coldspots of intervention coverage, and will be trialed in Tanzania and Niger to support efforts to identify residual hotspots of lymphatic filariasis.

To support the use, integration and extension of these algorithms by the wider community, the DiSARM project has just released an updated version of their documentation site (, allowing modelers and developers to make use of DiSARM algorithms and code. Included in the documentation is information on how to integrate DiSARM, and similar algorithms, with DHIS2, allowing health programs to leverage machine learning and geospatial algorithms from within DHIS2.

All the resources and code of the DiSARM project will continue to be hosted and freely accessible via the project sites and