Zzapp is a software system that supports the planning and implementation of malaria elimination operations. Zzapp uses artificial intelligence to identify malaria hotspots and optimize interventions for maximum impact. Zzapp's map-based mobile app conveys the AI strategies to field workers as simple instructions, ensuring thorough implementation.
Malaria transmission takes place where water bodies and human populations converge: water bodies are necessary for mosquito larvae to develop, and humans act as the reservoir for the Plasmodium parasites responsible for malaria, and as a source of blood for mosquitoes.
To determine the population density in different areas, Zzapp uses machine vision to identify houses and other structures from satellite imagery. Topographical maps as well as satellite images are analyzed using convolutional neural networks (a form of AI) to predict the presence of water bodies suitable for mosquito reproduction. Through a combination of these data, the malaria transmission hotspots are flagged and prioritized, and appropriate interventions are matched for each location.
Since many of the water bodies where mosquitoes breed are seasonal, and form with rains, weather is an important factor in timing interventions targeting these water bodies. In addition, rains can wash away both the larvae and the larvicides (larva-specific insecticides) from the water bodies. In collaboration with Zzapp, IBM Watson’s AI and Data Science Elite Team has developed a weather analysis module that predicts the abundance of water bodies based on weather data, allowing Zzapp to better time interventions, and more accurately determine the resources required to implement them.
Mobile app and dashboard
A unique element of Zzapp’s system is the ability to effectively convey detailed AI strategies to large teams of field workers, as well as verify their implementation. First, a web dashboard is used to delineate the areas field workers will be tasked with covering. A map-based mobile app then guides workers through the two stages of the operation: mapping the areas for water bodies, and treating these water bodies at the correct intervals.
The app — To ensure workers do not miss a single water body, the app highlights their path on the map display so they can keep track of the coverage achieved. Each water body identified is reported on the app, including its GPS location, a photo, size, and other characteristics. Once the area has been mapped, the app keeps track of the spraying schedule automatically, so that no larvae get a chance to develop in the water bodies. During the workday, the app requires no network connectivity, storing all data on the worker’s mobile device. When the workers are able to obtain an internet connection, all data is synced to the dashboard, allowing operation managers to keep track of progress on a daily basis.
The dashboard — The dashboard shows both high-level information such as neighborhoods sprayed and coverage achieved, as well as details about the individual water body, such as how many times it was sprayed, when, and by whom. This empowers managers to identify any issues early and find appropriate solutions.
Zzapp malaria pilot in Ghana | 2017
The rich data collected in the field on the locations of water bodies, as well as their sizes, types, larval positivity and other information, can be used to retrain our algorithms, both to improve their predictions, as well as expand their scope.
To request a pilot or for any inquiry — contact us at: firstname.lastname@example.org