The Israeli based team uses AI technology to locate malaria hotspots and works to eliminate the risk
Malaria is, according to Zzapp Malaria, “one of the world's biggest solvable problems.” A tropical infectious disease carried by mosquitos, malaria kills over 400,000 humans every year, but could, in the near future, be fully eliminated. Zzapp Malaria’s technology uses complex data mapping to identify risk areas and dangerous bodies of water, allowing them to be sprayed, killing off the parasite.
ZZapp Malaria is based in Jerusalem, Israel, where local villages are prey to malaria – one of the motivations for creating this technology. Yet, the team understands that malaria spreads differently in various locations, which is why they’re not only focussing on a solution but a tailored solution that takes local conditions into account. This isn’t the first work in fighting malaria the team have conducted, either. Prior to founding Zzapp, CEO Arnon Houri-Yafin had a key role in developing Parasight, an accurate machine vision-based malaria diagnostics device currently sold in more than 20 countries.
Below, Houri-Yafin answers our questions on behalf of the team.
Where is your team based, how big is the team, and who is involved?
We are based in Tel Aviv, Israel, and have six workers: Arnon Houri-Yafin (CEO), Lea Leiman (algorithm developer), Eugene Rozenberg (software developer), Arbel Vigodny (biology and operation), Michael Ben Aharon (business development) and Yonatan Fialkoff (strategy and community-engagement).
Please can you tell us more about your AI technology and how it can tackle Malaria?
We created an AI-based mobile app for planning, executing, and monitoring of large-scale malaria elimination campaigns. It uses a neural network to extract the location of houses from satellite imagery. Next, it analyzes topography, synthetic-aperture radar (SAR), and satellite imagery to create a heat map of water body probabilities. From there, it moves to an optimization model, combining house locations and water body probabilities, defining the areas to be scanned.
We also developed a software that simulates multiple interventions on a given area and chooses the one that is most cost-effective by choosing the optimal season for launching the operation and deciding which houses should be sprayed. The mobile app (which can work offline) then helps implement the chosen strategy by dividing the area to workable units and allocating treatment areas to workers. It enables them, while in the field, to pinpoint the location of puddles and to easily report all relevant information. Data obtained in the field is later uploaded to the dashboard, where it is automatically processed to identify underperformance or other abnormalities; these are reported in clear graphics to operation managers. Data is also used to help formulate recommendations for ongoing and future operations.
What made you want to enter the $5M IBM Watson AI XPRIZE?
We see a strong fit between our goals and methods to those of the AI XPRIZE – the ambition to address the grandest global challenges by combining cutting-edge science and technology with a real-world problem-solving approach. AI for good is a powerful idea because it enables tailoring and localizing solutions for humanity’s many problems that are too complex for a one-size-fits-all solution. For example, our AI tailors different intervention strategies for a village in the mountainous area of Madagascar compared to a village in the coastal area of Kenya.
Why is what you’re doing different from other current solutions?
In recent years, existing methods for fighting malaria have exhausted their efficacy with scientists warning of a further increasing mosquito resilience, coupled with the threat of the expansion of their population and activity due to global warming. Our approach, on the other hand, addresses the problem at its root and is based on the only proven method to fully eliminate the disease, if applied properly.
Currently, we began devising a plan to eliminate malaria from the island country of São Tomé and Príncipe. Assuming we are successful there, we will be in the unique position of being the first to eliminate malaria from a country in Subsaharan Africa. We then plan to use this unique achievement to receive an endorsement from the WHO and countrywide contracts from African governments and the big donating organizations.
How has the current climate impacted your work?
In light of the recent surge in both COVID-19 cases and malaria cases in Africa, the need for our solution – which is compatible with social distancing guidelines — has become even more compelling. Forced to compromise our presence in the field, we nevertheless were able to have our system used successfully in Ghana and in Zanzibar based on our improved remote training methods
What team accomplishment are you most proud of?
In August 2020, we began a large-scale operation in Obuasi, Ghana in which we are protecting over 170,000 people. Using our technology, fieldworkers located twice as many water bodies compared to the previous operation with extremely low cost, even compared to bed nets.
How has the competition furthered your success? How has it changed you?
Having been going through rigorous technical examinations by XPRIZE's reviewers has helped us define to ourselves the technological and visionary venues we should be seeking as well as to approach investors and apply for grants. While the competition has been contributing
significantly to our progression, it has not changed who we are. All along, we have been feeling that the XPRIZE judging process genuinely seeks to help us accomplish our goals, rather than adjust ourselves to their own predetermined objectives.
Have you collaborated with IBM in any way? How has it impacted your work?
Seasonal timing is crucial in strategizing larviciding operations. However, the intricacy and volatility of meteorological systems – especially in transitional seasons when these operations are preferably conducted – complicate the decision over the best time to start the operation. The IBM team has picked up the gauntlet of processing several sets of complex and rich data for us.
Through IBM's willingness and responsiveness we also had the chance to be exposed to interesting AutoAI tools.
Learn more about the prize here