Unveiling the Power of AI and Citizen Science: A New Approach to Detecting Invasive Mosquitoes in Madagascar
In a groundbreaking development, researchers have harnessed the power of Artificial Intelligence (AI) and the collective efforts of citizens to identify the Anopheles stephensi mosquito in Madagascar. This discovery comes at a critical time as the mosquito poses a growing threat to malaria control across Africa. The World Health Organization (WHO) has identified the spread of Anopheles stephensi as a significant challenge to malaria elimination efforts, particularly in hard-to-reach urban areas.
The study, published in the journal Insects, showcases how local communities and digital technology can be leveraged to fill gaps in mosquito surveillance. It all began with a single smartphone photo submitted by residents of Antananarivo, Madagascar's capital, through the GLOBE Observer app. This photo, taken in 2020, revealed a mosquito larva collected from a tyre, and its discovery two years later sparked concern about the potential spread of Anopheles stephensi.
The researchers developed a citizen science tool using AI image recognition algorithms, trained on thousands of smartphone photos of Anopheles stephensi and other species. This tool successfully confirmed the larva's species, demonstrating the potential of citizen science and AI in combating invasive species. Citizen-generated photos can provide early warning data, enabling authorities to quickly identify new or expanding mosquito populations, according to lead author Ryan Carney.
Traditional mosquito surveillance methods, such as trapping, are costly, time-consuming, and not easily scalable, especially across jurisdictional boundaries. Carney emphasizes the importance of citizen science in complementing ongoing surveillance efforts, especially in densely populated and hard-to-reach regions. Anopheles stephensi is particularly concerning due to its high susceptibility to the malaria parasite and resistance to pesticides. It thrives in cities, breeding in artificial containers, allowing for year-round malaria transmission in densely populated areas.
A previous study indicated that the spread of Anopheles stephensi could put an additional 126 million people at risk of malaria across Africa. To address this, researchers recommend freely available apps like iNaturalist, Mosquito Alert, and NASA's GLOBE Observer for scaling mosquito monitoring in African countries. These apps are accessible globally and offer multiple language options, including Swahili and Arabic, to encourage community engagement.
Citizen scientists can submit close-up photos of mosquitoes or their larvae using smartphones and 60x magnifying clip-on lenses. The Global Mosquito Observations Dashboard aggregates these photos and location data, helping officials identify areas requiring targeted surveillance or control efforts. However, the study also acknowledges limitations, such as the need for public awareness and the technical requirements of using specific lenses for valid AI results.
Despite these challenges, the technology has practical value in Madagascar, according to entomologist Andrianjafy Mbolatiana. He highlights the lack of financial and human resources for vector-borne disease monitoring, making citizen science a valuable complement to standard surveillance methods. However, the study also notes that many Malagasy people may lack smartphones or reliable internet access, potentially limiting participation.
To address these barriers, researchers suggest that public health agencies and malaria control programs should support citizen scientists in obtaining the correct lenses and raise awareness of the technology. Mohga Kamal-Yanni, a global health consultant, emphasizes the importance of placing tools directly in community hands, especially during times of reduced international aid. She believes that citizen science approaches are crucial for tackling malaria from the ground up, filling the gap left by international aid cuts.
