Researchers from the University of South Florida, in collaboration with the company 'Digitome', have developed a smart mosquito trap based on artificial intelligence, aiming to accelerate the identification of disease-carrying species, particularly mosquitoes associated with the spread of dengue fever. Laboratory tests showed the system's ability to identify mosquito species with an accuracy exceeding 98 percent.
Integration of Modern Technologies
The system combines built-in cameras, artificial intelligence algorithms, and cloud computing to determine the mosquito species and sex, and to detect whether it has fed on blood. The trap attracts mosquitoes using light or bait, then draws them inside via a fan and traps them on an adhesive surface, where two built-in cameras capture images of the insects, which are analyzed by algorithms trained on a large database.
Accelerating the Monitoring Process
This mechanism accelerates the identification of species compared to traditional methods that rely on examining samples under a microscope, a process that can take two to three days. Trap data can be linked to an interactive map to monitor mosquito spread and identify areas where disease-carrying species concentrate, such as the 'Aedes aegypti' mosquito known for transmitting the dengue virus.
Future Expansion
The team integrated trap data into a dedicated layer within the 'Global Mosquito Monitoring' dashboard, which includes more than half a million reports from insect monitoring participants. Researchers are conducting additional field tests in collaboration with mosquito control authorities in St. Johns and Hillsborough counties in Florida, paving the way for expanding the technology's use and making it commercially available.
The team hopes the system will enable the identification of mosquito species on the same day samples are collected, rather than waiting several days, thereby helping to accelerate response efforts and monitor areas requiring intensified control measures.
Source: Syrian Arab News Agency (SANA)
Comments (0)