UOW AI Device Detects Disease-Carrying Mosquitoes

Associate Professor Kiran Trivedi at the University of Wollongong has developed a low‑cost, portable device that uses artificial intelligence to identify three major disease‑carrying mosquito species—Aedes, Anopheles and Culex—by the distinctive frequency of their wingbeats. The system runs on Tiny Machine Learning (TinyML) on a small, low‑power chip and operates offline, offering a faster alternative to traditional surveillance that requires collecting larvae and sending samples to a laboratory.

In prototype tests, the AI model achieved 88.3 % accuracy using publicly available recordings. Ongoing work focuses on improving noise filtering so the device can reliably isolate wingbeat sounds in noisy field conditions. Trivedi said, “That’s a critical step if we want it to work in real‑world environments.” The technology aims to help remote and developing communities monitor malaria, dengue, Zika and other mosquito‑transmitted diseases more effectively.