.jpg)
India, July 22, 2026: Associate Professor Kiran Trivedi from the University of Wollongong has designed an affordable Artificial Intelligence-powered gadget that detects disease-carrying mosquito species by listening to the wing beats of the insect in a matter of seconds! As India's public health has been challenged with mosquito-borne infections like malaria and dengue, Trivedi's invention may pave the way for herculean efforts to alleviate these seasonal infections during the rains. Public health officials are now more responsive in attempts to control outbreaks and notify the public of the need for mosquito monitoring and identification of disease-carrying vector(s). Existing surveillance systems can be augmented with these types of technologies, especially to reduce the burden of monitoring and provide more timely mosquito monitoring, for operations and interventions based on the best current evidence.
The AI device of Trivedi is a Tiny Machine Learning (TinyML) based, stand-alone teaching machine that can learn classification of the three major disease-carrying mosquito species: Aedes, Anopheles, and Culex. Also, because of the architecture of the teaching device, it is possible to classify species based on diagnostics directly at the point of need or in the field.
Traditional surveillance techniques rely on the collection and laboratory analysis of mosquito larvae to verify their species. In contrast, an AI-powered device, created by Professor Trivedi and one of his former students, Harsh Shroff, identifies mosquito species through the analysis of distinct wingbeat sound using an embedded AI model. Unlike other deployed AI models, this one was able to achieve real recorded mosquito sound discrimination with an accuracy of 88.3% with the benefit of processing sound locally for the device in an economically viable way. With an integrated display, microphone, and based on a compact Arduino design, the device has the potential to serve as an affordable and easily transportable device for mosquito population monitoring in the field.
Most recently, the innovation was displayed at the United Nations’s Geneva-based AI for Good Global Summit. Trivedi was invited to illustrate how low-cost edge AI technology can serve the needs of public health through cost-effective and scalable solutions.
Regarding the innovation, Associate Professor Kiran Trivedi of the University of Wollongong India explained: “TinyML lets us migrate artificial intelligence to a typically tiny and portable device, which can be taken advantage of for instant mosquito species identification in the absence of both laboratory and internet resources.”
The technology’s real opportunity is in its scalability. Devices could be networked together, sustaining monitoring over mosquitoes. It would offer real-time information on hotspots for disease outbreaks. This would allow community/public health to be proactive about potential outbreaks, as opposed to being reactive after the outbreak. We are looking to show the capability of low-cost, accessible AI to address continental and global health challenges, specifically in cases where there are minimal to no resources for health surveillance systems.
The device may also be used as part of larger networks for continuously monitoring mosquitoes and producing real-time data for health planning and outbreak preparation. Since AI has opened up new opportunities in health care, the edge AI used in this case is a clear opportunity for health systems and for the AI to have a real return on important health goals such as effective health emergency surveillance
Click Here for More Institutional Activities