What happened
A research team at Dhaka University has developed 'MAST-CloudNet,' an AI-powered, internet-connected mosquito trap designed to detect Aedes mosquitoes, the primary vectors for dengue fever. The device uses acoustic lures to attract mosquitoes, captures images of them, and employs an AI model to identify the species and transmit data to a cloud-based dashboard for real-time monitoring.
The MAST-CloudNet device was developed through a collaboration between the Microsystems and Nanoengineering Lab and the Genome-Guided Vector Control and Innovation Lab at Dhaka University. The research was published on October 8 in the journal IEEE Transactions on Instrumentation and Measurement.
The trap functions by emitting specific sound frequencies (350-500 Hz) that mimic the wingbeat of female Aedes mosquitoes to attract males. Once trapped, an internal camera captures images of the insects, which are then processed by an AI model to confirm the species. The system displays these counts on a web dashboard, allowing for remote monitoring of mosquito activity.
According to the researchers, the device costs approximately Tk12,000 to manufacture, which is significantly lower than the cost of traditional CDC light traps (Tk25,000) or BG-Sentinel traps (Tk20,000).
Initial testing included a three-day outdoor trial on the Dhaka University campus, where the device successfully captured 49 Aedes mosquitoes despite adverse weather conditions, including heavy rain and wind.
Why it matters
This development offers a potentially cost-effective and scalable method for public health surveillance in regions prone to dengue. By automating the identification process and providing real-time data on mosquito population density, the system could enable more precise, proactive vector control measures. The researchers report that the device is significantly cheaper to produce than conventional traps, which could facilitate wider deployment in urban areas to identify emerging hotspots before outbreaks escalate.
Dengue fever remains a significant public health challenge in many tropical regions. Current surveillance methods often rely on manual labor and expensive equipment, which can limit the frequency and geographic scope of monitoring.
By leveraging AI for automated species identification, the MAST-CloudNet system reduces the need for manual entomological analysis, potentially allowing health authorities to monitor larger areas with greater efficiency.
The use of male Aedes mosquitoes as an indicator is a strategic choice, as they emerge earlier than females and remain near breeding sites, providing a 'leading indicator' for potential dengue risk in a specific area.
The ability to network multiple traps to a central server further enhances the utility of the system, as it allows for the mapping of population trends across entire city districts.
Interactive Mechanism: How It Actually Works
Explore the underlying technology behind this development interactively.
Which component of an AI application is the machine-learning model itself?
What to watch next
The research team plans to conduct a field deployment of 10-15 units in residential areas of the Dhaka South City Corporation between June and July 2027. Future development efforts are focused on improving the device's durability against environmental factors, integrating solar power, and implementing LoRaWAN connectivity to reduce reliance on local Wi-Fi networks for data transmission.
The primary milestone for the project is the planned deployment of 10-15 units in Dhaka South City Corporation during the 2027 dengue season. This trial will be critical for assessing the device's performance in real-world, high-density urban environments.
Technical improvements are ongoing, specifically regarding the device's physical resilience to sun and rain, which is essential for long-term outdoor operation.
The team is also working on hardware iterations, including a solar-powered version and the integration of LoRaWAN technology, which would allow the traps to function in areas without reliable Wi-Fi infrastructure.