O-61 Smart traps and internet of things-based entomological surveillance integrated with artificial intelligence for real-time mosquito density mapping

Methods: Peer-reviewed studies, technical reports, and grey literature published between 2015 and 2025 were screened. Eligible studies assessed smart traps, IoT-enabled surveillance, or AI applications for detecting, identifying, or estimating medically important mosquito species (Aedes, Anopheles, Culex). Data were extracted based on technology type, AI methods, outcomes, and implementation context, with thematic synthesis identifying patterns, challenges, and public health implications.

Results: Eight studies described smart-trap and IoT technologies, including acoustic-based surveillance, computervision and image-processing systems, IoT-enabled environmental monitoring, and integrated platforms. Acoustic systems detected wing-beat frequencies with 65-80% species identification accuracy, while AI-based computervision models, including YOLO and convolutional neural networks, achieved 91–97% accuracy under controlled conditions. IoT systems integrated environmental sensors for predictive modeling and geospatial mapping, enabling large-scale, near-real-time surveillance. In addition, six studies highlighted AI applications, with deep-learning architectures consistently achieving ≥90% accuracy for species identification and up to 100% for density estimation. Real-time applications included drone-based habitat mapping and automated traps.

Conclusion: AI- and IoT-enhanced surveillance enables automated, continuous data capture, high detection accuracy, and integration with environmental and spatial data, supporting targeted vector control. Challenges include field validation gaps, environmental variability, computational demands, and ethical considerations. These technologies offer a scalable, data-driven approach to modernizing mosquito surveillance and informing public health interventions.