Design, Development and Deployment of Customisable Mobile and IoT Systems to Enhance Mosquito Surveillance

Abstract

Mosquito-borne diseases pose significant public health challenges in tropical and subtropical regions, requiring precise and efficient surveillance methods. Traditional field data collection methods often lack the accuracy, timeliness and efficiency required to control outbreaks. This research advances mosquito field surveillance by designing, developing, and deploying two digital solutions customised for distinct environmental settings in Northeast Brazil and Madeira Island: a Mobile Surveillance System and an Internet of Things (IoT) Environmental Monitoring System. Developed through iterative processes and stakeholder engagement, these adaptable and scalable systems enhance the accuracy and granularity of data collection. The Mobile Surveillance System was adapted to regional requirements: in Brazil, it comprises a mobile app for field agents and a web platform for supervisors; in Madeira, it combines both functionalities into a single, unified mobile app. The implementation of both applications shares a common architecture that not only proves the systems’ generalisability but also boosts operational efficiency and data accuracy by digitising field data, supporting field agents, and aiding supervisors in managing activities. The IoT-based Environmental Monitoring System, featuring a five-layer architecture, including Arduino microcontrollers, weather and water sensors, and a water pump, autonomously and continuously captures high-resolution data, offering deeper insights into environmental influences on mosquito populations and supporting precise, location-specific predictive modelling. Detailed statistical analyses using Bayesian hierarchical models and correlation studies were conducted to pinpoint critical environmental predictors of mosquito breeding. Furthermore, a neural network classification-based predictive model was developed, enhancing weekly mosquito presence predictions by analysing temporal and sequential environmental patterns. This research distinguishes itself through its real-world deployments, addressing infrastructure, logistics, and technology challenges, while underscoring the importance of a process-oriented approach in tool development and recommending longitudinal deployments to assess the long-term impact of these technologies on mosquito population control and disease management in diverse environments.