Two presentations were delivered at the ISEM Global Conference:

1. Universal platform for mosquito population control planning using AI

This presentation focused on the development of a modular, AI-powered platform designed to support mosquito population control through data-driven decision-making. A system was presented that integrates real-time data from AI-enhanced CDC traps, remote sensing, and environmental monitoring with high-performance simulations and machine learning models. The platform was structured into several key modules, including a simulation engine, data mining tools, a graphical user interface, and an AI-based decision support system. Most components were implemented in Python, enabling rapid adaptation and scalability.

2. Spatio-temporal optimisation of SIT mosquito population control - reinforcement learning approach

In this work, a reinforcement learning approach was applied to optimise the spatio-temporal dynamics of the Sterile Insect Technique (SIT). A dynamic mosquito population model was used in combination with reinforcement learning to identify optimal timing, location, and quantity for the release of sterile male mosquitoes. Results from simulation experiments demonstrated that adaptive, data-driven strategies can significantly improve population suppression when compared to static, calendar-based approaches. The importance of integrating real-time monitoring and environmental data into SIT planning was emphasised.