Collaborative Research: eMB: Math-model Informed Neural Network (MINN) based on Emerging Mathematics in Mosquito-Dengue Biology

Currently, about half of the world's population, about 4 billion people, live in areas with a risk of dengue infection, with further increased public health concerns due to recent evolutionary adaptation of dengue-transmitting mosquitoes to colder places. This project develops and uses mathematical models and computational methods, including the novel Math-model Informed Neural Networks (MINN) based on emerging mathematics in mosquito-dengue biology. Proper management guidelines, identified through data-driven models, help healthcare professionals mitigate the burdens of dengue infection, thereby improving the quality of life for dengue-infected patients and their families. The outcomes of this project not only fundamentally advance the fields of mathematical biology and quantitative biology but also have a simultaneous broad and highly positive societal impact. In addition, this project offers extensive interdisciplinary research training opportunities for undergraduate and graduate students in mathematics and biology. The project will expand research and educational opportunities to various programs for students, as well as junior and senior researchers, and will incorporate the research into an interdisciplinary mathematical biology course. This project will focus on three aims: (a) Develop Math-model Informed Neural Networks (MINN) capturing emerging mathematics in climate-dependent mosquito-dengue biology. (b) Analyze models and develop MINN-based methods to estimate epidemic thresholds. (c) Develop MINN-based user-friendly online platforms for public health policy evaluations and healthcare accessibility. The novel models, validated using data from our collaborators (biologists/environmentalists) from Nepal, will incorporate an experimentally observed mosquito life cycle and dengue transmission. The models and related MINNs will be used through a user-friendly online platform to evaluate public health policies and calculate healthcare accessibility for dengue control in spatially heterogeneous environmental conditions. This contribution will have a significant positive and practical impact on developing public health policies to prevent dengue virus infection, as well as advance the development of sophisticated mathematical and machine learning models to help explain the role of the environment and mobility in the complex biological systems of mosquito-dengue interactions. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.