Mechanism-Empowered Multivariate Time Series Forecasting Model: Application to Tuberculosis Prediction

Li Danyu

Keywords: Multivariate Time Series Forecasting Model, Spatiotemporal framework, Source reduction, mechanism-driven dimensionality reduction

TL;DR: This study provides a fresh perspective for enhancing epidemic forecasting and exploring source reduction measures for industrial activities, demonstrating the feasibility of AI-assisted public health strategies and green production.

Abstract: Among the current global health challenges, tuberculosis, as a highly contagious chronic disease, remains one of the major public health problems worldwide. Despite significant progress made in the past decades, new challenges, including systematic and effective downscaling, accurate prediction of disease incidence, and implementation of source reduction measures, have added to the difficulty of tuberculosis control. In view of the limitations of the recently proposed EIGHT prediction models in terms of prediction accuracy, this study adopts the Learnable Decomposition and Dual Focus Module Model (Leddam) and then introduces a novel mechanism-supported multivariate spatiotemporal series framework, termed LCHHA-Leddam, to address the challenges in tuberculosis forecasting through an investigation of coal power generation in China. This framework substantially simplifies the complexity of tuberculosis prediction, enhances accurate dimensionality reduction, and improves traceability. It also enhances the explanatory power and accuracy of the Leddam model in the field of tuberculosis prediction. This study provides a fresh perspective for enhancing epidemic forecasting and exploring source reduction measures for industrial activities, demonstrating the feasibility of AI-assisted public health strategies and green production.

Primary Area: applications to physical sciences (physics, chemistry, biology, etc.)

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 6961