MAT-MPNN: A Mobility-Aware Transformer-MPNN Model for Dynamic Spatiotemporal Prediction of HIV Diagnoses in California, Florida, and New England

2025-11-17

AI Summary

This study addresses the limitation of conventional models in capturing non-adjacent spatiotemporal dependencies among counties for HIV diagnosis rate forecasting. We propose MAT-MPNN, a novel framework comprising a Mobility-aware Graph Generator (MGG) that dynamically constructs weighted spatial graphs integrating human mobility and geographic proximity—thereby relaxing the rigid fixed-neighborhood assumption—and jointly leverages Transformer architectures for temporal dynamics modeling and Message Passing Neural Networks (MPNNs) for spatial message propagation. Evaluated on county-level data from California, Florida, and New England, MAT-MPNN reduces mean squared prediction error by 39.1%, 27.9%, and 12.5%, respectively, and improves the Prediction Model Choice Criterion (PMCC) by 3.5%, 7.7%, and 3.9% over baseline models. To our knowledge, this is the first work to systematically integrate mobility-driven dynamic graph learning with spatiotemporal graph neural networks, achieving significantly enhanced fine-grained accuracy in forecasting HIV transmission trends.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Human Immunodeficiency Virus (HIV) has posed a major global health challenge for decades, and forecasting HIV diagnoses continues to be a critical area of research. However, capturing the complex spatial and temporal dependencies of HIV transmission remains challenging. Conventional Message Passing Neural Network (MPNN) models rely on a fixed binary adjacency matrix that only encodes geographic adjacency, which is unable to represent interactions between non-contiguous counties. Our study proposes a deep learning architecture Mobility-Aware Transformer-Message Passing Neural Network (MAT-MPNN) framework to predict county-level HIV diagnosis rates across California, Florida, and the New England region. The model combines temporal features extracted by a Transformer encoder with spatial relationships captured through a Mobility Graph Generator (MGG). The MGG improves conventional adjacency matrices by combining geographic and demographic information. Compared with the best-performing hybrid baseline, the Transformer MPNN model, MAT-MPNN reduced the Mean Squared Prediction Error (MSPE) by 27.9% in Florida, 39.1% in California, and 12.5% in New England, and improved the Predictive Model Choice Criterion (PMCC) by 7.7%, 3.5%, and 3.9%, respectively. MAT-MPNN also achieved better results than the Spatially Varying Auto-Regressive (SVAR) model in Florida and New England, with comparable performance in California. These results demonstrate that applying mobility-aware dynamic spatial structures substantially enhances predictive accuracy and calibration in spatiotemporal epidemiological prediction.

Problem

• Predicting county-level HIV diagnosis rates using spatiotemporal data
• Capturing complex spatial dependencies beyond geographic adjacency
• Improving HIV transmission forecasting with mobility-aware deep learning

Innovation

• Transformer encoder extracts temporal features for prediction
• Mobility Graph Generator combines geographic and demographic data
• Dynamic spatial structures enhance spatiotemporal epidemiological forecasting