MedGNN: Capturing the Links Between Urban Characteristics and Medical Prescriptions

2025-04-07

AI Summary

This study addresses three key challenges in modeling the impact of urban sociodemographic and environmental factors on health: (1) capturing complex nonlinear effects, (2) representing spatial dependence, and (3) ensuring model interpretability. To this end, we propose a spatiotemporally explicit graph neural network framework for medical prescription prediction. Leveraging data from 4,835 London neighborhoods and the MEDSAT multi-source dataset, we introduce a novel two-hop spatial graph incorporating location/locational embeddings and over 150 urban features. We further pioneer geographic principal component analysis (Geo-PCA) to enable interpretable discovery of medicine–urban associations, identifying new mechanistic clues—including greenness, NO₂ concentration, and canopy evapotranspiration. The model achieves >25% average improvement in predicting six prescription categories (e.g., depression, diabetes). It systematically uncovers three types of health mechanisms: consensus-supported, contested (requiring empirical validation), and previously unrecognized latent mechanisms—establishing an interpretable, generalizable urban health modeling paradigm for interdisciplinary public health decision-making.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Graphical Models
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Understanding how urban socio-demographic and environmental factors relate with health is essential for public health and urban planning. However, traditional statistical methods struggle with nonlinear effects, while machine learning models often fail to capture geographical (nearby areas being more similar) and topological (unequal connectivity between places) effects in an interpretable way. To address this, we propose MedGNN, a spatio-topologically explicit framework that constructs a 2-hop spatial graph, integrating positional and locational node embeddings with urban characteristics in a graph neural network. Applied to MEDSAT, a comprehensive dataset covering over 150 environmental and socio-demographic factors and six prescription outcomes (depression, anxiety, diabetes, hypertension, asthma, and opioids) across 4,835 Greater London neighborhoods, MedGNN improved predictions by over 25% on average compared to baseline methods. Using depression prescriptions as a case study, we analyzed graph embeddings via geographical principal component analysis, identifying findings that: align with prior research (e.g., higher antidepressant prescriptions among older and White populations), contribute to ongoing debates (e.g., greenery linked to higher and NO2 to lower prescriptions), and warrant further study (e.g., canopy evaporation correlated with fewer prescriptions). These results demonstrate MedGNN's potential, and more broadly, of carefully applied machine learning, to advance transdisciplinary public health research.

Problem

• Understanding urban factors' impact on health prescriptions
• Overcoming nonlinear and spatial modeling limitations
• Improving prediction accuracy for medical outcomes

Innovation

• Spatio-topological graph neural network framework
• 2-hop spatial graph with urban embeddings
• 25% prediction improvement over baselines