Patient-Centered, Graph-Augmented Artificial Intelligence-Enabled Passive Surveillance for Early Stroke Risk Detection in High-Risk Individuals

TL;DR

This paper presents a novel AI-driven passive surveillance system that uses patient-reported symptoms and graph neural networks to detect early stroke risk in high-risk individuals, achieving high specificity and positive predictive value.

Contribution

The study introduces a hybrid machine learning approach combining symptom language and temporal data for early stroke risk detection, grounded in patient language and validated with EHR data.

Findings

High specificity (1.00) and positive predictive value (1.00) in stroke risk detection

Good sensitivity (0.72) in 90-day window for early detection

Effective use of patient language for high-precision risk screening

Abstract

Stroke affected millions annually, yet poor symptom recognition often delayed care-seeking. To address risk recognition gap, we developed a passive surveillance system for early stroke risk detection using patient-reported symptoms among individuals with diabetes. Constructing a symptom taxonomy grounded in patients own language and a dual machine learning pipeline (heterogeneous GNN and EN/LASSO), we identified symptom patterns associated with subsequent stroke. We translated findings into a hybrid risk screening system integrating symptom relevance and temporal proximity, evaluated across 3-90 day windows through EHR-based simulations. Under conservative thresholds, intentionally designed to minimize false alerts, the screening system achieved high specificity (1.00) and prevalence-adjusted positive predictive value (1.00), with good sensitivity (0.72), an expected trade-off prioritizing precision, that was highest in 90-day window. Patient-reported language alone supported high-precision, low-burden early stroke risk detection, that could offer a valuable time window for clinical evaluation and intervention for high-risk individuals.