A Spatially Validated Machine-Learning Framework for Mapping Alpha-Gal Syndrome Exposure Potential in Illinois

Random forest and XGBoost–based county indices for lone star tick–linked alpha-gal syndrome exposure in Illinois show strong concordance with CDC suspected-AGS categories (Spearman rho = 0.710 and 0.721).

• Why it matters: Heterogeneous tick surveillance and limited clinical AGS data make it hard to pinpoint where people in Illinois may face higher exposure risk.
• What they did: Using data from 102 Illinois counties, eight ecological and machine-learning models were evaluated with geographic cross-validation to predict lone star tick relative abundance, then combined with reported ehrlichiosis rates to form county-level AGS exposure-risk indices.
• The result: Across 97 counties, both indices matched CDC suspected-AGS categories, and at the highest category threshold the odds ratios per 0.10-unit index increase were 5.02 (random forest) and 3.63 (XGBoost), enabling spatially explicit surveillance prioritization.

AI summary of the abstract · may contain errors · not medical advice · read the paper.