Bayesian Selective Latent Inference for Wastewater-First Influenza Monitoring
2026-06-08 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors look at how tracking the flu using wastewater can give early signs of the disease in a community but isn’t enough alone to measure true infection levels. They created a new method called Bayesian Selective Latent Inference (BSLI) that starts with wastewater data and smartly decides if other official health reports are needed or if it should avoid making uncertain conclusions. Their approach uses a Bayesian framework to weigh evidence and decide the best times to gather more data or to stop, aiming for accurate and cautious flu monitoring. They tested BSLI on real public data and found it balances performance and careful decision-making better than previous methods.
Wastewater surveillanceInfluenza monitoringBayesian inferenceSelective decision makingLatent burdenSource ambiguityCost-calibrated policyBellman optimalityScientific gatesForecasting
Authors
Yixuan Zhang, Yang Song, Hao Wang, Samir Bhatt, Hengguan Huang
Abstract
Wastewater influenza surveillance can reveal community circulation before clinical reporting, but wastewater alone is not a fully identifiable proxy for human burden. Existing wastewater models assume a fixed evidence set, while generic evidence-acquisition methods treat official surveillance streams as interchangeable costly features. We cast wastewater-first influenza monitoring as a selective decision problem: starting from mandatory wastewater evidence, the system must decide whether wastewater is sufficient, which delayed official stream to query next, and when abstention is the only scientifically defensible action under source ambiguity. We propose Bayesian Selective Latent Inference (BSLI), a principled Bayesian method that maintains a posterior over latent burden and identifiability, certifies answerability through explicit scientific gates, and optimizes query-stop decisions with an exact cost-calibrated Bellman policy. We prove the key variational, answerability, Bellman-optimality, and one-dimensional cost-calibration properties. On a fixed public-data benchmark with 5,933 forecasting episodes and 3,102 source-ambiguity episodes, BSLI improves the matched-budget cost-performance frontier while preserving conservative abstention under source ambiguity.