To develop and validate ERNIE, an AI-driven framework for early detection of infectious diseases using unstructured primary care data.
Key Findings:
ERNIE detected COVID-19-like clusters before the first confirmed case.
It identified RSV patterns during the 2021 surge and flagged simulated West Nile Virus cases.
The framework achieved a consultation-level recall of 0.97, cluster precision of 0.82, and cluster recall of 0.90.
Interpretation:
ERNIE enables early and interpretable detection of emerging health threats using routine clinical text, enhancing surveillance capabilities without reliance on diagnostic codes or laboratory confirmation.
Limitations:
The study's findings are based on data from a specific region, which may limit generalizability.
The reliance on unstructured text data may introduce variability in data quality and interpretation.
Conclusion:
ERNIE represents a significant advancement in AI-driven surveillance for infectious diseases, with potential applications in various health domains.
by Maarten Homburg, Gijs Danoe, Marjolein Y. Berger, Tim olde Hartman, Jean Muris, Andreas Voss, Axel Hamprecht, Maarten F. Brilman, Lilian L. Peters, Matthijs S. Berends
