Probabilistic Case Detection for Disease Surveillance Using Data in Electronic Medical Records

Fuchiang Tsui, Michael Wagner, Gregory Cooper, Jialan Que, Hendrik Harkema, John Dowling, Thomsun Sriburadej, Qi Li, Jeremy Espino, Ronald Voorhees

Abstract


This paper describes a probabilistic case detection system (CDS) that uses a Bayesian network model of medical diagnosis and natural language processing to compute the posterior probability of influenza and influenza-like illness from emergency department dictated notes and laboratory results. The diagnostic accuracy of CDS for these conditions, as measured by the area under the ROC curve, was 0.97, and the overall accuracy for NLP employed in CDS was 0.91.

Full Text:

PDF HTML


DOI: https://doi.org/10.5210/ojphi.v3i3.3793



Online Journal of Public Health Informatics * ISSN 1947-2579 * http://ojphi.org