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

Tsui F-CWagner MMCooper GQue J, Harkema H,  Dowling J, Sriburadej T,  Li Q, Espino JU, Voorhees R.  Probabilistic case detection for disease surveillance using data in electronic medical records.  Online J Public Health Inform.  2011 Dec;3(3) Epub 2011 Dec 22.  PubMed PMID: 23569615.  PMCID:PMC3615792.

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.    Key words: case detection, disease surveillance, influenza, electronic medical records.

Publication Year: 
2011
Publication Credits: 
Tsui F-C, Wagner MM, Cooper G, Que J, Harkema H, Dowling J, Sriburadej T, Li Q, Espino JU, Voorhees R.
AttachmentSize
PDF icon Tsui.pdf943.77 KB
^