A study of the transferability of influenza case detection systems between two large healthcare systems

Abstract

ObjectivesThis study evaluates the accuracy and transferability of Bayesian case detection systems (BCD) that use clinical notes from emergency department (ED) to detect influenza cases.MethodsA BCD uses natural language processing (NLP) to infer the presence or absence of clinical findings from ED notes, which are fed into a Bayesain network classifier (BN) to infer patients’ diagnoses. We developed BCDs at the University of Pittsburgh Medical Center (BCDUPMC) and Intermountain Healthcare in Utah (BCDIH). At each site, we manually built a rule-based NLP and trained a Bayesain network classifier from over 40,000 ED encounters between Jan. 2008 and May. 2010 using feature selection, machine learning, and expert debiasing approach. Transferability of a BCD in this study may be impacted by seven factors: development (source) institution, development parser, application (target) institution, application parser, NLP transfer, BN transfer, and classification task. We employed an ANOVA analysis to study their impacts on BCD performance.ResultsBoth BCDs discriminated well between influenza and non-influenza on local test cases (AUCs > 0.92). When tested for transferability using the other institution’s cases, BCDUPMC discriminations declined minimally (AUC decreased from 0.95 to 0.94, p<0.01), and BCDIH discriminations declined more (from 0.93 to 0.87, p<0.0001). We attributed the BCDIH decline to the lower recall of the IH parser on UPMC notes. The ANOVA analysis showed five significant factors: development parser, application institution, application parser, BN transfer, and classification task.ConclusionWe demonstrated high influenza case detection performance in two large healthcare systems in two geographically separated regions, providing evidentiary support for the use of automated case detection from routinely collected electronic clinical notes in national influenza surveillance. The transferability could be improved by training Bayesian network classifier locally and increasing the accuracy of the NLP parser.

Artificial intelligence and machine learning in emergency medicine

“…ML models have also been developed to classify suicide‐related communication on twitter . Natural language processing ML models have been applied to emergency medicine clinical documentation to detect influenza . ML may also assist with the detection of novel viral outbreaks, having been used to develop accurate and dynamic forecasting models for Zika virus using Google Trends …”

“…43 Natural language processing ML models have been applied to emergency medicine clinical documentation to detect influenza. 44 ML may also assist with the detection of novel BOX 1. What is… Artificial intelligence (AI) • Can be defined as 'the theory and development of computer systems able to perform tasks normally requiring human intelligence'.…”

Development and Evaluation of a Fully Automated Surveillance System for Influenza-Associated Hospitalization at a Multihospital Health System in Northeast Ohio

“…Automated surveillance systems developed at one health care system can potentially be replicated at other institutions. 20,21 With the vast majority of acute care hospitals having now adopted EMRs, 22 replication of this database development at other health care systems is feasible. Of the true cases not detected by the automatic system, 58 (83%) were patients who had a positive test collected at an outpatient visit, were sent home, and were later admitted within 14 days.…”

Deep learning facilitates rapid classification of human and veterinary clinical narratives

“…Domain adaptation . The portability of trained algorithms on independent domains has previously been used as a metric of model robustness in systems that leverage NLP and machine learning [58]. We evaluated the ability of our trained LSTM models to be used in a cross-species context.…”