Detection of Hospital Acquired Infections in Sparse and Noisy Swedish Patient Records
Claudia Ehrentraut, Jörg Tiedemann, Hercules Dalianis, Hideyuki Tanushi · Analytics for Noisy Unstructured Text Data · 2012
Hospital Acquired Infections (HAI) pose a significant risk on patients’ health while their surveillance is an additional work load for hospital medical staff and hospital management. Our overall aim is to build a system which reliably retrieves all patient records which potentially include HAI, to reduce the burden of manually checking patient records by the hospital staff. In other words, we emphasize recall when detecting HAI (aiming at 100%) with the highest precision possible. The present study is of experimental nature, focusing on the application of Naive Bayes (NB), Support Vector Machines (SVM) and a C4.5 Decision Tree as well as different preprocessing methods to the problem, and the evaluation of the efficiency of this approach. The three machine learning algorithms are applied in three drafted classification tasks: binary classification, twostep classification and multi-class classification. Our machine learning approach is presented as an alternative to rule-based systems which are more common in this task. The results of the three classification task were overall similar. SVM yielded the highest recall 91% with the best overall performance, i.e., a F2-score of 87.4%. In each classification task, the classifiers were applied on a small and noisy dataset, generating results which pinpoint the potentials of using learning algorithms for detecting HAI. Further research will have to focus on optimizing the performance of the classifiers and to test them on larger datasets.