Supervised learning for infection risk inference using pathology data

BMC Medical Informatics and Decision Making

Research Article

Bernard Hernandez¹  Pau Herrero¹  Benjamin Evans¹  Pantelis Georgiou¹  Christofer Toumazou¹  Luke S. P. Moore²  Timothy Miles Rawson²  Alison H. Holmes² 

[1] Centre for Bio-Inspired Technology, Department of Electrical and Electronic Engineering, Imperial College London, B422 Bessemer Building, South Kensington Campus, SW7 2AZ, London, UK;Health Protection Unit in Healthcare Associated infections and Antimicrobial Resistance, Imperial College London, 8th floor Commonwealth Building, Hammersmith Hospital Campus, Acton, W12 0NN, London, UK;

关键词: Antimicrobial resistance;    Infection;    Machine learning;    Supervised learning;    Predictive modelling;    Biochemical markers;    Decision support;    Behaviour change;   

DOI  :  10.1186/s12911-017-0550-1

 received in 2017-04-04, accepted in 2017-11-19,  发布年份 2017

来源: Springer

【 摘 要 】

BackgroundAntimicrobial Resistance is threatening our ability to treat common infectious diseases and overuse of antimicrobials to treat human infections in hospitals is accelerating this process. Clinical Decision Support Systems (CDSSs) have been proven to enhance quality of care by promoting change in prescription practices through antimicrobial selection advice. However, bypassing an initial assessment to determine the existence of an underlying disease that justifies the need of antimicrobial therapy might lead to indiscriminate and often unnecessary prescriptions.MethodsFrom pathology laboratory tests, six biochemical markers were selected and combined with microbiology outcomes from susceptibility tests to create a unique dataset with over one and a half million daily profiles to perform infection risk inference. Outliers were discarded using the inter-quartile range rule and several sampling techniques were studied to tackle the class imbalance problem. The first phase selects the most effective and robust model during training using ten-fold stratified cross-validation. The second phase evaluates the final model after isotonic calibration in scenarios with missing inputs and imbalanced class distributions.ResultsMore than 50% of infected profiles have daily requested laboratory tests for the six biochemical markers with very promising infection inference results: area under the receiver operating characteristic curve (0.80-0.83), sensitivity (0.64-0.75) and specificity (0.92-0.97). Standardization consistently outperforms normalization and sensitivity is enhanced by using the SMOTE sampling technique. Furthermore, models operated without noticeable loss in performance if at least four biomarkers were available.ConclusionThe selected biomarkers comprise enough information to perform infection risk inference with a high degree of confidence even in the presence of incomplete and imbalanced data. Since they are commonly available in hospitals, Clinical Decision Support Systems could benefit from these findings to assist clinicians in deciding whether or not to initiate antimicrobial therapy to improve prescription practices.

【 授权许可】

CC BY   
© The Author(s) 2017

【 图 表 】

12864_2017_3500_Article_IEq12.gif

12864_2016_3440_Article_IEq27.gif

12864_2017_4030_Article_IEq1.gif

12864_2016_2682_Article_IEq39.gif

12864_2016_3440_Article_IEq23.gif

12864_2016_2888_Article_IEq7.gif

12864_2017_3487_Article_IEq3.gif

12864_2016_2682_Article_IEq36.gif