Predictive models for pressure ulcers from intensive care unit electronic health records using Bayesian networks

Abstract

BackgroundWe develop predictive models enabling clinicians to better understand and explore patient clinical data along with risk factors for pressure ulcers in intensive care unit patients from electronic health record data. Identifying accurate risk factors of pressure ulcers is essential to determining appropriate prevention strategies; in this work we examine medication, diagnosis, and traditional Braden pressure ulcer assessment scale measurements as patient features. In order to predict pressure ulcer incidence and better understand the structure of related risk factors, we construct Bayesian networks from patient features. Bayesian network nodes (features) and edges (conditional dependencies) are simplified with statistical network techniques. Upon reviewing a network visualization of our model, our clinician collaborators were able to identify strong relationships between risk factors widely recognized as associated with pressure ulcers.MethodsWe present a three-stage framework for predictive analysis of patient clinical data: 1) Developing electronic health record feature extraction functions with assistance of clinicians, 2) simplifying features, and 3) building Bayesian network predictive models. We evaluate all combinations of Bayesian network models from different search algorithms, scoring functions, prior structure initializations, and sets of features.ResultsFrom the EHRs of 7,717 ICU patients, we construct Bayesian network predictive models from 86 medication, diagnosis, and Braden scale features. Our model not only identifies known and suspected high PU risk factors, but also substantially increases sensitivity of the prediction - nearly three times higher comparing to logistical regression models - without sacrificing the overall accuracy. We visualize a representative model with which our clinician collaborators identify strong relationships between risk factors widely recognized as associated with pressure ulcers.ConclusionsGiven the strong adverse effect of pressure ulcers on patients and the high cost for treating pressure ulcers, our Bayesian network based model provides a novel framework for significantly improving the sensitivity of the prediction model. Thus, when the model is deployed in a clinical setting, the caregivers can suitably respond to conditions likely associated with pressure ulcer incidence.

Accuracy of dengue clinical diagnosis with and without NS1 antigen rapid test: Comparison between human and Bayesian network model decision

“…Bayesian networks can represent complex relationships among variables involved in a disease process. Previous studies have shown that Bayesian networks can be used to assist diagnosis in a variety of diseases, such as breast cancer, hepato-biliary diseases, liver disorder, pneumonia, and pressure ulcer [20,[24][25][26][27]. Our study also demonstrated that BN models can be effectively applied for dengue diagnosis.…”

Clinical decision support system to assess the risk of sepsis using Tree Augmented Bayesian networks and electronic medical record data

“…Bayesian networks are very popular for capturing the uncertain knowledge in medicine and have been extensively used in the diagnosis of diseases. [12][13][14] The tree augmented naive (TAN) Bayesian network is a specific class of Bayesian networks used for classification problems and applied widely in health care. 15 The TAN Bayesian networks perform better than alternate classifiers when biomarkers are correlated, and at the same time maintain the mathematical simplicity.…”

Introduction: the International Conference on Intelligent Biology and Medicine (ICIBM) 2016: special focus on medical informatics and big data

“…The project detecting and quantifying new and redundant information over longitudinal clinical notes presented by Zhang et al discussed an automated technique with statistical language models to improve data quality [9]. Kaewprag et al [10] explored a predictive analysis of pressure ulcer in intensive care unit patient records, which helps clinicians identify relationships between risk factors associated with pressure ulcers.…”

“…These papers cover a wide range of topics including Knowledge and Data Personalization [1,2], Social Media Applications to Healthcare [3,4], Clinical Natural Language Processing [5,6], Patient Safety Analyses [7,8], and Data Mining Using Electronic Health Records [9,10].…”

“…The 2016 International Conference on Intelligent Biology and Medicine (ICIBM 2016) was held December [8][9][10]2016 at the Texas Medical Center, Houston, Texas, USA. ICIBM 2016 included four keynote lectures, four conference invited talks, four workshops/tutorials, eight scientific sessions, and a poster session.…”