Electronic Health Records (EHR) data can provide valuable insights into inpatient trajectories. By representing blood tests and vital signs as multivariate time-series (MVTS), unsupervised Hidden Markov Models (HMM) can be trained to classify each day of hospital admission as one of 17 states. Clinical interpretation of these HMM states revealed their associations with inpatient mortality and specific diagnoses. Machine learning models trained with MVTS data showed promising performance in predicting inpatient mortality, indicating the potential for developing decision-support tools for EHR systems.
Electronic Health Records (EHR) data can provide novel insights into inpatient trajectories. Blood tests and vital signs from de-identified patients' hospital admission episodes (AE) were represented as multivariate time-series (MVTS) to train unsupervised Hidden Markov Models (HMM) and represent each AE day as one of 17 states. All HMM states were clinically interpreted based on their patterns of MVTS variables and relationships with clinical information. Visualization differentiated patients progressing toward stable 'discharge-like' states versus those remaining at risk of inpatient mortality (IM). Chi-square tests confirmed these relationships (two states associated with IM; 12 states withR1 diagnosis). Logistic Regression and Random Forest (RF) models trained with MVTS data rather than states had higher prediction performances of IM, but results were comparable (best RFmodel AUC-ROC: MVTS data = 0.85; HMM states = 0.79). ML models extracted clinically interpretable signals from hospital data. The potential of ML to develop decision-support tools for EHR systems warrants investigation.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据