Use of deep learning to develop continuous-risk models for adverse event prediction from electronic health records.

Early prediction of patient outcomes is important for targeting preventive care. This protocol describes a practical workflow for developing deep-learning risk models that can predict various clinical and operational outcomes from structured electronic health record (EHR) data. The protocol comprises five main stages: formal problem definition, data pre-processing, architecture selection, calibration and uncertainty, and generalizability evaluation. We have applied the workflow to four endpoints (acute kidney injury, mortality, length of stay and 30-day hospital readmission). The workflow can enable continuous (e.g., triggered every 6 h) and static (e.g., triggered at 24 h after admission) predictions. We also provide an open-source codebase that illustrates some key principles in EHR modeling. This protocol can be used by interdisciplinary teams with programming and clinical expertise to build deep-learning prediction models with alternate data sources and prediction tasks.

A clinically applicable approach to continuous prediction of future acute kidney injury.

The early prediction of deterioration could have an important role in supporting healthcare professionals, as an estimated 11% of deaths in hospital follow a failure to promptly recognize and treat deteriorating patients1. To achieve this goal requires predictions of patient risk that are continuously updated and accurate, and delivered at an individual level with sufficient context and enough time to act. Here we develop a deep learning approach for the continuous risk prediction of future deterioration in patients, building on recent work that models adverse events from electronic health records2-17 and using acute kidney injury-a common and potentially life-threatening condition18-as an exemplar. Our model was developed on a large, longitudinal dataset of electronic health records that cover diverse clinical environments, comprising 703,782 adult patients across 172 inpatient and 1,062 outpatient sites. Our model predicts 55.8% of all inpatient episodes of acute kidney injury, and 90.2% of all acute kidney injuries that required subsequent administration of dialysis, with a lead time of up to 48 h and a ratio of 2 false alerts for every true alert. In addition to predicting future acute kidney injury, our model provides confidence assessments and a list of the clinical features that are most salient to each prediction, alongside predicted future trajectories for clinically relevant blood tests9. Although the recognition and prompt treatment of acute kidney injury is known to be challenging, our approach may offer opportunities for identifying patients at risk within a time window that enables early treatment.

Acute Kidney Injury Risk Prediction in Patients Undergoing Coronary Angiography in a National Veterans Health Administration Cohort With External Validation.

BACKGROUND: Acute kidney injury (AKI) occurs frequently after cardiac catheterization and percutaneous coronary intervention. Although a clinical risk model exists for percutaneous coronary intervention, no models exist for both procedures, nor do existing models account for risk factors prior to the index admission. We aimed to develop such a model for use in prospective automated surveillance programs in the Veterans Health Administration. METHODS AND RESULTS: We collected data on all patients undergoing cardiac catheterization or percutaneous coronary intervention in the Veterans Health Administration from January 01, 2009 to September 30, 2013, excluding patients with chronic dialysis, end-stage renal disease, renal transplant, and missing pre- and postprocedural creatinine measurement. We used 4 AKI definitions in model development and included risk factors from up to 1 year prior to the procedure and at presentation. We developed our prediction models for postprocedural AKI using the least absolute shrinkage and selection operator (LASSO) and internally validated using bootstrapping. We developed models using 115 633 angiogram procedures and externally validated using 27 905 procedures from a New England cohort. Models had cross-validated C-statistics of 0.74 (95% CI: 0.74-0.75) for AKI, 0.83 (95% CI: 0.82-0.84) for AKIN2, 0.74 (95% CI: 0.74-0.75) for contrast-induced nephropathy, and 0.89 (95% CI: 0.87-0.90) for dialysis. CONCLUSIONS: We developed a robust, externally validated clinical prediction model for AKI following cardiac catheterization or percutaneous coronary intervention to automatically identify high-risk patients before and immediately after a procedure in the Veterans Health Administration. Work is ongoing to incorporate these models into routine clinical practice.

National Veterans Health Administration inpatient risk stratification models for hospital-acquired acute kidney injury.

OBJECTIVE: Hospital-acquired acute kidney injury (HA-AKI) is a potentially preventable cause of morbidity and mortality. Identifying high-risk patients prior to the onset of kidney injury is a key step towards AKI prevention. MATERIALS AND METHODS: A national retrospective cohort of 1,620,898 patient hospitalizations from 116 Veterans Affairs hospitals was assembled from electronic health record (EHR) data collected from 2003 to 2012. HA-AKI was defined at stage 1+, stage 2+, and dialysis. EHR-based predictors were identified through logistic regression, least absolute shrinkage and selection operator (lasso) regression, and random forests, and pair-wise comparisons between each were made. Calibration and discrimination metrics were calculated using 50 bootstrap iterations. In the final models, we report odds ratios, 95% confidence intervals, and importance rankings for predictor variables to evaluate their significance. RESULTS: The area under the receiver operating characteristic curve (AUC) for the different model outcomes ranged from 0.746 to 0.758 in stage 1+, 0.714 to 0.720 in stage 2+, and 0.823 to 0.825 in dialysis. Logistic regression had the best AUC in stage 1+ and dialysis. Random forests had the best AUC in stage 2+ but the least favorable calibration plots. Multiple risk factors were significant in our models, including some nonsteroidal anti-inflammatory drugs, blood pressure medications, antibiotics, and intravenous fluids given during the first 48 h of admission. CONCLUSIONS: This study demonstrated that, although all the models tested had good discrimination, performance characteristics varied between methods, and the random forests models did not calibrate as well as the lasso or logistic regression models. In addition, novel modifiable risk factors were explored and found to be significant.