Impact of novel aggregation methods for flexible, time-sensitive EHR prediction without variable selection or cleaning
This addresses the need for automated, interpretable risk analysis in healthcare by eliminating variable selection and cleaning, potentially easing integration with existing EHR systems.
The paper tackled the problem of dynamic patient status assessment in the ICU by developing improved aggregation methods for a flexible deep learning architecture that processes all EHR events without pre-processing, achieving an AUROC of 0.87 for mortality classification at 48 hours on the MIMIC-III dataset.
Dynamic assessment of patient status (e.g. by an automated, continuously updated assessment of outcome) in the Intensive Care Unit (ICU) is of paramount importance for early alerting, decision support and resource allocation. Extraction and cleaning of expert-selected clinical variables discards information and protracts collaborative efforts to introduce machine learning in medicine. We present improved aggregation methods for a flexible deep learning architecture which learns a joint representation of patient chart, lab and output events. Our models outperform recent deep learning models for patient mortality classification using ICU timeseries, by embedding and aggregating all events with no pre-processing or variable selection. Our model achieves a strong performance of AUROC 0.87 at 48 hours on the MIMIC-III dataset while using 13,233 unique un-preprocessed variables in an interpretable manner via hourly softmax aggregation. This demonstrates how our method can be easily combined with existing electronic health record systems for automated, dynamic patient risk analysis.