Multi-Task Learning with Recurrent Neural Networks for ARDS Prediction using only EHR Data: Model Development and Validation Study (Preprint)
BACKGROUND Acute Respiratory Distress Syndrome (ARDS) is a condition that is often considered to have broad and subjective diagnostic criteria and is associated with significant mortality and morbidity. Early and accurate prediction of ARDS and related conditions such as hypoxemia and sepsis could allow timely administration of therapies, leading to improved patient outcomes. OBJECTIVE To perform an exploration of how multi-label classification in the clinical setting can take advantage of the underlying dependencies between ARDS and related conditions to improve early prediction of ARDS. METHODS The electronic health record dataset included 40,073 patient encounters from 7 hospitals from 4/20/2018 to 3/17/2021. A recurrent neural network (RNN) was trained using data from 5 hospitals, and external validation was conducted on data from 2 hospitals. In addition to ARDS, 12 target labels for related conditions such as sepsis, hypoxemia and Covid-19 were used to train the model to classify a total of 13 outputs. As a comparator, XGBoost models were developed for each of the 13 target labels. Model performance was assessed using the area under the receiver operating characteristic (AUROC). Heatmaps to visualize attention scores were generated to provide interpretability to the NNs. Finally, cluster analysis was performed to identify potential phenotypic subgroups of ARDS patients. RESULTS The single RNN model trained to classify 13 outputs outperformed the XGBoost model for ARDS prediction, achieving an AUROC of 0.842 on the external test sets. Models trained on an increasing number of tasks resulted in increasing performance. Earlier diagnosis of ARDS nearly doubled the rate of in-hospital survival. Cluster analysis revealed distinct ARDS subgroups, some of which had similar mortality rates but different clinical presentations. CONCLUSIONS The RNN model presented in this paper can be used as an early warning system to stratify patients who are at risk of developing one of the multiple risk outcomes, hence providing practitioners with means to take early action.