Predicting Changes in Pediatric Medical Complexity using Large Longitudinal Health Records.

Medically complex patients consume a disproportionate amount of care resources in hospitals but still often end up with sub-optimal clinical outcomes. Predicting dynamics of complexity in such patients can potentially help improve the quality of care and reduce utilization of hospital resources. In this work, we model the change prediction of medical complexity using a large dataset of 226K pediatric patients over 5 years from Children's Healthcare of Atlanta (CHOA). We compare different classification methods including logistic regression, random forest, gradient boosting trees, and multilayer perceptron in predicting whether patients will change their complexity status in the last year based on the data from previous years. We achieved an area under the ROC curve (AUC) of 88% for predicting noncomplex patients becoming complex and 74% for predicting complex patients staying complex. We also identify the factors associated with the change in complexity of patients.

GRAM: Graph-based Attention Model for Healthcare Representation Learning.

Deep learning methods exhibit promising performance for predictive modeling in healthcare, but two important challenges remain: Data insufficiency: Often in healthcare predictive modeling, the sample size is insufficient for deep learning methods to achieve satisfactory results.Interpretation: The representations learned by deep learning methods should align with medical knowledge. To address these challenges, we propose GRaph-based Attention Model (GRAM) that supplements electronic health records (EHR) with hierarchical information inherent to medical ontologies. Based on the data volume and the ontology structure, GRAM represents a medical concept as a combination of its ancestors in the ontology via an attention mechanism. We compared predictive performance (i.e. accuracy, data needs, interpretability) of GRAM to various methods including the recurrent neural network (RNN) in two sequential diagnoses prediction tasks and one heart failure prediction task. Compared to the basic RNN, GRAM achieved 10% higher accuracy for predicting diseases rarely observed in the training data and 3% improved area under the ROC curve for predicting heart failure using an order of magnitude less training data. Additionally, unlike other methods, the medical concept representations learned by GRAM are well aligned with the medical ontology. Finally, GRAM exhibits intuitive attention behaviors by adaptively generalizing to higher level concepts when facing data insufficiency at the lower level concepts.

Doctor AI: Predicting Clinical Events via Recurrent Neural Networks.

Leveraging large historical data in electronic health record (EHR), we developed Doctor AI, a generic predictive model that covers observed medical conditions and medication uses. Doctor AI is a temporal model using recurrent neural networks (RNN) and was developed and applied to longitudinal time stamped EHR data from 260K patients over 8 years. Encounter records (e.g. diagnosis codes, medication codes or procedure codes) were input to RNN to predict (all) the diagnosis and medication categories for a subsequent visit. Doctor AI assesses the history of patients to make multilabel predictions (one label for each diagnosis or medication category). Based on separate blind test set evaluation, Doctor AI can perform differential diagnosis with up to 79% recall@30, significantly higher than several baselines. Moreover, we demonstrate great generalizability of Doctor AI by adapting the resulting models from one institution to another without losing substantial accuracy.

Causal Phenotype Discovery via Deep Networks.

The rapid growth of digital health databases has attracted many researchers interested in using modern computational methods to discover and model patterns of health and illness in a research program known as computational phenotyping. Much of the work in this area has focused on traditional statistical learning paradigms, such as classification, prediction, clustering, pattern mining. In this paper, we propose a related but different paradigm called causal phenotype discovery, which aims to discover latent representations of illness that are causally predictive. We illustrate this idea with a two-stage framework that combines the latent representation learning power of deep neural networks with state-of-the-art tools from causal inference. We apply this framework to two large ICU time series data sets and show that it can learn features that are predictively useful, that capture complex physiologic patterns associated with critical illnesses, and that are potentially more clinically meaningful than manually designed features.