A Deep Learning Model for Pediatric Patient Risk Stratification

Artificial intelligence based on medical claims data outperforms traditional models in stratifying patient risk. ABSTRACT Objectives: Current models for patient risk prediction rely on practitioner expertise and domain knowledge. This study presents a deep learning model--a type of machine learning that does not require human inputs--to analyze complex clinical and financial data for population risk stratification. Methods: "Skip-Gram," an unsupervised deep learning approach that uses neural networks for prediction modeling, used data from 2014 and 2015 to predict the risk of hospitalization in 2016. The area under the curve (AUC) of the deep learning model was compared with that of both the Clinical Classifications Software and the commercial DxCG Intelligence predictive risk models, each with and without demographic and utilization features.