HotCSE Seminar
Computational Science & Engineering
Wednesday September 14, 12pm-1pm, 1116-E Klaus

RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanism

Edward Choi
Advisor: Prof. Jimeng Sun


Accuracy and interpretation are two goals of any successful predictive models. Most existing works have to suffer the tradeoff between the two by either picking complex black box models such as recurrent neural networks (RNN) or relying on less accurate traditional models with better interpretation such as logistic regression. To address this dilemma, we present REverse Time AttentIoN model (RETAIN) for analyzing EHR data that achieves high accuracy while remaining clinically interpretable. RETAIN is a two-level neural attention model that can find influential past visits and significant clinical variables within those visits (e.g,. key diagnoses). RETAIN mimics physician practice by attending the EHR data in a reverse time order so that more recent clinical visits will likely get higher attention. Experiments on a large real EHR dataset of 14 million visits from 263K patients over 8 years confirmed the comparable predictive accuracy and computational scalability to the state-of-the-art methods such as RNN. Finally, we demonstrate the clinical interpretation with concrete examples from RETAIN.


Edward Choi is a Ph.D. student in Computer Science, working with Prof. Jimeng Sun. Edward has a BS in CS from Seoul National University, and an MS in CS from KAIST, where he specialized in natural language processing. He has spent 4 years as member of the text mining team at ETRI (Electronics & Telecommunications Research Institute). Edward’s current interests are in applying representation learning methods in healthcare problems.