Description
Clinical outcome prediction based on Electronic Health Record (EHR) helps enable early interventions for high-risk
patients, and is thus a central task for smart healthcare. Conventional deep sequential models fail to capture the rich
temporal patterns encoded in the long and irregular clinical event sequences in EHR. We make the observation that
clinical events at a long time scale exhibit strong temporal patterns, while events within a short time period tend to be
disordered co-occurrence. We thus propose differentiated mechanisms to model clinical events at different time scales.
Our model learns hierarchical representations of event sequences, to adaptively distinguish between short-range and
long-range events, and accurately capture their core temporal dependencies. Experimental results on real clinical
data show that our model greatly improves over previous state-of-the-art models, achieving AUC scores of 0.94 and
0.90 for predicting death and ICU admission, respectively. Our model also successfully identifies important events for
different clinical outcome prediction tasks.
Learning Objective: 1. Understand the current data-driven technics for clinical outcome prediction
2. Learn challenges and possible solutions in maintaining and processing data elements for electronic health record systems
3. Learn a deep learning based model for learning hierarchical representation of long irregular event sequences. This model can deal with temporal irregularity and capture the long-term dependencies of clinical events.
4. Discovering some useful clinical insights, such as important clinical events for accurate prediction.
Authors:
Luchen Liu (Presenter)
Peking University
Haoran Li, Peking University
Zhiting Hu, Carnegie Mellon University
Haoran Shi, Peking University
Zichang Wang, Peking University
Jian Tang, Quebec AI Institute
Ming Zhang, Peking University
Presentation Materials: