深度可解释ICU死亡率预测模型

Deep Interpretable Mortality Model for Intensive Care Unit Risk Prediction
ADMA 2019 : The 15th International Conference on Advanced Data Mining and Applications
Nov 21, 2019 - Nov 23, 2019.
Dalian, China
Zhenkun Shi, Weitong Chen, Shining Liang, Wanli Zuo*, Lin Yue, Sen Wang.

Abstract

Estimating the mortality of patients plays a fundamental role in an intensive care unit (ICU). Currently, most learning approaches are based on deep learning models. However, these approaches in mor- tality prediction suffer from two problems: (i) the specificity of causes of death are not considered in the learning process due to the differ- ent diseases, and symptoms are mixed-used without diversification and localization; (ii) the learning outcome for the mortality prediction is not self-explainable for the clinicians. In this paper, we propose a Deep Inter- pretable Mortality Model (DIMM), which employs Multi-Source Embed- ding, Gated Recurrent Units (GRU), Attention mechanism and Focal Loss techniques to prognosticate mortality prediction. We intensified the mortality prediction by considering the different clinical measures, med- ical treatments and the heterogeneity of the disease. More importantly, for the first time, in this framework, we use a separate evidence-based interpreter named Highlighter to interpret the prediction model, which makes the prediction understandable and trustworthy to clinicians. We demonstrate that our approach achieves state-of-the-art performance in mortality prediction and can get an interpretable prediction on four dif- ferent diseases.

Attachment

Deep Interpretable Mortality Model for Intensive Care Unit Risk Prediction