Interpretable time-aware and co-occurrence-aware network for medical prediction

Abstract

Background Disease prediction based on electronic health records (EHRs) is essential for personalized healthcare. But it’s hard due to the special data structure and the interpretability requirement of methods. The structure of EHR is hierarchical: each patient has a sequence of admissions, and each admission has some co-occurrence diagnoses. However, the existing methods only partially model these characteristics and lack the interpretation for non-specialists. Methods This work proposes a time-aware and co-occurrence-aware deep learning network (TCoN), which is not only suitable for EHR data structure but also interpretable: the co-occurrence-aware self-attention (CS-attention) mechanism and time-aware gated recurrent unit (T-GRU) can model multilevel relations; the interpretation path and the diagnosis graph can make the result interpretable. Results The method is tested on a real-world dataset for mortality prediction, readmission prediction, disease prediction, and next diagnoses prediction. Experimental results show that TCoN is better than baselines with 2.01% higher accuracy. Meanwhile, the method can give the interpretation of causal relationships and the diagnosis graph of each patient. Conclusions This work proposes a novel model—TCoN. It is an interpretable and effective deep learning method, that can model the hierarchical medical structure and predict medical events. The experiments show that it outperforms all state-of-the-art methods. Future work can apply the graph embedding technology based on more knowledge data such as doctor notes.

TOO-BERT: A Trajectory Order Objective BERT for self-supervised representation learning of temporal healthcare data

“…Similarly, interventions often mitigate the severity of conditions at the cost of raising other health risks. Thus, every medical event, whether it involves diseases or medications, can serve as a cause, complication, or early symptom of the recorded codes [42]. This study will demonstrate that order objectives, besides the context, enhance the model performance by learning more structural information.…”

A multi-model architecture based on deep learning for aircraft load prediction

“…Unfortunately, the common opinion holds that deep learning models are blackboxes 20 . Through our previous investigation [21][22][23] , we found that an expert's understanding of a method is usually based on their knowledge, i.e., which input features are important to the result, and what is the contribution of each feature to the result. Based on this, we achieved the interpretation method by finding the key flight parameters and their contributions to load prediction, which is a preliminary exploration to explain the deep learning method in aviation engineering.…”