Neural Document Embeddings for Intensive Care Patient Mortality Prediction

Abstract

We present an automatic mortality prediction scheme based on the unstructured textual content of clinical notes. Proposing a convolutional document embedding approach, our empirical investigation using the MIMIC-III intensive care database shows significant performance gains compared to previously employed methods such as latent topic distributions or generic doc2vec embeddings. These improvements are especially pronounced for the difficult problem of post-discharge mortality prediction.

References in corpus (2)

• Learning to Diagnose with LSTM Recurrent Neural Networks
• Recurrent Neural Networks for Multivariate Time Series with Missing Values

Cited by in corpus (14)

• ISeeU: Visually interpretable deep learning for mortality prediction inside the ICU
• Patient representation learning and interpretable evaluation using clinical notes
• MeDAL: Medical Abbreviation Disambiguation Dataset for Natural Language Understanding Pretraining
• Combining LSTM and Latent Topic Modeling for Mortality Prediction
• Evaluating the Impact of Social Determinants on Health Prediction in the Intensive Care Unit
• Embedding Electronic Health Records for Clinical Information Retrieval
• Machine Learning and Visualization in Clinical Decision Support: Current State and Future Directions
• DC3 -- A Diagnostic Case Challenge Collection for Clinical Decision Support
• Learning Conceptual-Contextual Embeddings for Medical Text
• Fast Prototyping a Dialogue Comprehension System for Nurse-Patient Conversations on Symptom Monitoring
• Multimodal Ensemble Approach to Incorporate Various Types of Clinical Notes for Predicting Readmission
• Unsupervised patient representations from clinical notes with interpretable classification decisions
• Interpretable Structure-aware Document Encoders with Hierarchical Attention
• Hierarchical Semantic Correspondence Learning for Post-Discharge Patient Mortality Prediction