Improving Hospital Mortality Prediction with Medical Named Entities and Multimodal Learning

TL;DR

This paper introduces a multimodal neural network that combines structured data and extracted clinical text entities to improve hospital mortality prediction, achieving a 2% higher AUC than existing benchmarks.

Contribution

The study presents a novel approach integrating named entity extraction from clinical notes with multimodal learning to enhance mortality prediction accuracy.

Findings

Model outperforms benchmark by 2% AUC

Combining clinical text and time series improves prediction

Named entity features contribute significantly to model performance

Abstract

Clinical text provides essential information to estimate the acuity of a patient during hospital stays in addition to structured clinical data. In this study, we explore how clinical text can complement a clinical predictive learning task. We leverage an internal medical natural language processing service to perform named entity extraction and negation detection on clinical notes and compose selected entities into a new text corpus to train document representations. We then propose a multimodal neural network to jointly train time series signals and unstructured clinical text representations to predict the in-hospital mortality risk for ICU patients. Our model outperforms the benchmark by 2% AUC.