Improving Clinical Outcome Predictions Using Convolution over Medical Entities with Multimodal Learning

TL;DR

This paper introduces a convolutional multimodal learning approach that integrates medical entities from clinical notes with time-series ICU data to improve early prediction of patient mortality and length of stay, outperforming baseline models.

Contribution

It presents a novel convolution-based architecture that effectively combines medical entities and ICU signals, also comparing different embedding techniques for medical entities.

Findings

Outperforms baseline models on clinical prediction tasks

Effectively combines clinical notes and time-series data

Demonstrates the impact of embedding techniques like Word2vec and FastText

Abstract

Early prediction of mortality and length of stay(LOS) of a patient is vital for saving a patient's life and management of hospital resources. Availability of electronic health records(EHR) makes a huge impact on the healthcare domain and there has seen several works on predicting clinical problems. However, many studies did not benefit from the clinical notes because of the sparse, and high dimensional nature. In this work, we extract medical entities from clinical notes and use them as additional features besides time-series features to improve our predictions. We propose a convolution based multimodal architecture, which not only learns effectively combining medical entities and time-series ICU signals of patients, but also allows us to compare the effect of different embedding techniques such as Word2vec, FastText on medical entities. In the experiments, our proposed method robustly outperforms all other baseline models including different multimodal architectures for all clinical tasks. The code for the proposed method is available at https://github.com/tanlab/ConvolutionMedicalNer.