Neural Natural Language Processing for Unstructured Data in Electronic Health Records: a Review

TL;DR

This review discusses recent advances in neural NLP methods applied to unstructured data in electronic health records, highlighting improvements over traditional approaches across various tasks like classification, extraction, and generation.

Contribution

It provides a comprehensive summary of current neural NLP techniques for EHRs, covering a wide range of tasks and recent methodological developments.

Findings

Neural NLP methods outperform traditional systems on EHR tasks

Deep learning enables better extraction and prediction from unstructured medical text

The survey covers diverse applications including question answering and phenotyping.

Abstract

Electronic health records (EHRs), digital collections of patient healthcare events and observations, are ubiquitous in medicine and critical to healthcare delivery, operations, and research. Despite this central role, EHRs are notoriously difficult to process automatically. Well over half of the information stored within EHRs is in the form of unstructured text (e.g. provider notes, operation reports) and remains largely untapped for secondary use. Recently, however, newer neural network and deep learning approaches to Natural Language Processing (NLP) have made considerable advances, outperforming traditional statistical and rule-based systems on a variety of tasks. In this survey paper, we summarize current neural NLP methods for EHR applications. We focus on a broad scope of tasks, namely, classification and prediction, word embeddings, extraction, generation, and other topics such as question answering, phenotyping, knowledge graphs, medical dialogue, multilinguality, interpretability, etc.