Deep EHR: A Survey of Recent Advances in Deep Learning Techniques for Electronic Health Record (EHR) Analysis

By Benjamin Shickel, Patrick James Tighe, Azra Bihorac, Parisa Rashidi. 2017

1,486 citations recorded by the source corpus.

The past decade has seen an explosion in the amount of digital information stored in electronic health records (EHRs). While primarily designed for archiving patient information and performing administrative healthcare tasks like billing, many researchers have found secondary use of these records for various clinical informatics applications. Over the same period, the machine learning community has seen widespread advances in the field of deep learning. In this review, we survey the current research on applying deep learning to clinical tasks based on EHR data, where we find a variety of deep learning techniques and frameworks being applied to several types of clinical applications including information extraction, representation learning, outcome prediction, phenotyping, and deidentification. We identify several limitations of current research involving topics such as model interpretability, data heterogeneity, and lack of universal benchmarks. We conclude by summarizing the state of the field and identifying avenues of future deep EHR research.

Subfields: Artificial Intelligence, Health Information Management, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Health Information Management

In Molecular Biology

Citations in the corpus

Split learning for health: Distributed deep learning without sharing raw patient data (2018, background): “Deep learning methods in general have found a pervasive suite of applications in biology, clinical medicine, genomics and public health as surveyed in [24, 23, 25, 29, 30, 31].”
An overview of deep learning in medical imaging focusing on MRI (2018, background): “seizures [17] and cardiac arrests [18], to computer-aided detection [19] and diagnosis [20] supporting clinical decision making and survival analysis [21], to drug discovery [22] and as an aid in therapy selection and pharmacogenomics [23], to increased operational efficiency [24], stratified care d”
Scalable and accurate deep learning with electronic health records (2018, background): “Given rapid developments in this field, we point readers to a recent, comprehensive review [88].”
Artificial Intelligence and Surgical Decision-Making (2019, background): “61 Automatic feature extraction also promotes discovery of novel patterns and phenotypes that may have been overlooked by handcrafted feature selection techniques.”
A guide to deep learning in healthcare (2019, methodology): “The next generation of automatic speech recognition 32 and information extraction models will likely develop clinical voice assistants to accurately transcribe patient visits.”
The ethics of AI in health care: A mapping review. (2020, background): “…2019; Zacher and Czogiel, 2019); (5) prediction of adverse effects (Ding et al., 2019; Mortazavi et al., 2017); (6) prediction of outcome and/or likelihood of survival (Dong et al., 2019; Popkes et al., 2019; Topuz et al., 2018); and (7) analysing electronic health records (Shickel et al., 2018).”
Federated Learning for Internet of Things: A Comprehensive Survey (2021, neutral): “, intelligent imaging for disease detection [107].”
ClinicalBERT: Modeling Clinical Notes and Predicting Hospital Readmission (2019, background): “Machine learning techniques leverage structured features in ehr data, such as lab results and electrocardiography measurements, to uncover patterns and improve predictions (Shickel et al., 2018; Xiao et al., 2018a; Yu et al., 2018).”
Differential Privacy Techniques for Cyber Physical Systems: A Survey (2018, background): “Over the past decade, the trend of hospitals adopting electronic way of storing patient records has increased dramatically [272].”
Security and Privacy Issues in Deep Learning (2018, background): “Advances of deep learning (DL) algorithms have transformed the solution of data-driven problems in various applications in real life, including the use of large amounts of patient data for health prediction services [1]; autonomous security audits from system logs [2]; and unmanned car driving power”
BEHRT: Transformer for Electronic Health Records (2019, background): “A key contributing factor to this success, in addition to the developments in DL algorithms, was the massive influx of large multimodal biomedical data, including but not limited to, mega cohorts such as UK Biobank [5], and routinely-collected health data such as electronic health records (EHR) [6].”