Scalable and accurate deep learning with electronic health records

Масштабируемое и точное глубокое обучение с использованием электронных медицинских карт
Quoc Viet Le, Greg S. Corrado, Atul Janardhan Butte, Yi Zhang, Nigam Haresh Shah, Alvin Rajkomar, Greg S. Corrado, Alvin Rajkomar, Eyal Oren, Kai Chen, Andrew M. Dai, Nissan Hajaj, Michaela Hardt, Peter J. Liu, Xiaobing Liu, Jake Marcus, Mimi Sun, Patrik Sundberg, Hector Yee, Kun Zhang, Gerardo Flores, Gavin E. Duggan, Jamie Irvine, Quoc Le, Kurt Litsch, Alexander Mossin, Justin Tansuwan, De Wang, James Wexler, Jimbo Wilson, Dana Ludwig, Samuel L. Volchenboum, Katherine Chou, Michael Pearson, Srinivasan Madabushi, Nigam H. Shah, Atul J. Butte, Michael D. Howell, Claire Cui, Jeffrey Dean
2018-05-04

FHIR formatclinical event predictiondeep learningelectronic health recordspersonalized medicine
Predictive modeling with electronic health record (EHR) data is anticipated to drive personalized medicine and improve healthcare quality. Constructing predictive statistical models typically requires extraction of curated predictor variables from normalized EHR data, a labor-intensive process that discards the vast majority of information in each patient's record. We propose a representation of patients' entire raw EHR records based on the Fast Healthcare Interoperability Resources (FHIR) format. We demonstrate that deep learning methods using this representation are capable of accurately predicting multiple medical events from multiple centers without site-specific data harmonization. We validated our approach using de-identified EHR data from two US academic medical centers with 216,221 adult patients hospitalized for at least 24 h. In the sequential format we propose, this volume of EHR data unrolled into a total of 46,864,534,945 data points, including clinical notes. Deep learning models achieved high accuracy for tasks such as predicting: in-hospital mortality (area under the receiver operator curve [AUROC] across sites 0.93-0.94), 30-day unplanned readmission (AUROC 0.75-0.76), prolonged length of stay (AUROC 0.85-0.86), and all of a patient's final discharge diagnoses (frequency-weighted AUROC 0.90). These models outperformed traditional, clinically-used predictive models in all cases. We believe that this approach can be used to create accurate and scalable predictions for a variety of clinical scenarios. In a case study of a particular prediction, we demonstrate that neural networks can be used to identify relevant information from the patient's chart.
1
A FHIR-based sequential representation preserves patients’ entire raw EHRs, including clinical notes, avoiding labor-intensive predictor curation and information loss.
2
A case study showed that neural networks can identify clinically relevant information within patients’ raw charts, supporting scalable prediction across diverse clinical scenarios.
3
Deep learning models using raw EHR representations accurately predicted multiple medical outcomes across two centers without site-specific data harmonization.
4
The models predicted all final discharge diagnoses with a frequency-weighted AUROC of 0.90 and outperformed traditional clinically used predictive models across every evaluated task.
5
Using data from 216,221 hospitalized adults and 46,864,534,945 sequential data points, models achieved AUROCs of 0.93–0.94 for in-hospital mortality, 0.75–0.76 for 30-day unplanned readmission, and 0.85–0.86 for prolonged length of stay.

Entire raw electronic health record (EHR) data of hospitalized adult patients represented in FHIR sequential format

Accurate, scalable, and cross-center prediction of multiple medical events from sequential EHR data, including mortality, readmission, prolonged length of stay, and discharge diagnoses

Publication Details
Publication Date
2018-05-04
Journal
Publisher
ISSN
Cited by
3209
Access Type
Author Information
Authors
Quoc Viet Le
Greg S. Corrado
Atul Janardhan Butte
Yi Zhang
Nigam Haresh Shah
Alvin Rajkomar
Greg S. Corrado
Alvin Rajkomar
Eyal Oren
Kai Chen
Andrew M. Dai
Nissan Hajaj
Michaela Hardt
Peter J. Liu
Xiaobing Liu
Jake Marcus
Mimi Sun
Patrik Sundberg
Hector Yee
Kun Zhang
Gerardo Flores
Gavin E. Duggan
Jamie Irvine
Quoc Le
Kurt Litsch
Alexander Mossin
Justin Tansuwan
De Wang
James Wexler
Jimbo Wilson
Dana Ludwig
Samuel L. Volchenboum
Katherine Chou
Michael Pearson
Srinivasan Madabushi
Nigam H. Shah
Atul J. Butte
Michael D. Howell
Claire Cui
Jeffrey Dean
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%