A Multi-Task Framework for Monitoring Health Conditions via Attention-based Recurrent Neural Networks

Qiuling Suo, Fenglong Ma, Giovanni Canino, Jing Gao, Aidong Zhang, Pierangelo Veltri, Gnasso Agostino

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

Monitoring the future health status of patients from the historical Electronic Health Record (EHR) is a core research topic in predictive healthcare. The most important challenges are to model the temporality of sequential EHR data and to interpret the prediction results. In order to reduce the future risk of diseases, we propose a multi-task framework that can monitor the multiple status ofdiagnoses. Patients' historical records are directly fed into a Recurrent Neural Network (RNN) which memorizes all the past visit information, and then a task-specific layer is trained to predict multiple diagnoses. Moreover, three attention mechanisms for RNNs are introduced to measure the relationships between past visits and current status. Experimental results show that the proposed attention-based RNNs can significantly improve the prediction accuracy compared to widely used approaches. With the attention mechanisms, the proposed framework is able to identify the visit information which is important to the final prediction.

Original languageEnglish (US)
Pages (from-to)1665-1674
Number of pages10
JournalAMIA ... Annual Symposium proceedings. AMIA Symposium
Volume2017
StatePublished - Jan 1 2017

Fingerprint

Electronic Health Records
Health
Health Status
Delivery of Health Care
Research

All Science Journal Classification (ASJC) codes

  • Medicine(all)

Cite this

Suo, Qiuling ; Ma, Fenglong ; Canino, Giovanni ; Gao, Jing ; Zhang, Aidong ; Veltri, Pierangelo ; Agostino, Gnasso. / A Multi-Task Framework for Monitoring Health Conditions via Attention-based Recurrent Neural Networks. In: AMIA ... Annual Symposium proceedings. AMIA Symposium. 2017 ; Vol. 2017. pp. 1665-1674.
@article{4cf71996cb66419e97abbd228c7d7439,
title = "A Multi-Task Framework for Monitoring Health Conditions via Attention-based Recurrent Neural Networks",
abstract = "Monitoring the future health status of patients from the historical Electronic Health Record (EHR) is a core research topic in predictive healthcare. The most important challenges are to model the temporality of sequential EHR data and to interpret the prediction results. In order to reduce the future risk of diseases, we propose a multi-task framework that can monitor the multiple status ofdiagnoses. Patients' historical records are directly fed into a Recurrent Neural Network (RNN) which memorizes all the past visit information, and then a task-specific layer is trained to predict multiple diagnoses. Moreover, three attention mechanisms for RNNs are introduced to measure the relationships between past visits and current status. Experimental results show that the proposed attention-based RNNs can significantly improve the prediction accuracy compared to widely used approaches. With the attention mechanisms, the proposed framework is able to identify the visit information which is important to the final prediction.",
author = "Qiuling Suo and Fenglong Ma and Giovanni Canino and Jing Gao and Aidong Zhang and Pierangelo Veltri and Gnasso Agostino",
year = "2017",
month = "1",
day = "1",
language = "English (US)",
volume = "2017",
pages = "1665--1674",
journal = "AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium",
issn = "1559-4076",
publisher = "American Medical Informatics Association",

}

A Multi-Task Framework for Monitoring Health Conditions via Attention-based Recurrent Neural Networks. / Suo, Qiuling; Ma, Fenglong; Canino, Giovanni; Gao, Jing; Zhang, Aidong; Veltri, Pierangelo; Agostino, Gnasso.

In: AMIA ... Annual Symposium proceedings. AMIA Symposium, Vol. 2017, 01.01.2017, p. 1665-1674.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Multi-Task Framework for Monitoring Health Conditions via Attention-based Recurrent Neural Networks

AU - Suo, Qiuling

AU - Ma, Fenglong

AU - Canino, Giovanni

AU - Gao, Jing

AU - Zhang, Aidong

AU - Veltri, Pierangelo

AU - Agostino, Gnasso

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Monitoring the future health status of patients from the historical Electronic Health Record (EHR) is a core research topic in predictive healthcare. The most important challenges are to model the temporality of sequential EHR data and to interpret the prediction results. In order to reduce the future risk of diseases, we propose a multi-task framework that can monitor the multiple status ofdiagnoses. Patients' historical records are directly fed into a Recurrent Neural Network (RNN) which memorizes all the past visit information, and then a task-specific layer is trained to predict multiple diagnoses. Moreover, three attention mechanisms for RNNs are introduced to measure the relationships between past visits and current status. Experimental results show that the proposed attention-based RNNs can significantly improve the prediction accuracy compared to widely used approaches. With the attention mechanisms, the proposed framework is able to identify the visit information which is important to the final prediction.

AB - Monitoring the future health status of patients from the historical Electronic Health Record (EHR) is a core research topic in predictive healthcare. The most important challenges are to model the temporality of sequential EHR data and to interpret the prediction results. In order to reduce the future risk of diseases, we propose a multi-task framework that can monitor the multiple status ofdiagnoses. Patients' historical records are directly fed into a Recurrent Neural Network (RNN) which memorizes all the past visit information, and then a task-specific layer is trained to predict multiple diagnoses. Moreover, three attention mechanisms for RNNs are introduced to measure the relationships between past visits and current status. Experimental results show that the proposed attention-based RNNs can significantly improve the prediction accuracy compared to widely used approaches. With the attention mechanisms, the proposed framework is able to identify the visit information which is important to the final prediction.

UR - http://www.scopus.com/inward/record.url?scp=85051841819&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051841819&partnerID=8YFLogxK

M3 - Article

C2 - 29854237

AN - SCOPUS:85051841819

VL - 2017

SP - 1665

EP - 1674

JO - AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium

JF - AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium

SN - 1559-4076

ER -