Risk prediction on electronic health records with prior medical knowledge

Fenglong Ma, Quanzeng You, Jing Gao, Jing Zhou, Qiuling Suo, Aidong Zhang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

11 Citations (Scopus)

Abstract

Predicting the risk of potential diseases from Electronic Health Records (EHR) has attracted considerable attention in recent years, especially with the development of deep learning techniques. Compared with traditional machine learning models, deep learning based approaches achieve superior performance on risk prediction task. However, none of existing work explicitly takes prior medical knowledge (such as the relationships between diseases and corresponding risk factors) into account. In medical domain, knowledge is usually represented by discrete and arbitrary rules. Thus, how to integrate such medical rules into existing risk prediction models to improve the performance is a challenge. To tackle this challenge, we propose a novel and general framework called PRIME for risk prediction task, which can successfully incorporate discrete prior medical knowledge into all of the state-of-the-art predictive models using posterior regularization technique. Different from traditional posterior regularization, we do not need to manually set a bound for each piece of prior medical knowledge when modeling desired distribution of the target disease on patients. Moreover, the proposed PRIME can automatically learn the importance of different prior knowledge with a log-linear model. Experimental results on three real medical datasets demonstrate the effectiveness of the proposed framework for the task of risk prediction1

Original languageEnglish (US)
Title of host publicationKDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
PublisherAssociation for Computing Machinery
Pages1910-1919
Number of pages10
ISBN (Print)9781450355520
DOIs
StatePublished - Jul 19 2018
Event24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018 - London, United Kingdom
Duration: Aug 19 2018Aug 23 2018

Publication series

NameProceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Other

Other24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018
CountryUnited Kingdom
CityLondon
Period8/19/188/23/18

Fingerprint

Health
Learning systems
Deep learning

All Science Journal Classification (ASJC) codes

  • Software
  • Information Systems

Cite this

Ma, F., You, Q., Gao, J., Zhou, J., Suo, Q., & Zhang, A. (2018). Risk prediction on electronic health records with prior medical knowledge. In KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1910-1919). (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). Association for Computing Machinery. https://doi.org/10.1145/3219819.3220020
Ma, Fenglong ; You, Quanzeng ; Gao, Jing ; Zhou, Jing ; Suo, Qiuling ; Zhang, Aidong. / Risk prediction on electronic health records with prior medical knowledge. KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2018. pp. 1910-1919 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).
@inproceedings{43749ee200c2488cb279253d584aa644,
title = "Risk prediction on electronic health records with prior medical knowledge",
abstract = "Predicting the risk of potential diseases from Electronic Health Records (EHR) has attracted considerable attention in recent years, especially with the development of deep learning techniques. Compared with traditional machine learning models, deep learning based approaches achieve superior performance on risk prediction task. However, none of existing work explicitly takes prior medical knowledge (such as the relationships between diseases and corresponding risk factors) into account. In medical domain, knowledge is usually represented by discrete and arbitrary rules. Thus, how to integrate such medical rules into existing risk prediction models to improve the performance is a challenge. To tackle this challenge, we propose a novel and general framework called PRIME for risk prediction task, which can successfully incorporate discrete prior medical knowledge into all of the state-of-the-art predictive models using posterior regularization technique. Different from traditional posterior regularization, we do not need to manually set a bound for each piece of prior medical knowledge when modeling desired distribution of the target disease on patients. Moreover, the proposed PRIME can automatically learn the importance of different prior knowledge with a log-linear model. Experimental results on three real medical datasets demonstrate the effectiveness of the proposed framework for the task of risk prediction1",
author = "Fenglong Ma and Quanzeng You and Jing Gao and Jing Zhou and Qiuling Suo and Aidong Zhang",
year = "2018",
month = "7",
day = "19",
doi = "10.1145/3219819.3220020",
language = "English (US)",
isbn = "9781450355520",
series = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",
publisher = "Association for Computing Machinery",
pages = "1910--1919",
booktitle = "KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

}

Ma, F, You, Q, Gao, J, Zhou, J, Suo, Q & Zhang, A 2018, Risk prediction on electronic health records with prior medical knowledge. in KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Association for Computing Machinery, pp. 1910-1919, 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018, London, United Kingdom, 8/19/18. https://doi.org/10.1145/3219819.3220020

Risk prediction on electronic health records with prior medical knowledge. / Ma, Fenglong; You, Quanzeng; Gao, Jing; Zhou, Jing; Suo, Qiuling; Zhang, Aidong.

KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2018. p. 1910-1919 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Risk prediction on electronic health records with prior medical knowledge

AU - Ma, Fenglong

AU - You, Quanzeng

AU - Gao, Jing

AU - Zhou, Jing

AU - Suo, Qiuling

AU - Zhang, Aidong

PY - 2018/7/19

Y1 - 2018/7/19

N2 - Predicting the risk of potential diseases from Electronic Health Records (EHR) has attracted considerable attention in recent years, especially with the development of deep learning techniques. Compared with traditional machine learning models, deep learning based approaches achieve superior performance on risk prediction task. However, none of existing work explicitly takes prior medical knowledge (such as the relationships between diseases and corresponding risk factors) into account. In medical domain, knowledge is usually represented by discrete and arbitrary rules. Thus, how to integrate such medical rules into existing risk prediction models to improve the performance is a challenge. To tackle this challenge, we propose a novel and general framework called PRIME for risk prediction task, which can successfully incorporate discrete prior medical knowledge into all of the state-of-the-art predictive models using posterior regularization technique. Different from traditional posterior regularization, we do not need to manually set a bound for each piece of prior medical knowledge when modeling desired distribution of the target disease on patients. Moreover, the proposed PRIME can automatically learn the importance of different prior knowledge with a log-linear model. Experimental results on three real medical datasets demonstrate the effectiveness of the proposed framework for the task of risk prediction1

AB - Predicting the risk of potential diseases from Electronic Health Records (EHR) has attracted considerable attention in recent years, especially with the development of deep learning techniques. Compared with traditional machine learning models, deep learning based approaches achieve superior performance on risk prediction task. However, none of existing work explicitly takes prior medical knowledge (such as the relationships between diseases and corresponding risk factors) into account. In medical domain, knowledge is usually represented by discrete and arbitrary rules. Thus, how to integrate such medical rules into existing risk prediction models to improve the performance is a challenge. To tackle this challenge, we propose a novel and general framework called PRIME for risk prediction task, which can successfully incorporate discrete prior medical knowledge into all of the state-of-the-art predictive models using posterior regularization technique. Different from traditional posterior regularization, we do not need to manually set a bound for each piece of prior medical knowledge when modeling desired distribution of the target disease on patients. Moreover, the proposed PRIME can automatically learn the importance of different prior knowledge with a log-linear model. Experimental results on three real medical datasets demonstrate the effectiveness of the proposed framework for the task of risk prediction1

UR - http://www.scopus.com/inward/record.url?scp=85051564349&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051564349&partnerID=8YFLogxK

U2 - 10.1145/3219819.3220020

DO - 10.1145/3219819.3220020

M3 - Conference contribution

AN - SCOPUS:85051564349

SN - 9781450355520

T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

SP - 1910

EP - 1919

BT - KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

PB - Association for Computing Machinery

ER -

Ma F, You Q, Gao J, Zhou J, Suo Q, Zhang A. Risk prediction on electronic health records with prior medical knowledge. In KDD 2018 - Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery. 2018. p. 1910-1919. (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). https://doi.org/10.1145/3219819.3220020