AZpharm metaalert: A meta-learning framework for pharmacovigilance

Xiao Liu, Hsinchun Chen

Research output: ResearchConference contribution

Abstract

Pharmacovigilance is the research related to the detection, assessment, understanding, and prevention of adverse drug events. Despite the research efforts in pharmacovigilance in recent year, current approaches are insufficient in detecting adverse drug reaction (ADR) signals timely across different datasets. In this study, we develop an integrated and high-performance AZ Pharm Meta-Alert framework for efficient and accurate post-approval pharmacovigilance. Our approach extracts adverse drug events from patient social media, electronic health records, and FDA’s Adverse Event Reporting System (FAERS) and integrates ADR signals with stacking and bagging methods. Experiment results show that our approach achieves 71% in precision, 90% in recall, and 80% in f-measure for ADR signal detection and significantly outperforms the traditional signal detection methods.

LanguageEnglish (US)
Title of host publicationSmart Health - International Conference, ICSH 2016, Revised Selected Papers
PublisherSpringer Verlag
Pages147-154
Number of pages8
Volume10219 LNCS
ISBN (Print)9783319598574
StatePublished - 2017
EventInternational Conference for Smart Health, ICSH 2016 - Haikou, China
Duration: Dec 24 2016Dec 25 2016

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10219 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherInternational Conference for Smart Health, ICSH 2016
CountryChina
CityHaikou
Period12/24/1612/25/16

Fingerprint

Meta-learning
Drugs
Framework
Signal detection
Health
Experiments
Signal Detection
Bagging
Social Media
Stacking
High Performance
Integrate
Electronics
Experiment

Keywords

  • Adverse drug event
  • Deep-learning
  • Drug safety surveillance
  • Meta-learning
  • Pharmacovigilance

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Liu, X., & Chen, H. (2017). AZpharm metaalert: A meta-learning framework for pharmacovigilance. In Smart Health - International Conference, ICSH 2016, Revised Selected Papers (Vol. 10219 LNCS, pp. 147-154). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10219 LNCS). Springer Verlag.

AZpharm metaalert : A meta-learning framework for pharmacovigilance. / Liu, Xiao; Chen, Hsinchun.

Smart Health - International Conference, ICSH 2016, Revised Selected Papers. Vol. 10219 LNCS Springer Verlag, 2017. p. 147-154 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10219 LNCS).

Research output: ResearchConference contribution

Liu, X & Chen, H 2017, AZpharm metaalert: A meta-learning framework for pharmacovigilance. in Smart Health - International Conference, ICSH 2016, Revised Selected Papers. vol. 10219 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10219 LNCS, Springer Verlag, pp. 147-154, International Conference for Smart Health, ICSH 2016, Haikou, China, 12/24/16.
Liu X, Chen H. AZpharm metaalert: A meta-learning framework for pharmacovigilance. In Smart Health - International Conference, ICSH 2016, Revised Selected Papers. Vol. 10219 LNCS. Springer Verlag. 2017. p. 147-154. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Liu, Xiao ; Chen, Hsinchun. / AZpharm metaalert : A meta-learning framework for pharmacovigilance. Smart Health - International Conference, ICSH 2016, Revised Selected Papers. Vol. 10219 LNCS Springer Verlag, 2017. pp. 147-154 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inbook{f54d50d76b1b477e82a9511810dfb8db,
title = "AZpharm metaalert: A meta-learning framework for pharmacovigilance",
abstract = "Pharmacovigilance is the research related to the detection, assessment, understanding, and prevention of adverse drug events. Despite the research efforts in pharmacovigilance in recent year, current approaches are insufficient in detecting adverse drug reaction (ADR) signals timely across different datasets. In this study, we develop an integrated and high-performance AZ Pharm Meta-Alert framework for efficient and accurate post-approval pharmacovigilance. Our approach extracts adverse drug events from patient social media, electronic health records, and FDA’s Adverse Event Reporting System (FAERS) and integrates ADR signals with stacking and bagging methods. Experiment results show that our approach achieves 71% in precision, 90% in recall, and 80% in f-measure for ADR signal detection and significantly outperforms the traditional signal detection methods.",
keywords = "Adverse drug event, Deep-learning, Drug safety surveillance, Meta-learning, Pharmacovigilance",
author = "Xiao Liu and Hsinchun Chen",
year = "2017",
isbn = "9783319598574",
volume = "10219 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "147--154",
booktitle = "Smart Health - International Conference, ICSH 2016, Revised Selected Papers",
address = "Germany",

}

TY - CHAP

T1 - AZpharm metaalert

T2 - A meta-learning framework for pharmacovigilance

AU - Liu,Xiao

AU - Chen,Hsinchun

PY - 2017

Y1 - 2017

N2 - Pharmacovigilance is the research related to the detection, assessment, understanding, and prevention of adverse drug events. Despite the research efforts in pharmacovigilance in recent year, current approaches are insufficient in detecting adverse drug reaction (ADR) signals timely across different datasets. In this study, we develop an integrated and high-performance AZ Pharm Meta-Alert framework for efficient and accurate post-approval pharmacovigilance. Our approach extracts adverse drug events from patient social media, electronic health records, and FDA’s Adverse Event Reporting System (FAERS) and integrates ADR signals with stacking and bagging methods. Experiment results show that our approach achieves 71% in precision, 90% in recall, and 80% in f-measure for ADR signal detection and significantly outperforms the traditional signal detection methods.

AB - Pharmacovigilance is the research related to the detection, assessment, understanding, and prevention of adverse drug events. Despite the research efforts in pharmacovigilance in recent year, current approaches are insufficient in detecting adverse drug reaction (ADR) signals timely across different datasets. In this study, we develop an integrated and high-performance AZ Pharm Meta-Alert framework for efficient and accurate post-approval pharmacovigilance. Our approach extracts adverse drug events from patient social media, electronic health records, and FDA’s Adverse Event Reporting System (FAERS) and integrates ADR signals with stacking and bagging methods. Experiment results show that our approach achieves 71% in precision, 90% in recall, and 80% in f-measure for ADR signal detection and significantly outperforms the traditional signal detection methods.

KW - Adverse drug event

KW - Deep-learning

KW - Drug safety surveillance

KW - Meta-learning

KW - Pharmacovigilance

UR - http://www.scopus.com/inward/record.url?scp=85020453183&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85020453183&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9783319598574

VL - 10219 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 147

EP - 154

BT - Smart Health - International Conference, ICSH 2016, Revised Selected Papers

PB - Springer Verlag

ER -