IBM Research System at MADE 2018: Detecting Adverse Drug Events from Electronic Health Records

Bharath Dandala, Venkata Joopudi, Murthy Devarakonda
Proceedings of the 1st International Workshop on Medication and Adverse Drug Event Detection, PMLR 90:39-47, 2018.

Abstract

Adverse Drug Events (ADEs) are common and occur in approximately 2-5% of hospitalized adult patients. Each ADE is estimated to increase healthcare cost by more than $3,200. Severe ADEs rank among the top 5 or 6 leading causes of death in the United States. Prevention, early detection and mitigation of ADEs could save both lives and dollars. Employing Natural Language Processing (NLP) techniques on Electronic Health Records (EHRs) provides an effective way of real-time pharmacovigilance and drug safety surveillance. Thus, in this research, we developed a system for three different NLP tasks namely: Named Entity Recognition (NER), Relation Identification and Integrated task (integrative system to conduct NER and relation identification together). Our system achieved F-1 measures of 0.829 for Named Entity Recognition, 0.840 for Relation Identification and 0.617 for Integrated task. Our system ranked 1st in the integrated task and 2nd in both entity extraction and relation identification tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v90-dandala18a, title = {IBM Research System at MADE 2018: Detecting Adverse Drug Events from Electronic Health Records}, author = {Dandala, Bharath and Joopudi, Venkata and Devarakonda, Murthy}, booktitle = {Proceedings of the 1st International Workshop on Medication and Adverse Drug Event Detection}, pages = {39--47}, year = {2018}, editor = {Liu, Feifan and Jagannatha, Abhyuday and Yu, Hong}, volume = {90}, series = {Proceedings of Machine Learning Research}, month = {04 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v90/dandala18a/dandala18a.pdf}, url = {https://proceedings.mlr.press/v90/dandala18a.html}, abstract = {Adverse Drug Events (ADEs) are common and occur in approximately 2-5% of hospitalized adult patients. Each ADE is estimated to increase healthcare cost by more than $3,200. Severe ADEs rank among the top 5 or 6 leading causes of death in the United States. Prevention, early detection and mitigation of ADEs could save both lives and dollars. Employing Natural Language Processing (NLP) techniques on Electronic Health Records (EHRs) provides an effective way of real-time pharmacovigilance and drug safety surveillance. Thus, in this research, we developed a system for three different NLP tasks namely: Named Entity Recognition (NER), Relation Identification and Integrated task (integrative system to conduct NER and relation identification together). Our system achieved F-1 measures of 0.829 for Named Entity Recognition, 0.840 for Relation Identification and 0.617 for Integrated task. Our system ranked 1st in the integrated task and 2nd in both entity extraction and relation identification tasks.} }
Endnote
%0 Conference Paper %T IBM Research System at MADE 2018: Detecting Adverse Drug Events from Electronic Health Records %A Bharath Dandala %A Venkata Joopudi %A Murthy Devarakonda %B Proceedings of the 1st International Workshop on Medication and Adverse Drug Event Detection %C Proceedings of Machine Learning Research %D 2018 %E Feifan Liu %E Abhyuday Jagannatha %E Hong Yu %F pmlr-v90-dandala18a %I PMLR %P 39--47 %U https://proceedings.mlr.press/v90/dandala18a.html %V 90 %X Adverse Drug Events (ADEs) are common and occur in approximately 2-5% of hospitalized adult patients. Each ADE is estimated to increase healthcare cost by more than $3,200. Severe ADEs rank among the top 5 or 6 leading causes of death in the United States. Prevention, early detection and mitigation of ADEs could save both lives and dollars. Employing Natural Language Processing (NLP) techniques on Electronic Health Records (EHRs) provides an effective way of real-time pharmacovigilance and drug safety surveillance. Thus, in this research, we developed a system for three different NLP tasks namely: Named Entity Recognition (NER), Relation Identification and Integrated task (integrative system to conduct NER and relation identification together). Our system achieved F-1 measures of 0.829 for Named Entity Recognition, 0.840 for Relation Identification and 0.617 for Integrated task. Our system ranked 1st in the integrated task and 2nd in both entity extraction and relation identification tasks.
APA
Dandala, B., Joopudi, V. & Devarakonda, M.. (2018). IBM Research System at MADE 2018: Detecting Adverse Drug Events from Electronic Health Records. Proceedings of the 1st International Workshop on Medication and Adverse Drug Event Detection, in Proceedings of Machine Learning Research 90:39-47 Available from https://proceedings.mlr.press/v90/dandala18a.html.

Related Material