Developing a portable natural language processing based phenotyping system

Himanshu Sharma, Chengsheng Mao, Yizhen Zhang, Haleh Vatani, Liang Yao, Yizhen Zhong, Luke Rasmussen, Guoqian D Jiang, Jyotishman Pathak, Yuan Luo

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Background: This paper presents a portable phenotyping system that is capable of integrating both rule-based and statistical machine learning based approaches. Methods: Our system utilizes UMLS to extract clinically relevant features from the unstructured text and then facilitates portability across different institutions and data systems by incorporating OHDSI's OMOP Common Data Model (CDM) to standardize necessary data elements. Our system can also store the key components of rule-based systems (e.g., regular expression matches) in the format of OMOP CDM, thus enabling the reuse, adaptation and extension of many existing rule-based clinical NLP systems. We experimented with our system on the corpus from i2b2's Obesity Challenge as a pilot study. Results: Our system facilitates portable phenotyping of obesity and its 15 comorbidities based on the unstructured patient discharge summaries, while achieving a performance that often ranked among the top 10 of the challenge participants. Conclusion: Our system of standardization enables a consistent application of numerous rule-based and machine learning based classification techniques downstream across disparate datasets which may originate across different institutions and data systems.

Original languageEnglish (US)
Article number78
JournalBMC Medical Informatics and Decision Making
Volume19
DOIs
StatePublished - Apr 4 2019

Fingerprint

Natural Language Processing
Information Systems
Patient Discharge Summaries
Obesity
Unified Medical Language System
Comorbidity
Machine Learning
Datasets

ASJC Scopus subject areas

  • Health Policy
  • Health Informatics

Cite this

Developing a portable natural language processing based phenotyping system. / Sharma, Himanshu; Mao, Chengsheng; Zhang, Yizhen; Vatani, Haleh; Yao, Liang; Zhong, Yizhen; Rasmussen, Luke; Jiang, Guoqian D; Pathak, Jyotishman; Luo, Yuan.

In: BMC Medical Informatics and Decision Making, Vol. 19, 78, 04.04.2019.

Research output: Contribution to journalArticle

Sharma, H, Mao, C, Zhang, Y, Vatani, H, Yao, L, Zhong, Y, Rasmussen, L, Jiang, GD, Pathak, J & Luo, Y 2019, 'Developing a portable natural language processing based phenotyping system', BMC Medical Informatics and Decision Making, vol. 19, 78. https://doi.org/10.1186/s12911-019-0786-z
Sharma, Himanshu ; Mao, Chengsheng ; Zhang, Yizhen ; Vatani, Haleh ; Yao, Liang ; Zhong, Yizhen ; Rasmussen, Luke ; Jiang, Guoqian D ; Pathak, Jyotishman ; Luo, Yuan. / Developing a portable natural language processing based phenotyping system. In: BMC Medical Informatics and Decision Making. 2019 ; Vol. 19.
@article{89a8e83d67a9407db23be04568a6af86,
title = "Developing a portable natural language processing based phenotyping system",
abstract = "Background: This paper presents a portable phenotyping system that is capable of integrating both rule-based and statistical machine learning based approaches. Methods: Our system utilizes UMLS to extract clinically relevant features from the unstructured text and then facilitates portability across different institutions and data systems by incorporating OHDSI's OMOP Common Data Model (CDM) to standardize necessary data elements. Our system can also store the key components of rule-based systems (e.g., regular expression matches) in the format of OMOP CDM, thus enabling the reuse, adaptation and extension of many existing rule-based clinical NLP systems. We experimented with our system on the corpus from i2b2's Obesity Challenge as a pilot study. Results: Our system facilitates portable phenotyping of obesity and its 15 comorbidities based on the unstructured patient discharge summaries, while achieving a performance that often ranked among the top 10 of the challenge participants. Conclusion: Our system of standardization enables a consistent application of numerous rule-based and machine learning based classification techniques downstream across disparate datasets which may originate across different institutions and data systems.",
author = "Himanshu Sharma and Chengsheng Mao and Yizhen Zhang and Haleh Vatani and Liang Yao and Yizhen Zhong and Luke Rasmussen and Jiang, {Guoqian D} and Jyotishman Pathak and Yuan Luo",
year = "2019",
month = "4",
day = "4",
doi = "10.1186/s12911-019-0786-z",
language = "English (US)",
volume = "19",
journal = "BMC Medical Informatics and Decision Making",
issn = "1472-6947",
publisher = "BioMed Central",

}

TY - JOUR

T1 - Developing a portable natural language processing based phenotyping system

AU - Sharma, Himanshu

AU - Mao, Chengsheng

AU - Zhang, Yizhen

AU - Vatani, Haleh

AU - Yao, Liang

AU - Zhong, Yizhen

AU - Rasmussen, Luke

AU - Jiang, Guoqian D

AU - Pathak, Jyotishman

AU - Luo, Yuan

PY - 2019/4/4

Y1 - 2019/4/4

N2 - Background: This paper presents a portable phenotyping system that is capable of integrating both rule-based and statistical machine learning based approaches. Methods: Our system utilizes UMLS to extract clinically relevant features from the unstructured text and then facilitates portability across different institutions and data systems by incorporating OHDSI's OMOP Common Data Model (CDM) to standardize necessary data elements. Our system can also store the key components of rule-based systems (e.g., regular expression matches) in the format of OMOP CDM, thus enabling the reuse, adaptation and extension of many existing rule-based clinical NLP systems. We experimented with our system on the corpus from i2b2's Obesity Challenge as a pilot study. Results: Our system facilitates portable phenotyping of obesity and its 15 comorbidities based on the unstructured patient discharge summaries, while achieving a performance that often ranked among the top 10 of the challenge participants. Conclusion: Our system of standardization enables a consistent application of numerous rule-based and machine learning based classification techniques downstream across disparate datasets which may originate across different institutions and data systems.

AB - Background: This paper presents a portable phenotyping system that is capable of integrating both rule-based and statistical machine learning based approaches. Methods: Our system utilizes UMLS to extract clinically relevant features from the unstructured text and then facilitates portability across different institutions and data systems by incorporating OHDSI's OMOP Common Data Model (CDM) to standardize necessary data elements. Our system can also store the key components of rule-based systems (e.g., regular expression matches) in the format of OMOP CDM, thus enabling the reuse, adaptation and extension of many existing rule-based clinical NLP systems. We experimented with our system on the corpus from i2b2's Obesity Challenge as a pilot study. Results: Our system facilitates portable phenotyping of obesity and its 15 comorbidities based on the unstructured patient discharge summaries, while achieving a performance that often ranked among the top 10 of the challenge participants. Conclusion: Our system of standardization enables a consistent application of numerous rule-based and machine learning based classification techniques downstream across disparate datasets which may originate across different institutions and data systems.

UR - http://www.scopus.com/inward/record.url?scp=85063972802&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063972802&partnerID=8YFLogxK

U2 - 10.1186/s12911-019-0786-z

DO - 10.1186/s12911-019-0786-z

M3 - Article

VL - 19

JO - BMC Medical Informatics and Decision Making

JF - BMC Medical Informatics and Decision Making

SN - 1472-6947

M1 - 78

ER -