Ontology-guided feature engineering for clinical text classification.

Journal of Biomedical Informatics
Vijay Garla, C A Brandt

Abstract

In this study we present novel feature engineering techniques that leverage the biomedical domain knowledge encoded in the Unified Medical Language System (UMLS) to improve machine-learning based clinical text classification. Critical steps in clinical text classification include identification of features and passages relevant to the classification task, and representation of clinical text to enable discrimination between documents of different classes. We developed novel information-theoretic techniques that utilize the taxonomical structure of the Unified Medical Language System (UMLS) to improve feature ranking, and we developed a semantic similarity measure that projects clinical text into a feature space that improves classification. We evaluated these methods on the 2008 Integrating Informatics with Biology and the Bedside (I2B2) obesity challenge. The methods we developed improve upon the results of this challenge's top machine-learning based system, and may improve the performance of other machine-learning based clinical text classification systems. We have released all tools developed as part of this study as open source, available at http://code.google.com/p/ytex.

References

Nov 2, 2004·Neural Computation·David R HardoonJohn Shawe-Taylor
Aug 1, 2006·Journal of Biomedical Informatics·Ted PedersenChristopher G Chute
Aug 28, 2007·Bioinformatics·Yvan SaeysPedro Larrañaga
Apr 25, 2009·Journal of the American Medical Informatics Association : JAMIA·Ozlem Uzuner
Apr 25, 2009·Journal of the American Medical Informatics Association : JAMIA·Richárd FarkasRóbert Busa-Fekete
Apr 25, 2009·Journal of the American Medical Informatics Association : JAMIA·Kyle H Ambert, Aaron M Cohen
Sep 8, 2010·Journal of the American Medical Informatics Association : JAMIA·Guergana K SavovaChristopher G Chute
Apr 6, 2011·Journal of Biomedical Informatics·David Sánchez, Montserrat Batet
May 14, 2011·Journal of the American Medical Informatics Association : JAMIA·Berry de BruijnXiaodan Zhu
May 31, 2011·Journal of the American Medical Informatics Association : JAMIA·Vijay GarlaCynthia Brandt
Aug 19, 2011·Journal of the American Medical Informatics Association : JAMIA·Wendy W ChapmanOzlem Uzuner

❮ Previous
Next ❯

Citations

Oct 11, 2012·BMC Bioinformatics·Vijay N Garla, Cynthia Brandt
Mar 30, 2013·BMC Bioinformatics·Francisco M OrtuñoJean-Fred Fontaine
Oct 19, 2012·Journal of the American Medical Informatics Association : JAMIA·Vijay N Garla, Cynthia Brandt
Aug 5, 2015·Journal of Biomedical Informatics·Leyla Jael Garcia CastroAlexander Garcia
Apr 16, 2015·Journal of Biomedical Semantics·Kristina Doing-HarrisStephane Meystre
Apr 5, 2019·BMC Medical Informatics and Decision Making·Liang YaoYuan Luo
Oct 22, 2014·Journal of the American Medical Informatics Association : JAMIA·Christian M RochefortDavid L Buckeridge
Mar 14, 2018·Journal of Biomedical Informatics·Jejo D KoolaMichael E Matheny
Sep 2, 2021·Annual Review of Biomedical Data Science·Bethany Percha

❮ Previous
Next ❯

Related Concepts

Related Feeds

Antihypertensive Agents: Mechanisms of Action

Antihypertensive drugs are used to treat hypertension (high blood pressure) which aims to prevent the complications of high blood pressure, such as stroke and myocardial infarction. Discover the latest research on antihypertensive drugs and their mechanism of action here.

Cardiovascular Diseases: Risk Factors

Cardiovascular disease is a significant health concern. Risk factors include hypertension, obesity, dyslipidemia and smoking. Women who are postmenopausal are at an increased risk of heart disease. Here is the latest research for risk factors of cardiovascular disease.

Cardiovascular Disease Pathophysiology

Cardiovascular disease involves several different processes that contribute to the pathological mechanism, including hyperglycemia, inflammation, atherosclerosis, hypertension and more. Vasculature stability plays a critical role in the development of the disease. Discover the latest research on cardiovascular disease pathophysiology here.

Bioinformatics in Biomedicine

Bioinformatics in biomedicine incorporates computer science, biology, chemistry, medicine, mathematics and statistics. Discover the latest research on bioinformatics in biomedicine here.

Related Papers

Journal of the American Medical Informatics Association : JAMIA
Ozlem Uzuner
Journal of the American Medical Informatics Association : JAMIA
Cosmin Adrian BejanMeliha Yetisgen-Yildiz
© 2021 Meta ULC. All rights reserved