Missing data imputation for individualized CVD diagnostic and treatment

Sitalakshmi Venkatraman, Andrew Yatsko, Andrew Stranieri, Herbert Jelinek

Research output: Book chapter/Published conference paperConference paperpeer-review

Abstract

Cardiac health screening standards require increasingly more clinical tests consisting of blood, urine and anthropometric measures as well as an extensive clinical and medication history. To ensure optimal screening referrals, diagnostic determinants need to be highly accurate to reduce false positives and ensuing stress to individual patients. However, the data from individual patients partaking in population screening is often incomplete. The current study provides an imputation algorithm that has been applied to patient centered cardiac health screening. Missing values are iteratively imputed in conjunction with combinations of values on subsets of selected features. The approach was evaluated on the DiabHealth dataset containing 2800 records with over 180 attributes. The results for predicting CVD after data completion showed sensitivity and specificity of 94% and 99% respectively. Removing variables that define cardiac events and associated conditions directly, left ‘age’ followed by ‘use’ of antihypertensive and anti-cholesterol medication, especially statins among the best predictors.
Original languageEnglish
Title of host publicationComputing in Cardiology 2016
Subtitle of host publicationVolume 43
Place of PublicationUnited States
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages349
Number of pages352
Volume43
ISBN (Electronic)9781509008957
ISBN (Print)9781509008964
Publication statusPublished - 2016
Event43rd Computing in Cardiology Conference, CinC 2016 - Marriot Hotel and Simon Fraser University, Vancouver, Canada
Duration: 11 Sep 201614 Sep 2016
https://web.archive.org/web/20170603121743/http://cinc2016.org (Conference website)
http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=7861973 (Conference proceedings)

Publication series

Name
ISSN (Print)2325-887X

Conference

Conference43rd Computing in Cardiology Conference, CinC 2016
Country/TerritoryCanada
CityVancouver
Period11/09/1614/09/16
OtherComputing in Cardiology provides an international forum for scientists and professionals from the fields of medicine, physics, engineering and computer science, and has been held annually since 1974.
Internet address

Fingerprint

Dive into the research topics of 'Missing data imputation for individualized CVD diagnostic and treatment'. Together they form a unique fingerprint.

Cite this