Revett, Kenneth (2008) Diagnostic feature analysis of a dobutamine stress echocardiography dataset using rough sets. In: Transactions on rough sets IX. Lecture notes in computer science (5390). Springer, Berlin, Heidelberg, pp. 318-327. ISBN 9783540898757
Full text not available from this repository.
Official URL: http://dx.doi.org/10.1007/978-3-540-89876-4_17
Stress echocardiography is an important functional diagnostic and prognostic tool that is now routinely applied to evaluate the risk of cardiovascular artery disease (CAD). In patients who are unable to safely undergo a stress based test, dobutamine is administered which provides a similar effect to stress on the cardiovascular system. In this work, a complete dataset containing data on 558 subjects undergoing a prospective longitudinal study is employed to investigate what diagnostic features correlate with the final outcome. The dataset was examined using rough sets, which produced a series of decision rules that predicts which features influence the outcomes measured clinically and recorded in the dataset. The results indicate that the ECG attribute was the most informative diagnostic feature. In addition, prehistory information has a significant impact on the classification accuracy.
|Item Type:||Book Section|
|Research Community:||University of Westminster > Electronics and Computer Science, School of|
|Deposited On:||06 Jan 2010 12:09|
|Last Modified:||06 Jan 2010 12:09|
Repository Staff Only: item control page