Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
10355561 | Journal of Biomedical Informatics | 2011 | 7 Pages |
Abstract
The best performing model used ridge logistic regression and achieved a sensitivity of 100%, a specificity of 99.987% and a positive predictive value of 32% (recalibrated for a real population), obtained in a stratified cross-validation setting. These results were further validated on an independent test set. Using a method that combines ridge logistic regression with variable selection and threshold optimization, a significantly improved performance was achieved compared to the current state-of-the-art for derivatized data, while retaining more interpretability and requiring less variables. The results indicate the potential value of data mining methods as a diagnostic support tool.
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Science Applications
Authors
Tim Van den Bulcke, Paul Vanden Broucke, Viviane Van Hoof, Kristien Wouters, Seppe Vanden Broucke, Geert Smits, Elke Smits, Sam Proesmans, Toon Van Genechten, François Eyskens,