Combining biomarkers to improve diagnostic accuracy using the overlap coefficient

Tahani Coolen-Maturi

Communications in Statistics - Theory and Methods, 2025, vol. 54, issue 20, 6599-6615

Abstract: Measuring the accuracy of diagnostic tests is crucial in many application areas, including medicine, machine learning, and credit scoring. In practice, multiple diagnostic tests or biomarkers are combined to improve diagnostic accuracy. The area under the receiver operating characteristic curve (AUC) is a common measure of diagnostic test performance and can be used as an objective function to maximise when combining multiple biomarkers. Another useful measure is the overlap coefficient, which quantifies the similarity between two independent distributions by their overlapping area. The smaller the overlapping area, the better the biomarker is at discrimination. The aim of this article is to combine biomarkers to improve diagnostic accuracy by minimising the overlap coefficient. We approach this parametrically and non-parametrically using Kernel-based methods. We also present a probabilistic interpretation of the overlap coefficient, which gives more insight into this measure. The proposed methods are evaluated through a simulation study and illustrated via examples.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/03610926.2025.2460095 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:lstaxx:v:54:y:2025:i:20:p:6599-6615

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/lsta20

DOI: 10.1080/03610926.2025.2460095

Access Statistics for this article

Communications in Statistics - Theory and Methods is currently edited by Debbie Iscoe

More articles in Communications in Statistics - Theory and Methods from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().