Reference : Predicting Dichotomised Outcomes from High-Dimensional Data in Biomedicine

	Document type :	Scientific journals : Article
	Discipline(s) :	Life sciences : Multidisciplinary, general & others Engineering, computing & technology : Computer science Human health sciences : Multidisciplinary, general & others
	Focus Areas :	Systems Biomedicine; Computational Sciences
	To cite this reference:	http://hdl.handle.net/10993/55464

Title :	Predicting Dichotomised Outcomes from High-Dimensional Data in Biomedicine
Language :	English
Author, co-author :	Rauschenberger, Armin [University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > Biomedical Data Science >]
	Glaab, Enrico [University of Luxembourg > Luxembourg Centre for Systems Biomedicine (LCSB) > Biomedical Data Science >]
Publication date :	In press
Journal title :	Journal of Applied Statistics
Publisher :	Routledge
Peer reviewed :	Yes
Audience :	International
ISSN :	0266-4763
e-ISSN :	1360-0532
Country :	United Kingdom
Keywords :	[en] high-dimensional data ; logistic regression ; linear regression ; binary classification ; systems biomedicine ; numerical prediction ; dichotomisation ; binarization ; ridge regression ; lasso regression ; diagnosis ; prognosis
Abstract :	[en] In many biomedical applications, we are more interested in the predicted probability that a numerical outcome is above a threshold than in the predicted value of the outcome. For example, it might be known that antibody levels above a certain threshold provide immunity against a disease, or a threshold for a disease severity score might reflect conversion from the presymptomatic to the symptomatic disease stage. Accordingly, biomedical researchers often convert numerical to binary outcomes (loss of information) to conduct logistic regression (probabilistic interpretation). We address this bad statistical practice by modelling the binary outcome with logistic regression, modelling the numerical outcome with linear regression, transforming the predicted values from linear regression to predicted probabilities, and combining the predicted probabilities from logistic and linear regression. Analysing high-dimensional simulated and experimental data, namely clinical data for predicting cognitive impairment, we obtain significantly improved predictions of dichotomised outcomes. Thus, the proposed approach effectively combines binary with numerical outcomes to improve binary classification in high-dimensional settings. An implementation is available in the R package cornet on GitHub (https://github.com/rauschenberger/cornet) and CRAN (https://cran.r-project.org/package=cornet).
Research centres :	Luxembourg Centre for Systems Biomedicine (LCSB): Biomedical Data Science (Glaab Group)
Funders :	Fonds National de la Recherche - FnR
Target :	Researchers
Permalink :	http://hdl.handle.net/10993/55464
DOI :	10.1080/02664763.2023.2233057
Other URL :	https://doi.org/10.1080/02664763.2023.2233057
Mentions required by the publisher for OA :	The original publication is available at: https://doi.org/10.1080/02664763.2023.2233057
FnR project :	FnR ; FNR14599012 > Enrico Glaab > DIGIPD > Validating Digital Biomarkers For Better Personalized Treatment Of Parkinson’s Disease > 01/05/2021 > 30/04/2024 > 2020

	File(s) associated to this reference
	Fulltext file(s): File Commentary Version Size Access Open access Predicting dichotomised outcomes from high dimensional data in biomedicine.pdf Publisher postprint 2.04 MB View/Open

All documents in ORBi^lu are protected by a user license.

 

University of Luxembourg Library |  Feedback |  Legal notices Site Map