Počet záznamů: 1
A Sparse Pair-preserving Centroid-based Supervised Learning Method for High-dimensional Biomedical Data or Images
- 1.
SYSNO ASEP 0524330 Druh ASEP J - Článek v odborném periodiku Zařazení RIV J - Článek v odborném periodiku Poddruh J Článek ve WOS Název A Sparse Pair-preserving Centroid-based Supervised Learning Method for High-dimensional Biomedical Data or Images Tvůrce(i) Kalina, Jan (UIVT-O) RID, SAI, ORCID
Matonoha, Ctirad (UIVT-O) RID, SAICelkový počet autorů 2 Zdroj.dok. Biocybernetics and Biomedical Engineering. - : Elsevier - ISSN 0208-5216
Roč. 40, č. 2 (2020), s. 774-786Poč.str. 13 s. Forma vydání Tištěná - P Jazyk dok. eng - angličtina Země vyd. PL - Polsko Klíč. slova supervised learning ; high-dimensional data ; robustness ; sparsity ; nonlinear optimization Vědní obor RIV IN - Informatika Obor OECD Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8) CEP GA19-05704S GA ČR - Grantová agentura ČR Způsob publikování Omezený přístup Institucionální podpora UIVT-O - RVO:67985807 UT WOS 000547542400014 EID SCOPUS 85084491501 DOI 10.1016/j.bbe.2020.03.008 Anotace In various biomedical applications designed to compare two groups (e.g. patients and controls in matched case-control studies), it is often desirable to perform a dimensionality reduction in order to learn a classification rule over high-dimensional data. This paper considers a centroid-based classification method for paired data, which at the same time performs a supervised variable selection respecting the matched pairs design. We propose an algorithm for optimizing the centroid (prototype, template). A subsequent optimization of weights for the centroid ensures sparsity, robustness to outliers, and clear interpretation of the contribution of individual variables to the classification task. We apply the method to a simulated matched case-control study dataset, to a gene expression study of acute myocardial infarction, and to mouth localization in 2D facial images. The novel approach yields a comparable performance with standard classifiers and outperforms them if the data are contaminated by outliers. This robustness makes the method relevant for genomic, metabolomic or proteomic high-dimensional data (in matched case-control studies) or medical diagnostics based on images, as (excessive) noise and contamination are ubiquitous in biomedical measurements. Pracoviště Ústav informatiky Kontakt Tereza Šírová, sirova@cs.cas.cz, Tel.: 266 053 800 Rok sběru 2021 Elektronická adresa http://dx.doi.org/10.1016/j.bbe.2020.03.008
Počet záznamů: 1