EUVIMEDEUVIMED entdecken
Crossref

Clustering Hybrid Functional and Vector Data

Sekha D Daluwatumulle iD, Jeong Hoon Jang iD, Ying Cui iD, Amita K Manatunga

Biometrical Journal · 2026

DOI: 10.1002/bimj.70184 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biometrical Journal, 2026. Verfasst von Sekha D Daluwatumulle und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biostatistics and Bioinformatics, Emory University, Atlanta, Georgia, USA. Hauptthemen laut MEDLINE: Biometry. Die Arbeit nennt 32 Literaturangaben. Gefördert von: NIDDK NIH HHS, NHLBI NIH HHS, National Heart, Lung, and Blood Institute.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Modern biomedical studies increasingly collect hybrid data, consisting of functional observations (e.g., curves or images) and multiple scalar (vector‐valued) clinical variables. Integrating these modalities in an unsupervised learning framework can improve patient subgroup identification and interpretation, yet most existing clustering approaches handle only a single data type. We introduce a clustering framework for hybrid data combining multivariate functional and vector outcomes. The method first applies hybrid principal component analysis to capture joint mean structures and modes of variation across modalities, then iteratively reassigns subjects using a hybrid distance metric based on cluster‐specific Karhunen–Loéve expansions. This approach accommodates multiple functions of varying dimensions, integrates contributions from functional and vector components in mean and covariance structures, and avoids parametric distributional assumptions. Simulation studies show that the proposed method outperforms competing approaches when subgroup information, in terms of mean and/or variance differences, is either distributed across modalities or concentrated in a single modality. We illustrate the method using renal radionuclide imaging data, including baseline and diuretic renogram curves and pharmacokinetic variables from patients with suspected kidney obstruction. The method identifies clinically meaningful subgroups consistent with expert assessment, demonstrating its potential to support data‐driven decision‐making in multimodal biomedical studies.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

BiometryClustering AlgorithmsHumansCluster AnalysisPrincipal Component Analysis

Stichwörter der Autor:innen

clusteringfunctional datahybrid dataprincipal component analysisunsupervised learning

Autor:innen und Institutionen

  • Department of Biostatistics and Bioinformatics, Emory University, Atlanta, Georgia, USA.
  • Department of Biostatistics and Data Science, University of Texas Medical Branch, Galveston, Texas, USA.
  • Department of Biomedical Data Science, Stanford University, Stanford, California, USA.

ORCID-Profile: Sekha D Daluwatumulle, Jeong Hoon Jang, Ying Cui

Förderung

  • NIDDK NIH HHS
  • NHLBI NIH HHS
  • National Heart, Lung, and Blood Institute
  • National Institute of Diabetes and Digestive and Kidney Diseases

Zitieren

APA

Daluwatumulle, S. D., Jang, J. H., Cui, Y., & Manatunga, A. K. (2026). Clustering Hybrid Functional and Vector Data. Biometrical Journal. https://doi.org/10.1002/bimj.70184

Vancouver

Daluwatumulle SD, Jang JH, Cui Y, Manatunga AK. Clustering Hybrid Functional and Vector Data. Biometrical Journal. 2026. doi:10.1002/bimj.70184

BibTeX
@article{Daluwatumulle2026,
  author = {Daluwatumulle, Sekha D and Jang, Jeong Hoon and Cui, Ying and Manatunga, Amita K},
  title = {Clustering Hybrid Functional and Vector Data},
  journal = {Biometrical Journal},
  year = {2026},
  doi = {10.1002/bimj.70184}
}
Kostenlos starten →