EUVIMEDEUVIMED entdecken
Open Access

Characterizing daily physical activity patterns with unsupervised learning via functional mixture models

Ipek Ensari iD, Billy A Caceres, Kasey B Jackman, Jeff Goldsmith, Niurka M Suero-Tejeda, Michelle L Odlum, Suzanne Bakken

Journal of Behavioral Medicine · 2025

DOI: 10.1007/s10865-024-00519-w ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Behavioral Medicine, 2025. Verfasst von Ipek Ensari und 6 weiteren Autor:innen. Beteiligte Einrichtung: Windreich Department of Artificial Intelligence and Human Health, Icahn School of Medicine at Mount Sinai, Hasso Plattner Institute for Digital Healt…. Hauptthemen laut MEDLINE: Exercise, Unsupervised Machine Learning. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 61 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIA NIH HHS, NHLBI NIH HHS, NINR NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Physical inactivity is a significant public health concern. Consideration of inter-individual variations in physical activity (PA) trends can provide additional information about the groups under study to aid intervention design. This study aims to identify latent profiles ("phenotypes") based on daily PA trends among adults living in. This was a secondary analysis of 724 person-level days of accelerometry data from 133 urban-dwelling adults (89% Latinx, age = 19-77 years). We used Actigraph accelerometers and the Actilife software to collect and process 24-hour PA data. We implemented a probabilistic clustering technique based on functional mixture models. Multiple days of data per person were averaged for entry into the models. We evaluated step counts, moderate-intensity PA (MOD), total activity and sedentary minutes as potential model variables. Bayesian Information Criterion (BIC) index was used to select the model that provided the best fit for the data. A 4-cluster resolution provided the best fit for the data (i.e., BIC=-3257, improvements of Δ = 13 and Δ = 7 from 3- and 5-cluster models, respectively). MOD provided the greatest between-cluster discrimination. Phenotype 1 (N = 61) was characterized by a morning peak in PA that declined until bedtime. Later bedtimes and the highest daily PA volume were distinct for phenotype 2 (N = 18), along with a similar peak pattern. Phenotype 3 (N = 29) membership was associated with the lowest PA levels throughout the day. Phenotype 4 was characterized by a more evenly distributed PA during the day, and later waking/bedtimes. Our findings point to distinct, interpretable PA phenotypes based on temporal patterns. Functional clustering of PA data could provide additional actionable points for tailoring behavioral interventions.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ExerciseUnsupervised Machine LearningHumansFemaleMaleMiddle AgedAdultAccelerometryAgedBayes TheoremSedentary BehaviorYoung AdultCluster AnalysisPhenotype

Stichwörter der Autor:innen

AccelerometryClusteringFunctional dataPhysical activityUnsupervised learning

Autor:innen und Institutionen

  • Windreich Department of Artificial Intelligence and Human Health, Icahn School of Medicine at Mount Sinai, Hasso Plattner Institute for Digital Health at Mount Sinai, One Gustave L. Levy Place, Annenberg 11-02A, New Yor…
  • Columbia University Data Science Institute, New York, NY, 10025, USA. ipek.ensari@mssm.edu.
  • Columbia University Data Science Institute, New York, NY, 10025, USA.
  • Columbia University School of Nursing, New York, NY, 10032, USA.
  • NewYork-Presbyterian Hospital, New York, NY, 10032, USA.
  • Columbia University Mailman School of Public Health, New York, NY, 10032, USA.

ORCID-Profile: Ipek Ensari

Förderung

  • NIA NIH HHS
  • NHLBI NIH HHS
  • NINR NIH HHS
  • National Institute of Nursing Research
  • National Heart, Lung, and Blood Institute

Zitieren

APA

Ensari, I., Caceres, B. A., Jackman, K. B., Goldsmith, J., Suero-Tejeda, N. M., Odlum, M. L., & Bakken, S. (2025). Characterizing daily physical activity patterns with unsupervised learning via functional mixture models. Journal of Behavioral Medicine. https://doi.org/10.1007/s10865-024-00519-w

Vancouver

Ensari I, Caceres BA, Jackman KB, Goldsmith J, Suero-Tejeda NM, Odlum ML, et al. Characterizing daily physical activity patterns with unsupervised learning via functional mixture models. Journal of Behavioral Medicine. 2025. doi:10.1007/s10865-024-00519-w

BibTeX
@article{Ensari2025,
  author = {Ensari, Ipek and Caceres, Billy A and Jackman, Kasey B and Goldsmith, Jeff and Suero-Tejeda, Niurka M and Odlum, Michelle L and Bakken, Suzanne},
  title = {Characterizing daily physical activity patterns with unsupervised learning via functional mixture models},
  journal = {Journal of Behavioral Medicine},
  year = {2025},
  doi = {10.1007/s10865-024-00519-w}
}
Kostenlos starten →