Motif clustering and digital biomarker extraction for free-living physical activity analysis

BioData Mining, Journal Year: 2025, Volume and Issue: 18(1)

Published: Jan. 22, 2025

Abstract Background Analyzing free-living physical activity (PA) data presents challenges due to variability in daily routines and the lack of labels. Traditional approaches often rely on summary statistics, which may not capture nuances individual patterns. To address these limitations advance our understanding relationship between PA patterns health outcomes, we propose a novel motif clustering algorithm that identifies characterizes specific Methods This paper proposes an elastic distance-based for identifying (motifs) data. The segments long-term curves into short-term utilizes shape analysis measure similarity segments. enables discovery recurring motifs through pattern clustering. Then, functional principal component (FPCA) is then used extract digital biomarkers from each motif. These can subsequently be explore outcomes interest. Results We demonstrate efficacy method three real-world applications. show derived effectively association disease improving accuracy patient classification. Conclusions study introduced approach analyzing by characterizing (motifs). provide more nuanced its impact health, with potential applications personalized assessment detection, offering promising future healthcare.

Language: Английский

---

The Risk of Pediatric Overweight and Children’s Objectively Measured Sedentary Behaviors and Physical Activity by Area of Residence

Healthcare, Journal Year: 2025, Volume and Issue: 13(5), P. 462 - 462

Published: Feb. 21, 2025

Background/Objectives: Research considering objectively measured moderate-to-vigorous physical activity (MVPA) and sedentary behaviors (SB) health outcomes among urban children has produced equivocal findings. Therefore, this study was designed (i) to compare MVPA SB of non-urban (ii) analyze associations between the risk overweight by their degree urbanization. Methods: The sample comprised 389 (195 girls) aged 6 10 years. Measurements for height weight were taken, followed calculation body mass index (BMI). A motion sensor employed gather data on 7 days. Urban regions characterized as areas with a population density exceeding 500 inhabitants per square kilometer or having total greater than 50,000. ANOVAs, partial correlations, logistic regression analysis used, controlling potential confounders. Results: After wearing time sex, significantly less active (lower MVPA) peers [i.e., urban: 48 min/day vs. 51 (non-urban)]. On other hand, spent more counterparts weekend during all assessed associated being at both children. Inspection final model indicated that who engaged in behavior extended periods higher categorized overweight. Conclusions: Findings revealed association weekdays just significant location where child lives influences status, therefore, community-based programs [at schools (PE recess) after-school (organized sports)] should be developed encourage lifestyles tailored environments.

Language: Английский

---

Variability in moderate-to-vigorous physical activity assessment: comparing tri-axial accelerometry cut-off points in children

Sports Medicine and Health Science, Journal Year: 2025, Volume and Issue: unknown

Published: March 1, 2025

Language: Английский

---

Mortality prediction using data from wearable activity trackers and individual characteristics: An explainable artificial intelligence approach

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: unknown, P. 126195 - 126195

Published: Dec. 1, 2024

Language: Английский

---