跳到正文
原文
Nature 健康护理主题·· 25 天前AI 评分30

儿童非佩戴检测的无监督集成方法:基于三轴加速度计数据

Unsupervised ensemble methods for non-wear detection from triaxial accelerometer data in children following continuous monitoring

AI 导读

开发并评估了用于儿童连续监测数据非佩戴检测的无监督集成算法。在包含15名儿童的斯坦福睡眠加速度计研究及模拟数据中,SUMMA方法实现67.7%敏感性、87.1%特异性及77.5%平衡准确率,优于单一基线算法且跨受试者表现更稳定,为异质性活动模式提供了替代方案。

正文

Scientific Reports (2026) Cite this article

We’re sharing this article early to provide faster access to peer-reviewed, accepted research. It is citable and carries a permanent DOI. This version is subject to further edits and will be replaced automatically by the final Version of Record. All legal disclaimers apply.

Abstract

Accelerometer signals from non-wear periods are often misclassified as stationary time or sleep, potentially biasing interpretations of physical activity’s impact on health. Non-wear detection algorithms exist but their performance varies depending on the population and activity patterns. To address this, we developed and evaluated unsupervised ensemble algorithms using data from the Stanford Sleep Accelerometer Study (15 children, aged 4–18) and a simulated dataset (100 subjects) constructed from real-world pediatric data. We compared two unsupervised ensemble methods (SUMMA and BI) and a simple majority-voting ensemble against nine base classifiers derived from three algorithms: Choi, NHANES and a Hidden Markov Model (HMM). Non-wear was treated as the positive class throughout. Performance was evaluated using sensitivity, specificity, and balanced accuracy, and the variability of error across subjects. Individual algorithms exhibited performance trade-offs: Choi performed well for long stationary and sleep periods but often missed short non-wear intervals, while HMM identified short non-wear intervals but underestimated stationary time. Ensemble methods provided comparable balanced accuracy with a better balance between sensitivity and specificity. In the simulation dataset, SUMMA achieved 67.7% sensitivity, 87.1% specificity, and 77.5% balanced accuracy, compared to the best individual baseline algorithm, NHANES_VM (63.4%, 90.3%, 77.1%). Overall, the ensemble approach showed more stable performance across subjects and provides a useful alternative to relying on a single non-wear detection algorithm in settings with heterogeneous activity patterns and short non-wear periods.

Abbreviations

ADEMP:

Aims, Data-generating mechanisms, Estimands, Methods, and Performance measures

BI:

Bayesian integrator

CI:

Confidence interval

HMM:

Hidden Markov Model

LPA:

Light physical activity

MV:

Majority voting

MVPA:

Moderate-to-vigorous physical activity

NHANES:

National health and nutrition examination survey

SUMMA:

Strategy for unsupervised multiple method aggregation

VPA:

Vigorous physical activity

VM:

Vector magnitude

Funding

Research reported in this publication was supported in parts grants from the Stanford Maternal and Child Health Research Institute, the Department of Pediatrics at Stanford University, and the Li Ka Shing Foundation Stanford-Oxford Big Data for Human Health seed grant program. The content is solely the responsibility of the authors and does not necessarily represent the official views of Stanford University or the Li Ka Shing Foundation. This manuscript is partially supported by the following NIH grant: R01LM013355; Stanford’s Center for Clinical and Translational Education and Research award, under the Biostatistics, Epidemiology and Research Design (BERD) Program: UL1TR003142; the Clinical and Translational Core of the Stanford Diabetes Research Center: P30DK116074; the Biostatistics Shared Resource (B-SR) of the NCI-sponsored Stanford Cancer Institute: P30CA124435. Data were collected with funding from the National Heart, Lung, and Blood Institute of the National Institutes of Health under award number U01HL103629. The content expressed in this paper is solely the responsibility of the authors and does not necessarily represent the official views of the National Heart, Lung, and Blood Institute; the National Institutes of Health; the US Department of Health and Human Services; and the US Government. Research was also supported by the FDA grant “Improving the integrity of novel digital health technology-derived endpoints through rigorous simulation studies of multiple imputation” (U01FD005978). The content is solely the responsibility of the authors and does not necessarily represent the official views of the Food and Drug Administration.

Author information

Authors and Affiliations

  1. Quantitative Sciences Unit, Section of Biostatistics, Department of Medicine, Stanford University, Stanford, CA, USA

    Fatma Gunturkun, Alexandria Jensen & Manisha Desai

  2. Manning College of Information and Computer Science, University of Massachusetts Amherst, Amherst, MA, USA

    Madalina Fiterau

  3. Solutions Science Lab, Department of Pediatrics, Stanford University, Stanford, CA, USA

    Hyatt E. Moore IV, K. Farish Haydel & Thomas N. Robinson

  4. Department of Systems Engineering, Naval Postgraduate School, Monterey, CA, USA

    Hyatt E. Moore IV

  5. Office of Biostatistics, Office of Translational Sciences, Center for Drug Evaluation and Research, United States Food and Drug Administration (US FDA), Silver Spring, MD, USA

    Maria Matilde Kam, Lili Garrard, Andrew Potter & Paul Schuette

  6. Department of Medicine, Stanford University, Stanford, CA, USA

    Thomas N. Robinson

  7. Quantitative Sciences Unit, Section of Biostatistics, Department of Medicine, Stanford University, 3180 Porter Drive, Stanford, 94306, CA, USA

    Manisha Desai

Authors

  1. Fatma Gunturkun
  2. Madalina Fiterau
  3. Hyatt E. Moore IV
  4. Alexandria Jensen
  5. K. Farish Haydel
  6. Maria Matilde Kam
  7. Lili Garrard
  8. Andrew Potter
  9. Paul Schuette
  10. Thomas N. Robinson
  11. Manisha Desai

Corresponding author

Correspondence to Manisha Desai.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Gunturkun, F., Fiterau, M., Moore , H.E. et al. Unsupervised ensemble methods for non-wear detection from triaxial accelerometer data in children following continuous monitoring. Sci Rep (2026). https://doi.org/10.1038/s41598-026-75825-1

Download citation

  • Received: 19 February 2026

  • Accepted: 09 October 2026

  • Published: 26 October 2026

  • DOI: https://doi.org/10.1038/s41598-026-75825-1

Keywords

来源:Nature 健康护理主题 · nature.com