Performance Evaluation of a Machine Learning-based Methodology Using Dynamical Features to Detect Nonwear Intervals in Actigraphy Data in a Free-living Setting.

Details

• Alternative Title:
  Sleep Health, 2025, v. 11, no. 2: Performance Evaluation of a Machine Learning-based Methodology Using Dynamical Features to Detect Nonwear Intervals in Actigraphy Data in a Free-living Setting.
• Journal Article:
  Sleep Health
• Personal Author:
  Das, Jyotirmoy Nirupam ; Ji, Linying ; Shen, Yuqi ; Kumara, Soundar ; Buxton, Orfeu M. ; Chow, Sy-Miin
• Description:
  One challenge using wearable sensors is nonwear time. Without a nonwear (e.g., capacitive) sensor, actigraphy data quality can be biased by subjective determinations confounding sleep/wake classification. We developed and evaluated a machine learning algorithm supplemented by dynamic features to discern wear/nonwear episodes.
  
  Actigraphy data from wrist actigraph (Spectrum, Philips-Respironics).
  
  The built-in nonwear sensor as "ground truth" to classify nonwear periods using other data, mimicking features of Actiwatch 2.
  
  Data were collected over 1week from employed adults (n = 853).
  
  Extreme gradient boosting (XGBoost), a tree-based classifier algorithm, was used to classify wear/nonwear, supplemented by dynamic features calculated over various time windows.
  
  The performance of the proposed algorithm was tested over 30-second epochs. Additional analytics and exploratory analyses: Evaluation of the SHapley Additive exPlanations (SHAP) values to find the effectiveness of the dynamic features.
  
  The XGBoost classifier yielded substantial improvements in balanced accuracy, sensitivity, and specificity, including dynamic features and comparison to default actiwatch classification algorithms.
  
  The proposed classifier effectively distinguished between valid and invalid days, and the duration of contiguous periods of nonwear correctly identified.
  
  Our findings highlight the potential of XGBoost using dynamic features of varying activity levels across the time series to provide insights on wear/nonwear classification using a large dataset. The methodology provides an alternative to laborious manual benchmarking of the data for similar devices that do not have a nonwear sensor.
• Content Notes:
  Author manuscript
• Subjects:
  • Humans
  • Actigraphy
  • Machine Learning
  • Adult
  • Female
  • Male
  • Algorithms
  • Wearable Electronic Devices
  • Middle Aged
  • Sleep
• Keywords:
  • non-wear detection
  • machine learning
  • actigraphy
  • sleep
• Source:
  Sleep Health, 2025, v. 11, no. 2
• DOI:
  https://doi.org/10.1016/j.sleh.2024.10.003
• ISSN:
  2352-7218 ; 2352-7226
• Pubmed ID:
  39788836
• Pubmed Central ID:
  PMC12712857
• Document Type:
  Text
• Funding:
  U01 HD051256/HD/NICHD NIH HHSUnited States/ ; R01 DK134863/DK/NIDDK NIH HHSUnited States/ ; UL1 TR002014/TR/NCATS NIH HHSUnited States/ ; U01 OH008788/OH/NIOSH CDC HHSUnited States/ ; U01 AG027669/AG/NIA NIH HHSUnited States/ ; OT2 HL161847/HL/NHLBI NIH HHSUnited States/ ; R43 AG056250/AG/NIA NIH HHSUnited States/ ; U01 HD051217/HD/NICHD NIH HHSUnited States/ ; U24 AA027684/AA/NIAAA NIH HHSUnited States/ ; R44 AG056250/AG/NIA NIH HHSUnited States/ ; R01 HL107240/HL/NHLBI NIH HHSUnited States/ ; U01 HD059773/HD/NICHD NIH HHSUnited States/ ; U01 HD051276/HD/NICHD NIH HHSUnited States/ ; U01 HD051218/HD/NICHD NIH HHSUnited States/
• Genre:
  Journal Article
• Pages in Document:
  166-173 (22 pdf pages)
• Volume:
  11
• Issue:
  2
• Download URL:
  https://stacks.cdc.gov/view/cdc/261785/cdc_261785_DS1.pdf
• File Type:
  [PDF - 1000.65 KB]
• Collection(s):
  CDC Public Access
• Main Document Checksum:
  urn:sha-512:4b41eb866ed2a886e4a3b7b8b5b6028ef138274525020d6ec977f25149cb4e8cd1b8f2bc6d36bc1c619cf41df362723c9bff4b1f0524c4589d92592b26856486