KIDS: kinematics-based (in)activity detection and segmentation in a sleep case study

Elnaggar, Omar ORCID: https://orcid.org/0000-0001-7417-6023, Arelhi, Roselina ORCID: https://orcid.org/0000-0002-7638-0383, Coenen, Frans ORCID: https://orcid.org/0000-0003-1026-6649, Hopkinson, Andrew, Mason, Lyndon ORCID: https://orcid.org/0000-0002-0371-3183 and Paoletti, Paolo ORCID: https://orcid.org/0000-0001-6131-0377 (2023) KIDS: kinematics-based (in)activity detection and segmentation in a sleep case study. [Data Collection]

Original publication URL: https://doi.org/10.48550/arXiv.2301.03469

Description

Sleep behaviour and in-bed movements contain rich information on the neurophysiological health of people, and have a direct link to the general well-being and quality of life. Standard clinical practices rely on polysomnography for sleep assessment; however, it is intrusive, performed in unfamiliar environments and requires trained personnel. Progress has been made on less invasive sensor technologies, such as actigraphy, but clinical validation raises concerns over their reliability and precision. Additionally, the field lacks a widely acceptable algorithm, with proposed approaches ranging from raw signal or feature thresholding to data-hungry classification models, many of which are unfamiliar to medical staff. This paper proposes an online Bayesian probabilistic framework for objective (in)activity detection and segmentation based on clinically meaningful joint kinematics, measured by a custom-made wearable sensor. Intuitive three-dimensional visualisations of kinematic timeseries were accomplished through dimension reduction based preprocessing, offering out-of-the-box framework explainability potentially useful for clinical monitoring and diagnosis. The proposed framework attained up to 99.2% F1-score and 0.96 Pearson's correlation coefficient in, respectively, the posture change detection and inactivity segmentation tasks. The work paves the way for a reliable home-based analysis of movements during sleep which would serve patient-centred longitudinal care plans.

Keywords: Wearable sensors, inertial data, inactivity segmentation, activity detection, bayesian inference, dimension reduction, sleep biomechanics
Divisions: Faculty of Health and Life Sciences > Institute of Life Courses and Medical Sciences > School of Medicine
Faculty of Health and Life Sciences > Institute of Population Health > Psychology
Faculty of Science and Engineering > School of Electrical Engineering, Electronics and Computer Science
Faculty of Science and Engineering > School of Engineering > Mechanical, Materials and Aerospace Engineering
Depositing User: Omar Elnaggar
Date Deposited: 19 Jan 2023 12:47
Last Modified: 19 Jan 2023 13:09
DOI: 10.17638/datacat.liverpool.ac.uk/2127
Geography: United Kingdom
URI: https://datacat.liverpool.ac.uk/id/eprint/2127

Available Files

Data

Creative Commons: Attribution 4.0
Creative Commons: Attribution-Noncommercial 4.0

Metadata Export