Data-driven systems to detect physical weakening from daily routine: A pilot study on elderly over 80 years old

PLOS ONE RESEARCH ARTICLE Data-driven systems to detect physical weakening from daily routine: A pilot study on elderly over 80 years old Manuel Abbas ID1*, Majd Saleh1, Dominique Somme2, Régine Le Bouquin Jeannès1 1 Univ Rennes, Inserm, LTSI UMR 1099, Rennes, France, 2 Univ Rennes, CNRS, ARENES UMR6051, Rennes, France a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 * Abstract Published: January 30, 2023 The use of telemonitoring solutions via wearable sensors is believed to play a major role in the prevention and therapy of physical weakening in older adults. Despite the various studies found in the literature, some elements are still not well addressed, such as the study cohort, the experimental protocol, the type of research design, as well as the relevant features in this context. To this end, the objective of this pilot study was to investigate the efficacy of data-driven systems to characterize older individuals over 80 years of age with impaired physical function, during their daily routine and under unsupervised conditions. We propose a fully automated process which extracts a set of heterogeneous time-domain features from 24-hour files of acceleration and barometric data. After being statistically tested, the most discriminant features fed a group of machine learning classifiers to distinguish frail from non-frail subjects, achieving an accuracy up to 93.51%. Our analysis, conducted over 570 days of recordings, shows that a longitudinal study is important while using the proposed features, in order to ensure a highly specific diagnosis. This work may serve as a basis for the paradigm of future monitoring systems. Copyright: © 2023 Abbas et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Introduction OPEN ACCESS Citation: Abbas M, Saleh M, Somme D, Le Bouquin Jeannès R (2023) Data-driven systems to detect physical weakening from daily routine: A pilot study on elderly over 80 years old. PLoS ONE 18(1): e0274306. https://doi.org/10.1371/journal. pone.0274306 Editor: Emiliano Cè, Universita degli Studi di Milano, ITALY Received: July 1, 2021 Accepted: August 25, 2022 Data Availability Statement: All relevant data are within the paper or Supporting information files. Funding: This work was supported in part by the European Union through the European Regional Development Fund (ERDF), in part by the French Region of Brittany and Rennes Métropole under Grant EU000180, and in part by the French National Research Agency (ANR) in the context of the ACCORDS Project under Grant ANR-17-CE190024-01. - European Regional Development Fund: https://ec.europa.eu/regional_policy/en/funding/ With rising life expectancy, our society is getting older, and the majority of older adults would like to age at home. Nonetheless, frailty which leads to progressive physical weakening and increases the risk of falls and lack of autonomy [1, 2], threatens this lifestyle choice. This geriatric syndrome has been a trending topic in the past decades, and remains an emerging clinical and public health priority [3, 4]. Even though its definition is controversial [5], frailty is conceptualized as a clinical syndrome symbolized by reduced resistance to stressors, resulting from age-associated cumulative declines in physiological function across multiple organ systems [6, 7]. It is mainly identified by functional weakening and sarcopenia (i.e. loss of muscle tissue). Hence, the physical function is one of the most decisive elements in the prevention and therapy of old people’s health conditions [8, 9]. PLOS ONE | https://doi.org/10.1371/journal.pone.0274306 January 30, 2023 1 / 16 PLOS ONE erdf/ - French National Research Agency: https:// anr.fr/ The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. There was no additional external funding received for this study. Competing interests: The authors have declared that no competing interests exist. Signs of physical weakening in daily routine Fried et al. proposed the frailty phenotype (FrP) which is based on five criteria [10] and is one of the widely-used clinical tools for screening. Some other self-reported measures and performance tests like the Short Physical Performance Battery (SPPB) [11] and the International Physical Activity Questionnaire (IPAQ) [12] were also considered to evaluate mobility and to estimate the level of physical activity. While this kind of approaches is helpful in wellbeing assessment, it is not sufficient to define a full and accurate tracking system, seeing that most of these measurements (i) are subjective, (ii) are calculated under supervised conditions, and (iii) provide short-term estimation of physical capacities. This justifies the importance of telemonitoring solutions for elderly care. More recently, some works conducted observational studies by monitoring a set of different activity metrics from wearable sensors in relation to frailty status and health conditions. A pendant sensor was employed to monitor the physical activity of elderly (� 60 years old) over 48 hours, by observing the body posture (walking, sitting, standing), the activity behavior (sedentary, light, moderate-to-vigorous), steps parameters, and sleep parameters [13]. This study was further extended to assess cognitive frailty in community-dwelling older people [14]. The goal was to recognize the presence of both (i) physical frailty (defined by FrP) and (ii) cognitive impairment (defined by the score of Mini-Mental State Examination) using a chest-worn sensor over 48 hours. Another research has taken place to evaluate the reliability of wearable devices to assess frailty in older home care clients [15]. Acquired data of participants aged 55 years and older were exploited to derive variables including daily step count, sleep time and quality, and heart rate over a period of 9.43 (± 1.99) days. Besides, gait parameters, such as gait speed, cadence and stride time were inspected to assess frailty conditions [16]. Those parameters were calculated under supervised conditions, where participants walked a distance of 4 m at the self-selected usual pace, using wireless sensors fixed on the lower limb segments. Although the importance of remote monitoring was demonstrated in these works, we can identify several limitation in the literature. Firstly, wider sets of extracted variables are required. For instance, the muscle strength, the reduction of leisure activities (LAs), and instrumental activities of daily living (iADLs) are mainly associated to the functional capabilities limitations [17]. Therefore, it is important to observe the number of times an older person quits his/her house by taking the lift/stairs, a parameter that has not really been studied in the l (...truncated)