Smartwatch Step-Counting App for Older Adults: Development and Evaluation Study (Preprint)

2021 ◽  
Author(s):  
George Boateng ◽  
Curtis L. Petersen ◽  
David Kotz ◽  
Karen L. Fortuna ◽  
Rebecca Masutani ◽  
...  
Keyword(s):  
Physical Activity ◽  
Older Adults ◽  
Field Study ◽  
Ground Truth ◽  
Error Rates ◽  
Machine Learning Algorithms ◽  
Free Living ◽  
Step Counting ◽  
Best Parameter ◽  
Counting Algorithm

BACKGROUND Older adults who engage in physical activity can reduce their risk of mobility and disability. Short amounts of walking can improve their quality of life, physical function, and cardiovascular health. Various programs have been implemented to encourage older adults to engage in physical activity, but sustaining their motivation continues to be a challenge. Ubiquitous devices, such as mobile phones and smartwatches, coupled with machine-learning algorithms, can potentially encourage older adults to be more physically active. Current algorithms that are deployed in consumer devices (e.g., Fitbit) are proprietary, often are not tailored to the movements of older adults and have been shown to be inaccurate in clinical settings. Few studies have developed step-counting algorithms for smartwatches – but only using data from younger adults and often validating them only in controlled laboratory settings. OBJECTIVE In this work, we sought to develop and validate a smartwatch step-counting app targeting older adults that has been evaluated in free-living settings over a long period of time (24 weeks) with a large sample (N=42). METHODS the steps of older adults. The app includes algorithms to infer the level of physical activity and to count steps. We validated the step-counting algorithm with a total of 42 older adults in the lab (counting from a video recording, N= 20) and in free-living conditions — one 2-day field study (N=6) and two 12-week field studies (using the Fitbit as ground truth, N=16). During system development, we evaluated four kinds of walking patterns: normal, fast, up and down a staircase, and intermittent speed. For the field study, we evaluated various values for algorithm parameters, and subsequently evaluated the method’s performance using correlations and error rates. RESULTS The results from the evaluation showed that our step-counting algorithm performs well, highly correlated with the ground truth and with low error rate. For the lab study, there was stronger correlation for normal walking R2=0.5; across all activities, the Amulet was on average 3.2 (2.1%) steps lower (SD = 25.9) than video-validated steps. For the 2-day field study, the best parameter settings led to an association between Amulet and Fitbit (R2 of 0.989) and 3.1% (SD=25.1) steps lower than Fitbit, respectively. For the 12-week field study, the best parameter setting led to an R2 of 0.669. CONCLUSIONS Our findings demonstrate the importance of an iterative process in algorithm development in advance of field-based deployment. This work highlights various challenges and insights involved in developing and validating monitoring systems in real-world settings. Nonetheless, our step-counting app for older adults had good performance relative to the ground truth (a commercial Fitbit step-counter). Our app could potentially be used to improve the physical activity among older adults through accurate tracking of their step counts and in-app daily step-count goals.

scholarly journals A Machine Learning Classifier for Detection of Physical Activity Types and Postures During Free-Living

2021 ◽  
pp. 1-8
Author(s):  
Kerstin Bach ◽  
Atle Kongsvold ◽  
Hilde Bårdstu ◽  
Ellen Marie Bardal ◽  
Håkon S. Kjærnli ◽  
...  
Keyword(s):  
Physical Activity ◽  
Machine Learning ◽  
Ground Truth ◽  
Daily Physical Activity ◽  
Gradient Boosting ◽  
Free Living ◽  
Recording Time ◽  
Learning Classifier ◽  
Extreme Gradient Boosting ◽  
Activity Types

Introduction: Accelerometer-based measurements of physical activity types are commonly used to replace self-reports. To advance the field, it is desirable that such measurements allow accurate detection of key daily physical activity types. This study aimed to evaluate the performance of a machine learning classifier for detecting sitting, standing, lying, walking, running, and cycling based on a dual versus single accelerometer setups during free-living. Methods: Twenty-two adults (mean age [SD, range] 38.7 [14.4, 25–68] years) were wearing two Axivity AX3 accelerometers positioned on the low back and thigh along with a GoPro camera positioned on the chest to record lower body movements during free-living. The labeled videos were used as ground truth for training an eXtreme Gradient Boosting classifier using window lengths of 1, 3, and 5 s. Performance of the classifier was evaluated using leave-one-out cross-validation. Results: Total recording time was ∼38 hr. Based on 5-s windowing, the overall accuracy was 96% for the dual accelerometer setup and 93% and 84% for the single thigh and back accelerometer setups, respectively. The decreased accuracy for the single accelerometer setup was due to a poor precision in detecting lying based on the thigh accelerometer recording (77%) and standing based on the back accelerometer recording (64%). Conclusion: Key daily physical activity types can be accurately detected during free-living based on dual accelerometer recording, using an eXtreme Gradient Boosting classifier. The overall accuracy decreases marginally when predictions are based on single thigh accelerometer recording, but detection of lying is poor.

Analyzing Free-Living Physical Activity of Older Adults in Different Environments Using Body-Worn Activity Monitors

2010 ◽  
Vol 18 (2) ◽  
pp. 171-184 ◽  
Author(s):  
P. Margaret Grant ◽  
Malcolm H. Granat ◽  
Morag K. Thow ◽  
William M. Maclaren
Keyword(s):  
Physical Activity ◽  
Older Adults ◽  
Sedentary Time ◽  
Likelihood Estimation ◽  
Day Hospital ◽  
Free Living ◽  
Activity Monitors ◽  
The Third ◽  
Patient Groups ◽  
Restricted Maximum Likelihood Estimation

This study measured objectively the postural physical activity of 4 groups of older adults (≥65 yr). The participants (N= 70) comprised 3 patient groups—2 from rehabilitation wards (cityn= 20, 81.8 ± 6.7 yr; ruraln= 10, 79.4 ± 4.7 yr) and the third from a city day hospital (n= 20, 74.7 ± 7.9 yr)—and a healthy group to provide context (n= 20, 73.7 ± 5.5 yr). The participants wore an activity monitor (activPAL) for a week. A restricted maximum-likelihood-estimation analysis of hourly upright time (standing and walking) revealed significant differences between day, hour, and location and the interaction between location and hour (p< .001). Differences in the manner in which groups accumulated upright and sedentary time (sitting and lying) were found, with the ward-based groups sedentary for prolonged periods and upright for short episodes. This information may be used by clinicians to design appropriate rehabilitation interventions and monitor patient progress.

Validity, Reliability and Sensitivity to Change of Three Consumer-grade Activity Trackers in Controlled and Free-living Conditions among Older Adults (Preprint)

2021 ◽  
Author(s):  
Kaja Kastelic ◽  
Marina Dobnik ◽  
Stefan Loefler ◽  
Christian Hofer ◽  
Nejc Šarabon
Keyword(s):  
Physical Activity ◽  
Older Adults ◽  
Intraclass Correlation ◽  
Sleep Time ◽  
Living Conditions ◽  
Percentage Error ◽  
Sensitivity To Change ◽  
Free Living ◽  
Activity Trackers ◽  
Free Living Conditions

BACKGROUND Wrist worn consumer-grade activity trackers are popular devices, developed mainly for personal use, but with the potential to be used also for clinical and research purposes. OBJECTIVE The objective of this study was to explore the validity, reliability and sensitivity to change of movement behaviours metrics from three popular activity trackers (POLAR Vantage M, Garmin Vivosport and Garmin Vivoactive 4s) in controlled and free-living conditions when worn by older adults. METHODS Participants (n = 28; 74 ± 5 years) underwent a videotaped laboratory protocol while wearing all three activity trackers. On a separate occasion, participants wore one (randomly assigned) activity tracker and a research grade physical activity monitor ActiGraph wGT3X-BT simultaneously for six consecutive days for comparisons. RESULTS Both Garmin activity trackers showed excellent performance for step counts, with mean absolute percentage error (MAPE) below 20 % and intraclass correlation coefficient (ICC2,1) above 0.90 (P < .05), while Polar Vantage M substantially over counted steps (MAPE = 84 % and ICC2,1 = 0.37 for free-living conditions). MAPE for sleep time was within 10 % for all the trackers tested, while far beyond 20 % for all the physical activity and calories burned outputs. Both Garmin trackers showed fair agreement (ICC2,1 = 0.58–0.55) for measuring calories burned when compared with ActiGraph. CONCLUSIONS Garmin Vivoactive 4s showed overall best performance, especially for measuring steps and sleep time in healthy older adults. Minimal detectible change was consistently lower for an average day measures than for a single day measure, but still relatively high. The results provided in this study could be used to guide choice on activity trackers aiming for different purposes – individual use/care, longitudinal monitoring or in clinical trial setting.

Physical Activity Monitor Accuracy for Overground Walking and Free-Living Conditions Among Pregnant Women

2020 ◽  
Vol 3 (2) ◽  
pp. 100-109
Author(s):  
Christopher P. Connolly ◽  
Jordana Dahmen ◽  
Robert D. Catena ◽  
Nigel Campbell ◽  
Alexander H.K. Montoye
Keyword(s):  
Physical Activity ◽  
Pregnant Women ◽  
Living Conditions ◽  
Overground Walking ◽  
Third Trimester ◽  
Free Living ◽  
Activity Monitors ◽  
Step Counting ◽  
Physical Activity Monitors ◽  
Free Living Conditions

Purpose: We aimed to determine the step-count validity of commonly used physical activity monitors for pregnancy overground walking and during free-living conditions. Methods: Participants (n = 39, 12–38 weeks gestational age) completed six 100-step overground walking trials (three self-selected “normal pace”, three “brisk pace”) while wearing five physical activity monitors: Omron HJ-720 (OM), New Lifestyles 2000 (NL), Fitbit Flex (FF), ActiGraph Link (AG), and Modus StepWatch (SW). For each walking trial, monitor-recorded steps and criterion-measured steps were assessed. Participants also wore all activity monitors for an extended free-living period (72 hours), with the SW used as the criterion device. Mean absolute percent error (MAPE) was calculated for overground walking and free-living protocols and compared across monitors. Results: For overground walking, the OM, NL, and SW performed well (<5% MAPE) for normal and brisk pace walking trials, and also when trials were analyzed by actual speeds. The AG and FF had significantly greater MAPE for overground walking trials (11.9–14.7%). Trimester did affect device accuracy to some degree for the AG, FF, and SW, with error being lower in the third trimester compared to the second. For the free-living period, the OM, NL, AG, and FF significantly underestimated (>32% MAPE) actual steps taken per day as measured by the criterion SW (M [SD] = 9,350 [3,910]). MAPE for the OM was particularly high (45.3%). Conclusion: The OM, NL, and SW monitors are valid measures for overground step-counting during pregnancy walking. However, the OM and NL significantly underestimate steps by second and third trimester pregnant women in free-living conditions.

scholarly journals Physical Activity Intensity Cut-Points for Wrist-Worn GENEActiv in Older Adults

2021 ◽  
Vol 2 ◽  
Author(s):  
François Fraysse ◽  
Dannielle Post ◽  
Roger Eston ◽  
Daiki Kasai ◽  
Alex V. Rowlands ◽  
...  
Keyword(s):  
Physical Activity ◽  
Older Adults ◽  
Classification Accuracy ◽  
Operating Characteristic ◽  
Sedentary Time ◽  
Vigorous Physical Activity ◽  
Physical Activity Intensity ◽  
Characteristic Analysis ◽  
Free Living ◽  
Activity Intensity

Purpose: This study aims to (1) establish GENEActiv intensity cutpoints in older adults and (2) compare the classification accuracy between dominant (D) or non-dominant (ND) wrist, using both laboratory and free-living data.Methods: Thirty-one older adults participated in the study. They wore a GENEActiv Original on each wrist and performed nine activities of daily living. A portable gas analyzer was used to measure energy expenditure for each task. Testing was performed on two occasions separated by at least 8 days. Some of the same participants (n = 13) also wore one device on each wrist during 3 days of free-living. Receiver operating characteristic analysis was performed to establish the optimal cutpoints.Results: For sedentary time, both dominant and non-dominant wrist had excellent classification accuracy (sensitivity 0.99 and 0.97, respectively; specificity 0.91 and 0.86, respectively). For Moderate to Vigorous Physical Activity (MVPA), the non-dominant wrist device had better accuracy (ND sensitivity: 0.90, specificity 0.79; D sensitivity: 0.90, specificity 0.64). The corresponding cutpoints for sedentary-to-light were 255 and 375 g · min (epoch independent: 42.5 and 62.5 mg), and those for the light-to-moderate were 588 and 555 g · min (epoch-independent: 98.0 and 92.5 mg) for the non-dominant and dominant wrist, respectively. For free-living data, the dominant wrist device resulted in significantly more sedentary time and significantly less light and MVPA time compared to the non-dominant wrist.

scholarly journals DIETARY PATTERNS, NUTRIENT INTAKES, AND NUTRITIONAL AND PHYSICAL ACTIVITY STATUS OF SAUDI OLDER ADULTS: A NARRATIVE REVIEW

2015 ◽  
pp. 1-10
Author(s):  
H. M. Alsufiani ◽  
T.A. Kumosani ◽  
D. Ford ◽  
J.C. Mathers
Keyword(s):  
Physical Activity ◽  
Older Adults ◽  
Vitamin D ◽  
Saudi Arabia ◽  
Iron Deficiency ◽  
Iron Deficiency Anemia ◽  
Dietary Patterns ◽  
Deficiency Anemia ◽  
Nutrient Intakes ◽  
Free Living

Objective: to review the dietary patterns, nutrient intakes, and nutritional and physical activity status of older adults living in Saudi Arabia, to examine geographical differences in such patterns and to identify research gaps in respect of nutrition and physical activity for this population group. Design: Databases and websites (including Pubmed, Scopus, Proquest, Google Scholar and Arab Center for Nutrition) were searched in English and Arabic languages using the following key words: nutritional status, dietary pattern, food pattern, dietary habits, micronutrient intake and status, macronutrients intake, obesity, malnutrition, iron deficiency anemia, vitamin D, physical activity, exercise, Saudi older adults and Saudi elderly. All relevant and available data for both free-living and institutionalized Saudi older adults (> 50 years old or with mean age > 50 years) published in the last 20 years were included in this review. Results: We found that free-living females consumed fewer meals, and less fruits and vegetables, but their reported energy intake was higher than for males. Low intake of vitamins C and D were common in both genders and in those who lived in western and northern regions while low intake of folate and fiber were common in institutionalized people. Omega-3 fatty acids and fish were more highly consumed by older adults living in the coastal region compared with residents in the internal region. Obesity, overweight, vitamin D deficiency and insufficiency and physical inactivity were prevalent in free living older adults throughout the country while underweight and iron deficiency anemia were prevalent in institutionalized persons. Conclusion: Information on dietary patterns, nutrient intakes, and nutritional and physical activity status of older adults living in Saudi Arabia is fragmentary and interpretation of the findings is hampered by the lack of population-representative sampling frames and the use of heterogeneous data collection tools. More systematic studies are essential to facilitate objective assessment of these important lifestyle-related factors and to inform public health policies.

scholarly journals IS VARIABILITY OF FREE-LIVING ACTIVITY ASSOCIATED WITH PHYSICAL AND MENTAL FATIGABILITY IN OLDER ADULTS?

2019 ◽  
Vol 3 (Supplement_1) ◽  
pp. S233-S234
Author(s):  
Jessica L Graves ◽  
Robert T Krafty ◽  
Jaroslaw Harezlak ◽  
Eric J Shiroma ◽  
Nancy W Glynn
Keyword(s):  
Physical Activity ◽  
Older Adults ◽  
Cohort Study ◽  
Standard Deviation ◽  
Rise Time ◽  
Activity Patterns ◽  
Relative Activity ◽  
Energy Reserves ◽  
Free Living ◽  
The Relationship

Abstract Greater fatigability in older adults may be moderated by physical activity (PA). However, what features of PA timing are most strongly related to fatigability remains unknown. We examined the relationship between variability of free-living activity patterns and perceived physical and mental fatigability using the Pittsburgh Fatigability Scale (PFS, 0-50pts, higher=greater fatigability) in older adults from the Developmental Epidemiologic Cohort Study (DECOS, n=57, age=70-91yrs, 61% female). We assessed PA using ActiGraph GT3X+ over 7 days. Mean activity, standard deviation (SD) of mean activity across days, and relative activity [(mean at each bin)/(total mean)] were calculated across 24-hours in 4-hour bins , adjusting for estimated rise-time. Lower SD of PA from 0-4 hours after rising was associated with greater PFS physical scores (r=-0.27, p=0.05). No measures of PA correlated with PFS mental scores. In older adults with lower physical fatigability, associations with greater variability in activity may indicate larger energy reserves.

Free-living standing activity as assessed by seismic accelerometers and cognitive function in community-dwelling older adults: The MIND trial

2021 ◽  
Author(s):  
Shannon Halloway ◽  
Klodian Dhana ◽  
Pankaja Desai ◽  
Puja Agarwal ◽  
Thomas Holland ◽  
...  
Keyword(s):  
Physical Activity ◽  
Older Adults ◽  
Cognitive Function ◽  
Daily Life ◽  
Vigorous Physical Activity ◽  
Community Dwelling ◽  
Perceptual Speed ◽  
Free Living ◽  
Activity Intensity ◽  
The Impact

Abstract Background Few older adults are able to achieve recommended levels of moderate-vigorous physical activity despite known cognitive benefits. Alternatively, less intense activities such as standing can be easily integrated into daily life. No existing study has examined the impact of free-living standing activity during daily life as measured by a device on cognition in older adults. Our purpose was to examine the association between free-living standing activity and cognitive function in cognitively healthy older adults. Methods Participants were 98 adult participants aged 65 years or older from the ongoing MIND trial (NCT02817074) without diagnoses or symptoms of mild cognitive impairment or dementia. Linear regression analyses tested cross-sectional associations between standing activity (duration and intensity from the MoveMonitor+ accelerometer/gyroscope) and cognition (4 cognitive domains constructed from 12 cognitive performance tests). Results Participants were on average 69.7 years old (SD = 3.7), 69.4% women, and 73.5% had a college degree or higher. Higher mean intensity of standing activity was significantly associated with higher levels of perceptual speed when adjusting for age, gender, and education level. Each log unit increase in standing activity intensity was associated with 0.72 units higher of perceptual speed (p=.023). When we additionally adjusted for cognitive activities and moderate-vigorous physical activity, and then also for body mass index, depressive symptoms, prescription medication use, and device wear time, the positive association remained. Conclusions These findings should be further explored in longitudinal analyses and interventions for cognition that incorporate small changes to free-living activity in addition to promoting moderate-vigorous physical activity.

Sign in / Sign up

Export Citation Format

Share Document