Predicting lying, sitting, walking and running using Apple Watch and Fitbit data

Fuller, Daniel; Anaraki, Javad Rahimipour; Simango, Bongai; Rayner, Machel; Dorani, Faramarz; Bozorgi, Arastoo; Luan, Hui; Basset, Fabien A.

doi:10.1136/bmjsem-2020-001004

Cited by 23 publications

(12 citation statements)

References 32 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, the device may encourage us to stand up if we sit still for a long time or request an SOS if we fall while alone. With the recent advances in sensors and wearable technologies, many studies have investigated using smartwatches as data-collection equipment [ 1 , 2 , 3 , 4 ]. To date, many HAR studies have focused on the coarse-grained classification of human movements, such as walking, running, sitting, and lying, each of which is a distinct activity.…”

Section: Introductionmentioning

confidence: 99%

Recognition of Fine-Grained Walking Patterns Using a Smartwatch with Deep Attentive Neural Networks

Kim

Park

et al. 2021

Sensors

View full text Add to dashboard Cite

Generally, people do various things while walking. For example, people frequently walk while looking at their smartphones. Sometimes we walk differently than usual; for example, when walking on ice or snow, we tend to waddle. Understanding walking patterns could provide users with contextual information tailored to the current situation. To formulate this as a machine-learning problem, we defined 18 different everyday walking styles. Noting that walking strategies significantly affect the spatiotemporal features of hand motions, e.g., the speed and intensity of the swinging arm, we propose a smartwatch-based wearable system that can recognize these predefined walking styles. We developed a wearable system, suitable for use with a commercial smartwatch, that can capture hand motions in the form of multivariate timeseries (MTS) signals. Then, we employed a set of machine learning algorithms, including feature-based and recent deep learning algorithms, to learn the MTS data in a supervised fashion. Experimental results demonstrated that, with recent deep learning algorithms, the proposed approach successfully recognized a variety of walking patterns, using the smartwatch measurements. We analyzed the results with recent attention-based recurrent neural networks to understand the relative contributions of the MTS signals in the classification process.

show abstract

Section: Introductionmentioning

confidence: 99%

Recognition of Fine-Grained Walking Patterns Using a Smartwatch with Deep Attentive Neural Networks

Kim

Park

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…The wrist-worn activity monitors collected physical activity data via triaxial accelerometers (25) and average heart rate (HR) each minute (26). Participants were instructed to wear the activity monitor as often as possible, particularly during all physical activities during the week after the initial evaluation.…”

Section: Methodsmentioning

confidence: 99%

Postconcussion Moderate to Vigorous Physical Activity Predicts Anxiety Severity among Adolescent Athletes

SMULLIGAN,

WINGERSON,

MAGLIATO

et al. 2023

Medicine &Amp; Science in Sports &Amp; Exercise

View full text Add to dashboard Cite

Purpose To examine how moderate-to-vigorous physical activity (MVPA) during concussion recovery influences self-reported anxiety symptoms at follow-up assessment. We hypothesized more MVPA after concussion would be associated with lower anxiety rating at follow-up. Methods We performed a prospective study of participants ages 13-18 years initially assessed within 14 days of diagnosed concussion. Participants rated concussion symptoms using the Post-Concussion Symptom Inventory (PCSI) and were provided a wrist-worn actigraphy device to track activity for 1-week after assessment. At follow-up assessment, participants rated anxiety symptoms using the 4-question Patient-Reported Outcome Measurement Information System (PROMIS) anxiety subscale. Each question ranged from 1 (never) to 5 (almost always) with an overall score range of 4-20. For univariable analysis, we calculated correlation coefficients between MVPA and PROMIS anxiety subscale scores. We then created a multiple linear regression model with follow-up PROMIS anxiety subscale score as the outcome and MVPA, sex, initial symptom severity, and pre-concussion anxiety as predictors. Results We enrolled and initially tested 55 participants, and 48 were included in the final analysis (age = 14.6±2.7 years; 56% female; initial assessment = 7.3±3.1 days, follow-up assessment = 42.0±29.7 days). We observed an inverse and low correlation between MVPA and follow-up PROMIS anxiety subscale T-scores (r = -0.30, p = 0.04). Multivariable regression results indicated that MVPA (β = -5.30, 95%CI = -10.58, -0.01), initial PCSI score (β = 0.11, 95%CI = 0.03, 0.19), and pre-concussion anxiety (β = 5.56, 95%CI = 0.12, 11.0), but not sex (β = -2.60, 95%CI = -7.14, 1.94), were associated with follow-up PROMIS anxiety subscale T-scores. Conclusions After adjusting for covariates, more MVPA early after concussion predicted lower PROMIS anxiety subscale scores at follow-up. While initial concussion symptom severity and pre-concussion anxiety were also associated with follow-up PROMIS anxiety subscale score, MVPA represents a modifiable factor that may contribute to lower anxiety symptoms.

show abstract

“…Although they measure the gait (Fuller et al 2021), they only perform sensing and do not have any function of actuation, such as transferring sensations directly to the body to promote walking or facilitating the walking motion by providing external stimulus or force to the body. Additionally, the battery requires recharging and replacement.…”

Section: Market Applications To Support Walking Promotionmentioning

confidence: 99%

Ankle-Worn Sensor Sleeve to increase walking motivation

2022

View full text Add to dashboard Cite

In this study, we propose the Ankle-Worn Sensor Sleeve to help people maintain their health. Furthermore, our approach uses this sensor sleeve to improve walking motivation by making the walking motion more sensible. In this research, we pursue the design of a wearable device that users can wear as a “shoe accessory,” with the aim of promoting people’s walking motivation. This device is ankle-worn and generates electricity using the physical flexion and extension of the ankle, detects and acquires the amount of electricity generated, and uses the resulting power to produce sound and light feedback while walking. The Sleeve is designed to support health problems, such as locomotive syndrome and dementia, which are caused by physical inactivity. This article presents the results of a questionnaire survey conducted on 172 people who experienced or observed the device during exhibitions. The results reveal that this device increased people’s interest and motivation to walk, including those who did not like to walk.

show abstract

Predicting lying, sitting, walking and running using Apple Watch and Fitbit data

Cited by 23 publications

References 32 publications

Recognition of Fine-Grained Walking Patterns Using a Smartwatch with Deep Attentive Neural Networks

Recognition of Fine-Grained Walking Patterns Using a Smartwatch with Deep Attentive Neural Networks

Postconcussion Moderate to Vigorous Physical Activity Predicts Anxiety Severity among Adolescent Athletes

Ankle-Worn Sensor Sleeve to increase walking motivation

Contact Info

Product

Resources

About