Accurate walking and running speed estimation using wrist inertial data

Bertschi, Mattia; Celka, P.; Delgado-Gonzalo, Ricard; Lemay, Mathieu; Calvo, Enric M.; Grossenbacher, Olivier; Renevey, Ph.

doi:10.1109/embc.2015.7320269

Cited by 12 publications

(2 citation statements)

References 22 publications

(14 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These methods extracted several features from raw sensor data and then mapped them to gait speed though a linear or non-linear modeling. The features were in general chosen to be indicator of intensity, energy, cadence, mean crossing rate, but statistical features such as mean, standard deviation, mode, and median of acceleration norm have also been employed [20], [40]- [45]. Altitude changes, measured by a barometric pressure sensor, were also used as a feature to improve gait speed estimation in [20].…”

Section: Introductionmentioning

confidence: 99%

Real-World Gait Speed Estimation Using Wrist Sensor: A Personalized Approach

Soltani¹,

Dejnabadi²,

Savary³

et al. 2020

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

Gait speed is an important parameter to characterize people's daily mobility. For real-world speed measurement, inertial sensors or global navigation satellite system (GNSS) can be used on wrist, possibly integrated in a wristwatch. However, power consumption of GNSS is high and data are only available outdoor. Gait speed estimation using wrist-mounted inertial sensors is generally based on machine learning and suffers from low accuracy because of the inadequacy of using limited training data to build a general speed model that would be accurate for the whole population. To overcome this issue, a personalized model was proposed, which took unique gait style of each subject into account. Cadence and other biomechanically derived gait features were extracted from a wrist-mounted accelerometer and barometer. Gait features were fused with few GNSS data (sporadically sampled during gait) to calibrate the step length model of each subject through online learning. The proposed method was validated on 30 healthy subjects where it has achieved a median [Interquartile Range] of root mean square error of 0.05 [0.04-0.06] (m/s) and 0.14 [0.11-0.17] (m/s) for walking and running, respectively. Results demonstrated that the personalized model provided similar performance as GNSS. It used 50 times less training GNSS data than nonpersonalized method and achieved even better results. This parsimonious GNSS usage allowed extending battery life. The proposed algorithm met requirements for applications which need accurate, long, real-time, lowpower, and indoor/outdoor speed estimation in daily life. Index Terms-Low-power, online learning, personalization, real-world gait speed, running and walking.

show abstract

Section: Introductionmentioning

confidence: 99%

Real-World Gait Speed Estimation Using Wrist Sensor: A Personalized Approach

Soltani¹,

Dejnabadi²,

Savary³

et al. 2020

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

show abstract

“…Locomotion speed can already be accurately extracted from IMU readings over shorter distances [73, 74]. Since GPS and IMUs are standard components of contemporary smartphones, there is potential for smartphone-based measurements to replace or complement traditional – usually stopwatch timed – walking tests over short distances (e.g.…”

Section: Introductionmentioning

confidence: 99%

MOBIlity assessment with modern TEChnology in older patients’ real-life by the General Practitioner: the MOBITEC-GP study protocol

et al. 2019

View full text Add to dashboard Cite

BackgroundMobility limitations in older adults are associated with poor clinical outcomes including higher mortality and disability rates. A decline in mobility (including physical function and life-space) is detectable and should be discovered as early as possible, as it can still be stabilized or even reversed in early stages by targeted interventions. General practitioners (GPs) would be in the ideal position to monitor the mobility of their older patients. However, easy-to-use and valid instruments for GPs to conduct mobility assessment in the real-life practice setting are missing. Modern technologies such as the global positioning system (GPS) and inertial measurement units (IMUs) - nowadays embedded in every smartphone - could facilitate monitoring of different aspects of mobility in the GP's practice.MethodsThis project’s aim is to provide GPs with a novel smartphone application that allows them to quantify their older patients’ mobility. The project consists of three parts: development of the GPS- and IMU-based application, evaluation of its validity and reliability (Study 1), and evaluation of its applicability and acceptance (Study 2).In Study 1, participants (target N = 72, aged 65+, ≥2 chronic diseases) will perform a battery of walking tests (varying distances; varying levels of standardization). Besides videotaping and timing (gold standard), a high-end GPS device, a medium-accuracy GPS/IMU logger and three different smartphone models will be used to determine mobility parameters such as gait speed. Furthermore, participants will wear the medium-accuracy GPS/IMU logger and a smartphone for a week to determine their life-space mobility. Participants will be re-assessed after 1 week. In Study 2, participants (target N = 60, aged 65+, ≥2 chronic diseases) will be instructed on how to use the application by themselves. Participants will perform mobility assessments independently at their own homes. Aggregated test results will also be presented to GPs. Acceptance of the application will be assessed among patients and GPs. The application will then be finalized and publicly released.DiscussionIf successful, the MOBITEC-GP application will offer health care providers the opportunity to follow their patients’ mobility over time and to recognize impending needs (e.g. for targeted exercise) within pre-clinical stages of decline.

show abstract