Gait parameter and event estimation using smartphones

Pepa, Lucia; Verdini, Federica; Spalazzi, Luca

doi:10.1016/j.gaitpost.2017.06.011

Cited by 43 publications

(33 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Body-worn sensors, including those contained within smartphones, can be used to capture the kinematic properties of gait. Previous work has typically secured the smartphone or sensor tightly to the individual’s trunk [ 11 , 22 - 24 ] or lower extremities [ 10 , 25 , 26 ]. While that approach has been proven to enable measurement of gait metrics with enough sensitivity to distinguish between disease states, it has used additional equipment (eg, Velcro or elastic straps) together with trained personnel in a laboratory setting to provide assessment instructions.…”

Section: Discussionmentioning

confidence: 99%

“…Smartphones contain a 3-dimensional accelerometer, a 3-dimensional gyroscope, and a digital compass that are similar in sensitivity to research-grade biomechanical instrumentation. The smartphone, when secured to an individual’s lower back or sternum as they walk, can detect gait events such as heel strikes [ 10 ], as well as kinematic differences between those with and those without movement disorders, such as Parkinson disease [ 11 , 12 ]. Still, studies to date have been limited to laboratory environments and have required trained personnel to administer assessments, provide instructions, and secure the phone to the participant’s trunk.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Smartphone App–Based Assessment of Gait During Normal and Dual-Task Walking: Demonstration of Validity and Reliability

Manor

Zhu

et al. 2018

JMIR Mhealth Uhealth

View full text Add to dashboard Cite

BackgroundWalking is a complex cognitive motor task that is commonly completed while performing another task such as talking or making decisions. Gait assessments performed under normal and “dual-task” walking conditions thus provide important insights into health. Such assessments, however, are limited primarily to laboratory-based settings.ObjectiveThe objective of our study was to create and test a smartphone-based assessment of normal and dual-task walking for use in nonlaboratory settings.MethodsWe created an iPhone app that used the phone’s motion sensors to record movements during walking under normal conditions and while performing a serial-subtraction dual task, with the phone placed in the user’s pants pocket. The app provided the user with multimedia instructions before and during the assessment. Acquired data were automatically uploaded to a cloud-based server for offline analyses. A total of 14 healthy adults completed 2 laboratory visits separated by 1 week. On each visit, they used the app to complete three 45-second trials each of normal and dual-task walking. Kinematic data were collected with the app and a gold-standard–instrumented GAITRite mat. Participants also used the app to complete normal and dual-task walking trials within their homes on 3 separate days. Within laboratory-based trials, GAITRite-derived heel strikes and toe-offs of the phone-side leg aligned with smartphone acceleration extrema, following filtering and rotation to the earth coordinate system. We derived stride times—a clinically meaningful metric of locomotor control—from GAITRite and app data, for all strides occurring over the GAITRite mat. We calculated stride times and the dual-task cost to the average stride time (ie, percentage change from normal to dual-task conditions) from both measurement devices. We calculated similar metrics from home-based app data. For these trials, periods of potential turning were identified via custom-developed algorithms and omitted from stride-time analyses.ResultsAcross all detected strides in the laboratory, stride times derived from the app and GAITRite mat were highly correlated (P<.001, r2=.98). These correlations were independent of walking condition and pocket tightness. App- and GAITRite-derived stride-time dual-task costs were also highly correlated (P<.001, r2=.95). The error of app-derived stride times (mean 16.9, SD 9.0 ms) was unaffected by the magnitude of stride time, walking condition, or pocket tightness. For both normal and dual-task trials, average stride times derived from app walking trials demonstrated excellent test-retest reliability within and between both laboratory and home-based assessments (intraclass correlation coefficient range .82-.94).ConclusionsThe iPhone app we created enabled valid and reliable assessment of stride timing—with the smartphone in the pocket—during both normal and dual-task walking and within both laboratory and nonlaboratory environments. Additional work is warranted to expand the functionality of this tool to older adults and other pa...

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Smartphone App–Based Assessment of Gait During Normal and Dual-Task Walking: Demonstration of Validity and Reliability

Manor

Zhu

et al. 2018

JMIR Mhealth Uhealth

View full text Add to dashboard Cite

show abstract

“…A more recent work [ 32 ] evaluated the use of a smartphone’s accelerometer to detect foot contacts. However, like in the works described above, the device was placed over the trunk.…”

Section: Related Workmentioning

confidence: 99%

Smart Shoe-Assisted Evaluation of Using a Single Trunk/Pocket-Worn Accelerometer to Detect Gait Phases

Avvenuti

Carbonaro

Cimino

et al. 2018

Sensors

View full text Add to dashboard Cite

Wearable sensors may enable the continuous monitoring of gait out of the clinic without requiring supervised tests and costly equipment. This paper investigates the use of a single wearable accelerometer to detect foot contact times and estimate temporal gait parameters (stride time, swing and stance duration). The experiments considered two possible body positions for the accelerometer: over the lower trunk and inside a trouser pocket. The latter approach could be implemented using a common smartphone. Notably, during the experiments, the ground truth was obtained by using a pair of sensorized shoes. Unlike ambient sensors and camera-based systems, sensorized shoes enable the evaluation of body-worn sensors even during longer walks. Experiments showed that both trunk and pocket positions achieved promising results in estimating gait parameters, with a mean absolute error below 50 ms.

show abstract

“…This being a low-cost solution also takes away the need to mount an external sensor/device to capture motion related data. Use of inertial data obtained using smartphones has shown reasonable estimate of person's age [8], identity [22], gait [23,24], step count [25], stride length [26], walk distance [27], etc.…”

Section: Introductionmentioning

confidence: 99%

Person Re-Identification Using Deep Modeling of Temporally Correlated Inertial Motion Patterns

Gohar

Riaz

Shahzad

et al. 2020

Sensors

View full text Add to dashboard Cite

Person re-identification (re-ID) is among the essential components that play an integral role in constituting an automated surveillance environment. Majorly, the problem is tackled using data acquired from vision sensors using appearance-based features, which are strongly dependent on visual cues such as color, texture, etc., consequently limiting the precise re-identification of an individual. To overcome such strong dependence on visual features, many researchers have tackled the re-identification problem using human gait, which is believed to be unique and provide a distinctive biometric signature that is particularly suitable for re-ID in uncontrolled environments. However, image-based gait analysis often fails to extract quality measurements of an individual’s motion patterns owing to problems related to variations in viewpoint, illumination (daylight), clothing, worn accessories, etc. To this end, in contrast to relying on image-based motion measurement, this paper demonstrates the potential to re-identify an individual using inertial measurements units (IMU) based on two common sensors, namely gyroscope and accelerometer. The experiment was carried out over data acquired using smartphones and wearable IMUs from a total of 86 randomly selected individuals including 49 males and 37 females between the ages of 17 and 72 years. The data signals were first segmented into single steps and strides, which were separately fed to train a sequential deep recurrent neural network to capture implicit arbitrary long-term temporal dependencies. The experimental setup was devised in a fashion to train the network on all the subjects using data related to half of the step and stride sequences only while the inference was performed on the remaining half for the purpose of re-identification. The obtained experimental results demonstrate the potential to reliably and accurately re-identify an individual based on one’s inertial sensor data.

show abstract

Gait parameter and event estimation using smartphones

Cited by 43 publications

References 20 publications

Smartphone App–Based Assessment of Gait During Normal and Dual-Task Walking: Demonstration of Validity and Reliability

Smartphone App–Based Assessment of Gait During Normal and Dual-Task Walking: Demonstration of Validity and Reliability

Smart Shoe-Assisted Evaluation of Using a Single Trunk/Pocket-Worn Accelerometer to Detect Gait Phases

Person Re-Identification Using Deep Modeling of Temporally Correlated Inertial Motion Patterns

Contact Info

Product

Resources

About