Dataset Information

A machine learning approach for gait speed estimation using skin-mounted wearable sensors: From healthy controls to individuals with multiple sclerosis.

ABSTRACT: Gait speed is a powerful clinical marker for mobility impairment in patients suffering from neurological disorders. However, assessment of gait speed in coordination with delivery of comprehensive care is usually constrained to clinical environments and is often limited due to mounting demands on the availability of trained clinical staff. These limitations in assessment design could give rise to poor ecological validity and limited ability to tailor interventions to individual patients. Recent advances in wearable sensor technologies have fostered the development of new methods for monitoring parameters that characterize mobility impairment, such as gait speed, outside the clinic, and therefore address many of the limitations associated with clinical assessments. However, these methods are often validated using normal gait patterns; and extending their utility to subjects with gait impairments continues to be a challenge. In this paper, we present a machine learning method for estimating gait speed using a configurable array of skin-mounted, conformal accelerometers. We establish the accuracy of this technique on treadmill walking data from subjects with normal gait patterns and subjects with multiple sclerosis-induced gait impairments. For subjects with normal gait, the best performing model systematically overestimates speed by only 0.01 m/s, detects changes in speed to within less than 1%, and achieves a root-mean-square-error of 0.12 m/s. Extending these models trained on normal gait to subjects with gait impairments yields only minor changes in model performance. For example, for subjects with gait impairments, the best performing model systematically overestimates speed by 0.01 m/s, quantifies changes in speed to within 1%, and achieves a root-mean-square-error of 0.14 m/s. Additional analyses demonstrate that there is no correlation between gait speed estimation error and impairment severity, and that the estimated speeds maintain the clinical significance of ground truth speed in this population. These results support the use of wearable accelerometer arrays for estimating walking speed in normal subjects and their extension to MS patient cohorts with gait impairment.

SUBMITTER: McGinnis RS

PROVIDER: S-EPMC5453431 | biostudies-literature | 2017

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A machine learning approach for gait speed estimation using skin-mounted wearable sensors: From healthy controls to individuals with multiple sclerosis.

McGinnis Ryan S RS Mahadevan Nikhil N Moon Yaejin Y Seagers Kirsten K Sheth Nirav N Wright John A JA DiCristofaro Steven S Silva Ikaro I Jortberg Elise E Ceruolo Melissa M Pindado Jesus A JA Sosnoff Jacob J Ghaffari Roozbeh R Patel Shyamal S

PloS one 20170601 6

Gait speed is a powerful clinical marker for mobility impairment in patients suffering from neurological disorders. However, assessment of gait speed in coordination with delivery of comprehensive care is usually constrained to clinical environments and is often limited due to mounting demands on the availability of trained clinical staff. These limitations in assessment design could give rise to poor ecological validity and limited ability to tailor interventions to individual patients. Recent ...[more]

PMID: 28570570

Similar Datasets

Project description:BackgroundMobility impairment is common in people with multiple sclerosis (PwMS) and there is a need to assess mobility in remote settings. Here, we apply a novel wireless, skin-mounted, and conformal inertial sensor (BioStampRC, MC10 Inc.) to examine gait characteristics of PwMS under controlled conditions. We determine the accuracy and precision of BioStampRC in measuring gait kinematics by comparing to contemporary research-grade measurement devices.MethodsA total of 45 PwMS, who presented with diverse walking impairment (Mild MS = 15, Moderate MS = 15, Severe MS = 15), and 15 healthy control subjects participated in the study. Participants completed a series of clinical walking tests. During the tests participants were instrumented with BioStampRC and MTx (Xsens, Inc.) sensors on their shanks, as well as an activity monitor GT3X (Actigraph, Inc.) on their non-dominant hip. Shank angular velocity was simultaneously measured with the inertial sensors. Step number and temporal gait parameters were calculated from the data recorded by each sensor. Visual inspection and the MTx served as the reference standards for computing the step number and temporal parameters, respectively. Accuracy (error) and precision (variance of error) was assessed based on absolute and relative metrics. Temporal parameters were compared across groups using ANOVA.ResultsMean accuracy±precision for the BioStampRC was 2±2 steps error for step number, 6±9ms error for stride time and 6±7ms error for step time (0.6-2.6% relative error). Swing time had the least accuracy±precision (25±19ms error, 5±4% relative error) among the parameters. GT3X had the least accuracy±precision (8±14% relative error) in step number estimate among the devices. Both MTx and BioStampRC detected significantly distinct gait characteristics between PwMS with different disability levels (p<0.01).ConclusionBioStampRC sensors accurately and precisely measure gait parameters in PwMS across diverse walking impairment levels and detected differences in gait characteristics by disability level in PwMS. This technology has the potential to provide granular monitoring of gait both inside and outside the clinic.

Project description:Distinct gait characteristics like short steps and shuffling gait are prototypical signs commonly observed in Parkinson's disease. Routinely assessed by observation through clinicians, gait is rated as part of categorical clinical scores. There is an increasing need to provide quantitative measurements of gait, e.g. to provide detailed information about disease progression. Recently, we developed a wearable sensor-based gait analysis system as diagnostic tool that objectively assesses gait parameter in Parkinson's disease without the need of having a specialized gait laboratory. This system consists of inertial sensor units attached laterally to both shoes. The computed target of measures are spatiotemporal gait parameters including stride length and time, stance phase time, heel-strike and toe-off angle, toe clearance, and inter-stride variation from gait sequences. To translate this prototype into medical care, we conducted a cross-sectional study including 190 Parkinson's disease patients and 101 age-matched controls and measured gait characteristics during a 4x10 meter walk at the subjects' preferred speed. To determine intraindividual changes in gait, we monitored the gait characteristics of 63 patients longitudinally. Cross-sectional analysis revealed distinct spatiotemporal gait parameter differences reflecting typical Parkinson's disease gait characteristics including short steps, shuffling gait, and postural instability specific for different disease stages and levels of motor impairment. The longitudinal analysis revealed that gait parameters were sensitive to changes by mirroring the progressive nature of Parkinson's disease and corresponded to physician ratings. Taken together, we successfully show that wearable sensor-based gait analysis reaches clinical applicability providing a high biomechanical resolution for gait impairment in Parkinson's disease. These data demonstrate the feasibility and applicability of objective wearable sensor-based gait measurement in Parkinson's disease reaching high technological readiness levels for both, large scale clinical studies and individual patient care.

Project description:BackgroundThe step-by-step determination of the spatio-temporal parameters of gait is clinically relevant since it provides an estimation of the variability of specific gait patterns associated with frequent geriatric syndromes. In recent years, several methods, based on the use of magneto-inertial units (MIMUs), have been developed for the step-by-step estimation of the gait temporal parameters. However, most of them were applied to the gait of healthy subjects and/or of a single pathologic population. Moreover, spatial parameters in pathologic populations have been rarely estimated step-by-step using MIMUs. The validity of clinically suitable MIMU-based methods for the estimation of spatio-temporal parameters is therefore still an open issue. The aim of this study was to propose and validate a method for the determination of both temporal and spatial parameters that could be applied to normal and heavily compromised gait patterns.MethodsTwo MIMUs were attached above each subject's ankles. An instrumented gait mat was used as gold standard. Gait data were acquired from ten hemiparetic subjects, ten choreic subjects, ten subjects with Parkinson's disease and ten healthy older adults walking at two different gait speeds. The method detects gait events (GEs) taking advantage of the cyclic nature of gait and exploiting some lower limb invariant kinematic characteristics. A combination of a MIMU axes realignment along the direction of progression and of an optimally filtered direct and reverse integration is used to determine the stride length.ResultsOver the 4,514 gait cycles analyzed, neither missed nor extra GEs were generated. The errors in identifying both initial and final contact at comfortable speed ranged between 0 and 11 ms for the different groups analyzed. The stride length was estimated for all subjects with less than 3% error.ConclusionsThe proposed method is apparently extremely robust since gait speed did not substantially affect its performance and both missed and extra GEs were avoided. The spatio-temporal parameters estimates showed smaller errors than those reported in previous studies and a similar level of precision and accuracy for both healthy and pathologic gait patterns. The combination of robustness, precision and accuracy suggests that the proposed method is suitable for routine clinical use.

Project description:Inappropriate physical inactivity is a global health problem increasing the risk of cardiometabolic diseases. Wearable sensors show great potential to promote physical activity and thus a healthier lifestyle. While commercial activity trackers are available to estimate energy expenditure (EE) in non-disabled individuals, they are not designed for reliable assessments in individuals with an incomplete spinal cord injury (iSCI). Furthermore, activity recommendations for this population are currently rather vague and not tailored to their individual needs, and activity guidelines provided for the non-disabled population may not be easily translated for this population. However, especially in iSCI individuals with impaired abilities to stand and walk, the assessment of physical activities and appropriate recommendations for a healthy lifestyle are challenging. Therefore, the study aimed at developing an EE estimation model for iSCI individuals able to walk based on wearable sensor data. Additionally, the data collected within this study was used to translate common activity recommendations for the non-disabled population to easily understandable activity goals for ambulatory individuals with an iSCI. In total, 30 ambulatory individuals with an iSCI were equipped with wearable sensors while performing 12 different physical activities. EE was measured continuously and demographic and anthropometric variables, clinical assessment scores as well as wearable-sensor-derived features were used to develop different EE estimation models. The best EE estimation model comprised the estimation of resting EE using the updated Harris-Benedict equation, classifying activities using a k-nearest neighbor algorithm, and applying a multiple linear regression-based EE estimation model for each activity class. The mean absolute estimation error of this model was 15.2 ± 6.3% and the corresponding mean signed error was -3.4 ± 8.9%. Translating activity recommendations of global health institutions, we suggest a minimum of 2,000-3,000 steps per day for ambulatory individuals with an iSCI. If ambulatory individuals with an iSCI targeted the popular 10,000 steps a day recommendation for the non-disabled population, their equivalent would be around 8,000 steps a day. The combination of the presented dedicated EE estimation model for ambulatory individuals with an iSCI and the translated activity recommendations is an important step toward promoting an active lifestyle in this population.

Dataset Information

A machine learning approach for gait speed estimation using skin-mounted wearable sensors: From healthy controls to individuals with multiple sclerosis.

Publications

A machine learning approach for gait speed estimation using skin-mounted wearable sensors: From healthy controls to individuals with multiple sclerosis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets