sensors-logo

Journal Browser

Journal Browser

Wearable Movement Sensors for Rehabilitation: From Technology to Clinical Practice

A special issue of Sensors (ISSN 1424-8220). This special issue belongs to the section "Physical Sensors".

Deadline for manuscript submissions: closed (1 June 2021) | Viewed by 67645

Printed Edition Available!
A printed edition of this Special Issue is available here.

Special Issue Editors


E-Mail Website
Guest Editor
Department of Rehabilitation Medicine, Erasmus MC University Medical Centre Rotterdam, 3015 GD Rotterdam, The Netherlands
Interests: wearable movement sensors; signal analysis; motion analysis; algorithms; sensor-based activity recognition; machine learning; deep neural networks; physical behavior; upper limb kinematics; stroke rehabilitation

E-Mail Website
Guest Editor
Department of Rehabilitation Medicine, Erasmus MC University Medical Centre Rotterdam, 3015 GD Rotterdam, The Netherlands
Interests: physical behavior in people with chronic conditions; accelerometry; e-health; stroke rehabilitation; ambulatory monitoring

E-Mail Website
Guest Editor
Department of Rehabilitation Medicine, Erasmus MC University Medical Centre Rotterdam, Rotterdam, The Netherlands
Interests: neurorehabilitation

Special Issue Information

Dear Colleagues,

The rapid innovations in wearable movement sensors in recent years provide an opportunity to translate these innovations into the field of rehabilitation. Wearable movement sensors can provide objective and precise measurements of the quantity and quality of physical activities, body postures, and movements in clinical as well as normal daily life environments, thereby providing clinicians with data that can be used to guide, personalize, and optimize therapy. Since wearable sensors are portable, inexpensive, and unobtrusive, and also have the ability to provide information that cannot be obtained otherwise, they have an enormous potential for the tracking of patient functioning and recovery in rehabilitation. In addition, wearables can play a crucial role in the existing tendency towards at-home monitoring and treatment.

Although the opportunities are enormous, wearable movement sensors are relatively scarcely applied in the rehabilitation clinic. Important challenges remain, such as the reliability and validity of wearable movement sensors in clinical populations and free-living environments, barriers in the clinical application of wearable movement sensors, the development of sensor configurations and machine learning-based algorithms that enable detection of specific activities and movements in free-living conditions, and the development of disease-specific sensor-based outcome measures that are interpretable by patients and clinicians.

This Special Issue, entitled “Wearable Movement Sensors for Rehabilitation: From Technology to Clinical Practice”, aims to facilitate the application of wearable movement sensors in clinical practice. It intends to explore the opportunities for the application of wearable movement sensors in rehabilitation.

The main topics of this Special Issue include, but are not limited to, the following:

  • State-of-the-art and next-generation wearable movement sensors, e.g., accelerometry, hybrid sensors, soft sensors, smart/intelligent sensors, and body sensor networks
  • Smartphone-based movement monitoring
  • Technical validity, reliability, and clinical validity of sensor-based measurements
  • Usability and patient acceptance of wearable sensor systems
  • Wearable sensor-based feedback systems
  • Telerehabilitation based on wearable movement sensors
  • Monitoring of physical behavior
  • Instrumented clinical assessments
  • Machine learning-based movement/activity recognition in free-living environments
  • Clinical application in chronic conditions, such as stroke, neurological disorders, musculoskeletal injury/impairments, spinal cord injury, multiple sclerosis, cerebral palsy, or related conditions

Dr. G.R.H. (Ruben) Regterschot
Dr. J.B.J. (Hans) Bussmann
Prof. Dr. Gerard M. Ribbers
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Sensors is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • wearable movement sensors
  • monitoring
  • accelerometry
  • telemonitoring
  • feedback
  • machine learning
  • usability
  • rehabilitation
  • stroke
  • neurological disorders
  • musculoskeletal impairments

Published Papers (18 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Editorial

Jump to: Research, Review, Other

5 pages, 184 KiB  
Editorial
Wearable Movement Sensors for Rehabilitation: From Technology to Clinical Practice
by Gerrit Ruben Hendrik Regterschot, Gerard M. Ribbers and Johannes B. J. Bussmann
Sensors 2021, 21(14), 4744; https://doi.org/10.3390/s21144744 - 12 Jul 2021
Cited by 6 | Viewed by 2724
Abstract
Motor disorders are a common and age-related problem in the general community [...] Full article

Research

Jump to: Editorial, Review, Other

12 pages, 2617 KiB  
Article
Whole-Body Movements Increase Arm Use Outcomes of Wrist-Worn Accelerometers in Stroke Patients
by Gerrit Ruben Hendrik Regterschot, Ruud W. Selles, Gerard M. Ribbers and Johannes B. J. Bussmann
Sensors 2021, 21(13), 4353; https://doi.org/10.3390/s21134353 - 25 Jun 2021
Cited by 12 | Viewed by 1767
Abstract
Wrist-worn accelerometers are often applied to measure arm use after stroke. They measure arm movements during all activities, including whole-body movements, such as walking. Whole-body movements may influence clinimetric properties of arm use measurements—however, this has not yet been examined. This study investigates [...] Read more.
Wrist-worn accelerometers are often applied to measure arm use after stroke. They measure arm movements during all activities, including whole-body movements, such as walking. Whole-body movements may influence clinimetric properties of arm use measurements—however, this has not yet been examined. This study investigates to what extent arm use measurements with wrist-worn accelerometers are affected by whole-body movements. Assuming that arm movements during whole-body movements are non-functional, we quantify the effect of whole-body movements by comparing two methods: Arm use measured with wrist-worn accelerometers during all whole-body postures and movements (P&M method), and during sitting/standing only (sit/stand method). We have performed a longitudinal observational cohort study with measurements in 33 stroke patients during weeks 3, 12, and 26 poststroke. The P&M method shows higher daily paretic arm use outcomes than the sit/stand method (p < 0.001), the mean difference increased from 31% at week three to 41% at week 26 (p < 0.001). Differences in daily paretic arm use between methods are strongly related to daily walking time (r = 0.83–0.92). Changes in the difference between methods are strongly related to changes in daily walking time (r = 0.89). We show that not correcting arm use measurements for whole-body movements substantially increases arm use outcomes, thereby threatening the validity of arm use outcomes and measured arm use changes. Full article
Show Figures

Figure 1

11 pages, 1190 KiB  
Communication
Wearable Activity Monitoring in Day-to-Day Stroke Care: A Promising Tool but Not Widely Used
by Hanneke E. M. Braakhuis, Johannes B. J. Bussmann, Gerard M. Ribbers and Monique A. M. Berger
Sensors 2021, 21(12), 4066; https://doi.org/10.3390/s21124066 - 12 Jun 2021
Cited by 9 | Viewed by 3066
Abstract
Physical activity monitoring with wearable technology has the potential to support stroke rehabilitation. Little is known about how physical therapists use and value the use of wearable activity monitors. This cross-sectional study explores the use, perspectives, and barriers to wearable activity monitoring in [...] Read more.
Physical activity monitoring with wearable technology has the potential to support stroke rehabilitation. Little is known about how physical therapists use and value the use of wearable activity monitors. This cross-sectional study explores the use, perspectives, and barriers to wearable activity monitoring in day-to-day stroke care routines amongst physical therapists. Over 300 physical therapists in primary and geriatric care and rehabilitation centers in the Netherlands were invited to fill in an online survey that was developed based on previous studies and interviews with experts. In total, 103 complete surveys were analyzed. Out of the 103 surveys, 27% of the respondents were already using activity monitoring. Of the suggested treatment purposes of activity monitoring, 86% were perceived as useful by more than 55% of the therapists. The most recognized barriers to clinical implementation were lack of skills and knowledge of patients (65%) and not knowing what brand and type of monitor to choose (54%). Of the non-users, 79% were willing to use it in the future. In conclusion, although the concept of remote activity monitoring was perceived as useful, it was not widely adopted by physical therapists involved in stroke care. To date, skills, beliefs, and attitudes of individual therapists determine the current use of wearable technology. Full article
Show Figures

Figure 1

18 pages, 2968 KiB  
Article
Automatically Determining Lumbar Load during Physically Demanding Work: A Validation Study
by Charlotte Christina Roossien, Christian Theodoor Maria Baten, Mitchel Willem Pieter van der Waard, Michiel Felix Reneman and Gijsbertus Jacob Verkerke
Sensors 2021, 21(7), 2476; https://doi.org/10.3390/s21072476 - 02 Apr 2021
Cited by 4 | Viewed by 2035
Abstract
A sensor-based system using inertial magnetic measurement units and surface electromyography is suitable for objectively and automatically monitoring the lumbar load during physically demanding work. The validity and usability of this system in the uncontrolled real-life working environment of physically active workers are [...] Read more.
A sensor-based system using inertial magnetic measurement units and surface electromyography is suitable for objectively and automatically monitoring the lumbar load during physically demanding work. The validity and usability of this system in the uncontrolled real-life working environment of physically active workers are still unknown. The objective of this study was to test the discriminant validity of an artificial neural network-based method for load assessment during actual work. Nine physically active workers performed work-related tasks while wearing the sensor system. The main measure representing lumbar load was the net moment around the L5/S1 intervertebral body, estimated using a method that was based on artificial neural network and perceived workload. The mean differences (MDs) were tested using a paired t-test. During heavy tasks, the net moment (MD = 64.3 ± 13.5%, p = 0.028) and the perceived workload (MD = 5.1 ± 2.1, p < 0.001) observed were significantly higher than during the light tasks. The lumbar load had significantly higher variances during the dynamic tasks (MD = 33.5 ± 36.8%, p = 0.026) and the perceived workload was significantly higher (MD = 2.2 ± 1.5, p = 0.002) than during static tasks. It was concluded that the validity of this sensor-based system was supported because the differences in the lumbar load were consistent with the perceived intensity levels and character of the work tasks. Full article
Show Figures

Figure 1

10 pages, 1054 KiB  
Communication
Quantifying Circadian Aspects of Mobility-Related Behavior in Older Adults by Body-Worn Sensors—An “Active Period Analysis”
by Tim Fleiner, Rieke Trumpf, Anna Hollinger, Peter Haussermann and Wiebren Zijlstra
Sensors 2021, 21(6), 2121; https://doi.org/10.3390/s21062121 - 18 Mar 2021
Cited by 2 | Viewed by 1962
Abstract
Disruptions of circadian motor behavior cause a significant burden for older adults as well as their caregivers and often lead to institutionalization. This cross-sectional study investigates the association between mobility-related behavior and subjectively rated circadian chronotypes in healthy older adults. The physical activity [...] Read more.
Disruptions of circadian motor behavior cause a significant burden for older adults as well as their caregivers and often lead to institutionalization. This cross-sectional study investigates the association between mobility-related behavior and subjectively rated circadian chronotypes in healthy older adults. The physical activity of 81 community-dwelling older adults was measured over seven consecutive days and nights using lower-back-worn hybrid motion sensors (MM+) and wrist-worn actigraphs (MW8). A 30-min and 120-min active period for the highest number of steps (MM+) and activity counts (MW8) was derived for each day, respectively. Subjective chronotypes were classified by the Morningness-Eveningness Questionnaire into 40 (50%) morning types, 35 (43%) intermediate and six (7%) evening types. Analysis revealed significantly earlier starts for the 30-min active period (steps) in the morning types compared to the intermediate types (p ≤ 0.01) and the evening types (p ≤ 0.01). The 120-min active period (steps) showed significantly earlier starts in the morning types compared to the intermediate types (p ≤ 0.01) and the evening types (p = 0.02). The starting times of active periods determined from wrist-activity counts (MW8) did not reveal differences between the three chronotypes (p = 0.36 for the 30-min and p = 0.12 for the 120-min active period). The timing of mobility-related activity, i.e., periods with the highest number of steps measured by hybrid motion sensors, is associated to subjectively rated chronotypes in healthy older adults. The analysis of individual active periods may provide an innovative approach for early detecting and individually tailoring the treatment of circadian disruptions in aging and geriatric healthcare. Full article
Show Figures

Figure 1

25 pages, 6503 KiB  
Article
An Inertial Measurement Unit-Based Wireless System for Shoulder Motion Assessment in Patients with Cervical Spinal Cord Injury: A Validation Pilot Study in a Clinical Setting
by Riccardo Bravi, Stefano Caputo, Sara Jayousi, Alessio Martinelli, Lorenzo Biotti, Ilaria Nannini, Erez James Cohen, Eros Quarta, Stefano Grasso, Giacomo Lucchesi, Gabriele Righi, Giulio Del Popolo, Lorenzo Mucchi and Diego Minciacchi
Sensors 2021, 21(4), 1057; https://doi.org/10.3390/s21041057 - 04 Feb 2021
Cited by 8 | Viewed by 3568
Abstract
Residual motion of upper limbs in individuals who experienced cervical spinal cord injury (CSCI) is vital to achieve functional independence. Several interventions were developed to restore shoulder range of motion (ROM) in CSCI patients. However, shoulder ROM assessment in clinical practice is commonly [...] Read more.
Residual motion of upper limbs in individuals who experienced cervical spinal cord injury (CSCI) is vital to achieve functional independence. Several interventions were developed to restore shoulder range of motion (ROM) in CSCI patients. However, shoulder ROM assessment in clinical practice is commonly limited to use of a simple goniometer. Conventional goniometric measurements are operator-dependent and require significant time and effort. Therefore, innovative technology for supporting medical personnel in objectively and reliably measuring the efficacy of treatments for shoulder ROM in CSCI patients would be extremely desirable. This study evaluated the validity of a customized wireless wearable sensors (Inertial Measurement Units—IMUs) system for shoulder ROM assessment in CSCI patients in clinical setting. Eight CSCI patients and eight healthy controls performed four shoulder movements (forward flexion, abduction, and internal and external rotation) with dominant arm. Every movement was evaluated with a goniometer by different testers and with the IMU system at the same time. Validity was evaluated by comparing IMUs and goniometer measurements using Intraclass Correlation Coefficient (ICC) and Limits of Agreement (LOA). inter-tester reliability of IMUs and goniometer measurements was also investigated. Preliminary results provide essential information on the accuracy of the proposed wireless wearable sensors system in acquiring objective measurements of the shoulder movements in CSCI patients. Full article
Show Figures

Graphical abstract

17 pages, 1434 KiB  
Article
3D Motion Capture May Detect Spatiotemporal Changes in Pre-Reaching Upper Extremity Movements with and without a Real-Time Constraint Condition in Infants with Perinatal Stroke and Cerebral Palsy: A Longitudinal Case Series
by Julia Mazzarella, Mike McNally, Daniel Richie, Ajit M. W. Chaudhari, John A. Buford, Xueliang Pan and Jill C. Heathcock
Sensors 2020, 20(24), 7312; https://doi.org/10.3390/s20247312 - 19 Dec 2020
Cited by 10 | Viewed by 2624
Abstract
Perinatal stroke (PS), occurring between 20 weeks of gestation and 28 days of life, is a leading cause of hemiplegic cerebral palsy (HCP). Hallmarks of HCP are motor and sensory impairments on one side of the body—especially the arm and hand contralateral to [...] Read more.
Perinatal stroke (PS), occurring between 20 weeks of gestation and 28 days of life, is a leading cause of hemiplegic cerebral palsy (HCP). Hallmarks of HCP are motor and sensory impairments on one side of the body—especially the arm and hand contralateral to the stroke (involved side). HCP is diagnosed months or years after the original brain injury. One effective early intervention for this population is constraint-induced movement therapy (CIMT), where the uninvolved arm is constrained by a mitt or cast, and therapeutic activities are performed with the involved arm. In this preliminary investigation, we used 3D motion capture to measure the spatiotemporal characteristics of pre-reaching upper extremity movements and any changes that occurred when constraint was applied in a real-time laboratory simulation. Participants were N = 14 full-term infants: N = six infants with typical development; and N = eight infants with PS (N = three infants with PS were later diagnosed with cerebral palsy (CP)) followed longitudinally from 2 to 6 months of age. We aimed to evaluate the feasibility of using 3D motion capture to identify the differences in the spatiotemporal characteristics of the pre-reaching upper extremity movements between the diagnosis group, involved versus uninvolved side, and with versus and without constraint applied in real time. This would be an excellent application of wearable sensors, allowing some of these measurements to be taken in a clinical or home setting. Full article
Show Figures

Figure 1

17 pages, 5219 KiB  
Article
Lower Limb Exoskeleton Gait Planning Based on Crutch and Human-Machine Foot Combined Center of Pressure
by Wei Yang, Jiyu Zhang, Sheng Zhang and Canjun Yang
Sensors 2020, 20(24), 7216; https://doi.org/10.3390/s20247216 - 16 Dec 2020
Cited by 12 | Viewed by 5277
Abstract
With the help of wearable robotics, the lower limb exoskeleton becomes a promising solution for spinal cord injury (SCI) patients to recover lower body locomotion ability. However, fewer exoskeleton gait planning methods can meet the needs of patient in real time, e.g., stride [...] Read more.
With the help of wearable robotics, the lower limb exoskeleton becomes a promising solution for spinal cord injury (SCI) patients to recover lower body locomotion ability. However, fewer exoskeleton gait planning methods can meet the needs of patient in real time, e.g., stride length or step width, etc., which may lead to human-machine incoordination, limit comfort, and increase the risk of falling. This work presents a human-exoskeleton-crutch system with the center of pressure (CoP)-based gait planning method to enable the balance control during the exoskeleton-assisted walking with crutches. The CoP generated by crutches and human-machine feet makes it possible to obtain the overall stability conditions of the system in the process of exoskeleton-assisted quasi-static walking, and therefore, to determine the next stride length and ensure the balance of the next step. Thus, the exoskeleton gait is planned with the guidance of stride length. It is worth emphasizing that the nominal reference gait is adopted as a reference to ensure that the trajectory of the swing ankle mimics the reference one well. This gait planning method enables the patient to adaptively interact with the exoskeleton gait. The online gait planning walking tests with five healthy volunteers proved the method’s feasibility. Experimental results indicate that the algorithm can deal with the sensed signals and plan the landing point of the swing leg to ensure balanced and smooth walking. The results suggest that the method is an effective means to improve human–machine interaction. Additionally, it is meaningful for the further training of independent walking stability control in exoskeletons for SCI patients with less assistance of crutches. Full article
Show Figures

Figure 1

28 pages, 1291 KiB  
Article
Estimating Lower Limb Kinematics Using a Lie Group Constrained Extended Kalman Filter with a Reduced Wearable IMU Count and Distance Measurements
by Luke Wicent F. Sy, Nigel H. Lovell and Stephen J. Redmond
Sensors 2020, 20(23), 6829; https://doi.org/10.3390/s20236829 - 29 Nov 2020
Cited by 8 | Viewed by 3271
Abstract
Tracking the kinematics of human movement usually requires the use of equipment that constrains the user within a room (e.g., optical motion capture systems), or requires the use of a conspicuous body-worn measurement system (e.g., inertial measurement units (IMUs) attached to each body [...] Read more.
Tracking the kinematics of human movement usually requires the use of equipment that constrains the user within a room (e.g., optical motion capture systems), or requires the use of a conspicuous body-worn measurement system (e.g., inertial measurement units (IMUs) attached to each body segment). This paper presents a novel Lie group constrained extended Kalman filter to estimate lower limb kinematics using IMU and inter-IMU distance measurements in a reduced sensor count configuration. The algorithm iterates through the prediction (kinematic equations), measurement (pelvis height assumption/inter-IMU distance measurements, zero velocity update for feet/ankles, flat-floor assumption for feet/ankles, and covariance limiter), and constraint update (formulation of hinged knee joints and ball-and-socket hip joints). The knee and hip joint angle root-mean-square errors in the sagittal plane for straight walking were 7.6±2.6 and 6.6±2.7, respectively, while the correlation coefficients were 0.95±0.03 and 0.87±0.16, respectively. Furthermore, experiments using simulated inter-IMU distance measurements show that performance improved substantially for dynamic movements, even at large noise levels (σ=0.2 m). However, further validation is recommended with actual distance measurement sensors, such as ultra-wideband ranging sensors. Full article
Show Figures

Figure 1

15 pages, 3154 KiB  
Article
Classification of Neurological Patients to Identify Fallers Based on Spatial-Temporal Gait Characteristics Measured by a Wearable Device
by Yuhan Zhou, Rana Zia Ur Rehman, Clint Hansen, Walter Maetzler, Silvia Del Din, Lynn Rochester, Tibor Hortobágyi and Claudine J. C. Lamoth
Sensors 2020, 20(15), 4098; https://doi.org/10.3390/s20154098 - 23 Jul 2020
Cited by 19 | Viewed by 4202
Abstract
Neurological patients can have severe gait impairments that contribute to fall risks. Predicting falls from gait abnormalities could aid clinicians and patients mitigate fall risk. The aim of this study was to predict fall status from spatial-temporal gait characteristics measured by a wearable [...] Read more.
Neurological patients can have severe gait impairments that contribute to fall risks. Predicting falls from gait abnormalities could aid clinicians and patients mitigate fall risk. The aim of this study was to predict fall status from spatial-temporal gait characteristics measured by a wearable device in a heterogeneous population of neurological patients. Participants (n = 384, age 49–80 s) were recruited from a neurology ward of a University hospital. They walked 20 m at a comfortable speed (single task: ST) and while performing a dual task with a motor component (DT1) and a dual task with a cognitive component (DT2). Twenty-seven spatial-temporal gait variables were measured with wearable sensors placed at the lower back and both ankles. Partial least square discriminant analysis (PLS-DA) was then applied to classify fallers and non-fallers. The PLS-DA classification model performed well for all three gait tasks (ST, DT1, and DT2) with an evaluation of classification performance Area under the receiver operating characteristic Curve (AUC) of 0.7, 0.6 and 0.7, respectively. Fallers differed from non-fallers in their specific gait patterns. Results from this study improve our understanding of how falls risk-related gait impairments in neurological patients could aid the design of tailored fall-prevention interventions. Full article
Show Figures

Figure 1

17 pages, 1535 KiB  
Article
Machine Learning to Quantify Physical Activity in Children with Cerebral Palsy: Comparison of Group, Group-Personalized, and Fully-Personalized Activity Classification Models
by Matthew N. Ahmadi, Margaret E. O’Neil, Emmah Baque, Roslyn N. Boyd and Stewart G. Trost
Sensors 2020, 20(14), 3976; https://doi.org/10.3390/s20143976 - 17 Jul 2020
Cited by 19 | Viewed by 3592
Abstract
Pattern recognition methodologies, such as those utilizing machine learning (ML) approaches, have the potential to improve the accuracy and versatility of accelerometer-based assessments of physical activity (PA). Children with cerebral palsy (CP) exhibit significant heterogeneity in relation to impairment and activity limitations; however, [...] Read more.
Pattern recognition methodologies, such as those utilizing machine learning (ML) approaches, have the potential to improve the accuracy and versatility of accelerometer-based assessments of physical activity (PA). Children with cerebral palsy (CP) exhibit significant heterogeneity in relation to impairment and activity limitations; however, studies conducted to date have implemented “one-size fits all” group (G) models. Group-personalized (GP) models specific to the Gross Motor Function Classification (GMFCS) level and fully-personalized (FP) models trained on individual data may provide more accurate assessments of PA; however, these approaches have not been investigated in children with CP. In this study, 38 children classified at GMFCS I to III completed laboratory trials and a simulated free-living protocol while wearing an ActiGraph GT3X+ on the wrist, hip, and ankle. Activities were classified as sedentary, standing utilitarian movements, or walking. In the cross-validation, FP random forest classifiers (99.0–99.3%) exhibited a significantly higher accuracy than G (80.9–94.7%) and GP classifiers (78.7–94.1%), with the largest differential observed in children at GMFCS III. When evaluated under free-living conditions, all model types exhibited significant declines in accuracy, with FP models outperforming G and GP models in GMFCS levels I and II, but not III. Future studies should evaluate the comparative accuracy of personalized models trained on free-living accelerometer data. Full article
Show Figures

Figure 1

Review

Jump to: Editorial, Research, Other

30 pages, 742 KiB  
Review
Machine Learning Approaches for Activity Recognition and/or Activity Prediction in Locomotion Assistive Devices—A Systematic Review
by Floriant Labarrière, Elizabeth Thomas, Laurine Calistri, Virgil Optasanu, Mathieu Gueugnon, Paul Ornetti and Davy Laroche
Sensors 2020, 20(21), 6345; https://doi.org/10.3390/s20216345 - 06 Nov 2020
Cited by 27 | Viewed by 3764
Abstract
Locomotion assistive devices equipped with a microprocessor can potentially automatically adapt their behavior when the user is transitioning from one locomotion mode to another. Many developments in the field have come from machine learning driven controllers on locomotion assistive devices that recognize/predict the [...] Read more.
Locomotion assistive devices equipped with a microprocessor can potentially automatically adapt their behavior when the user is transitioning from one locomotion mode to another. Many developments in the field have come from machine learning driven controllers on locomotion assistive devices that recognize/predict the current locomotion mode or the upcoming one. This review synthesizes the machine learning algorithms designed to recognize or to predict a locomotion mode in order to automatically adapt the behavior of a locomotion assistive device. A systematic review was conducted on the Web of Science and MEDLINE databases (as well as in the retrieved papers) to identify articles published between 1 January 2000 to 31 July 2020. This systematic review is reported in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines and is registered on Prospero (CRD42020149352). Study characteristics, sensors and algorithms used, accuracy and robustness were also summarized. In total, 1343 records were identified and 58 studies were included in this review. The experimental condition which was most often investigated was level ground walking along with stair and ramp ascent/descent activities. The machine learning algorithms implemented in the included studies reached global mean accuracies of around 90%. However, the robustness of those algorithms seems to be more broadly evaluated, notably, in everyday life. We also propose some guidelines for homogenizing future reports. Full article
Show Figures

Figure 1

Other

28 pages, 2548 KiB  
Systematic Review
Wearable Sensor-Based Real-Time Gait Detection: A Systematic Review
by Hari Prasanth, Miroslav Caban, Urs Keller, Grégoire Courtine, Auke Ijspeert, Heike Vallery and Joachim von Zitzewitz
Sensors 2021, 21(8), 2727; https://doi.org/10.3390/s21082727 - 13 Apr 2021
Cited by 107 | Viewed by 10235
Abstract
Gait analysis has traditionally been carried out in a laboratory environment using expensive equipment, but, recently, reliable, affordable, and wearable sensors have enabled integration into clinical applications as well as use during activities of daily living. Real-time gait analysis is key to the [...] Read more.
Gait analysis has traditionally been carried out in a laboratory environment using expensive equipment, but, recently, reliable, affordable, and wearable sensors have enabled integration into clinical applications as well as use during activities of daily living. Real-time gait analysis is key to the development of gait rehabilitation techniques and assistive devices such as neuroprostheses. This article presents a systematic review of wearable sensors and techniques used in real-time gait analysis, and their application to pathological gait. From four major scientific databases, we identified 1262 articles of which 113 were analyzed in full-text. We found that heel strike and toe off are the most sought-after gait events. Inertial measurement units (IMU) are the most widely used wearable sensors and the shank and foot are the preferred placements. Insole pressure sensors are the most common sensors for ground-truth validation for IMU-based gait detection. Rule-based techniques relying on threshold or peak detection are the most widely used gait detection method. The heterogeneity of evaluation criteria prevented quantitative performance comparison of all methods. Although most studies predicted that the proposed methods would work on pathological gait, less than one third were validated on such data. Clinical applications of gait detection algorithms were considered, and we recommend a combination of IMU and rule-based methods as an optimal solution. Full article
Show Figures

Figure 1

11 pages, 1078 KiB  
Letter
Maximal Walking Distance in Persons with a Lower Limb Amputation
by Cheriel J. Hofstad, Kim T.J. Bongers, Mark Didden, René F. van Ee and Noël L.W. Keijsers
Sensors 2020, 20(23), 6770; https://doi.org/10.3390/s20236770 - 26 Nov 2020
Cited by 7 | Viewed by 2509
Abstract
The distance one can walk at a time could be considered an important functional outcome in people with a lower limb amputation. In clinical practice, walking distance in daily life is based on self-report (SIGAM mobility grade (Special Interest Group in Amputee Medicine)), [...] Read more.
The distance one can walk at a time could be considered an important functional outcome in people with a lower limb amputation. In clinical practice, walking distance in daily life is based on self-report (SIGAM mobility grade (Special Interest Group in Amputee Medicine)), which is known to overestimate physical activity. The aim of this study was to assess the number of consecutive steps and walking bouts in persons with a lower limb amputation, using an accelerometer sensor. The number of consecutive steps was related to their SIGAM mobility grade and to the consecutive steps of age-matched controls in daily life. Twenty subjects with a lower limb amputation and ten age-matched controls participated in the experiment for two consecutive days, in their own environment. Maximal number of consecutive steps and walking bouts were obtained by two accelerometers in the left and right trouser pocket, and one accelerometer on the sternum. In addition, the SIGAM mobility grade was determined and the 10 m walking test (10 MWT) was performed. The maximal number of consecutive steps and walking bouts were significantly smaller in persons with a lower limb amputation, compared to the control group (p < 0.001). Only 4 of the 20 persons with a lower limb amputation had a maximal number of consecutive steps in the range of the control group. Although the maximal covered distance was moderately correlated with the SIGAM mobility grade in participants with an amputation (r = 0.61), for 6 of them, the SIGAM mobility grade did not match with the maximal covered distance. The current study indicated that mobility was highly affected in most persons with an amputation and that the SIGAM mobility grade did not reflect what persons with a lower limb amputation actually do in daily life. Therefore, objective assessment of the maximal number of consecutive steps of maximal covered distance is recommended for clinical treatment. Full article
Show Figures

Figure 1

13 pages, 576 KiB  
Letter
Use of Functional Linear Models to Detect Associations between Characteristics of Walking and Continuous Responses Using Accelerometry Data
by William F. Fadel, Jacek K. Urbanek, Nancy W. Glynn and Jaroslaw Harezlak
Sensors 2020, 20(21), 6394; https://doi.org/10.3390/s20216394 - 09 Nov 2020
Cited by 1 | Viewed by 2115
Abstract
Various methods exist to measure physical activity. Subjective methods, such as diaries and surveys, are relatively inexpensive ways of measuring one’s physical activity; however, they are prone to measurement error and bias due to self-reporting. Wearable accelerometers offer a non-invasive and objective measure [...] Read more.
Various methods exist to measure physical activity. Subjective methods, such as diaries and surveys, are relatively inexpensive ways of measuring one’s physical activity; however, they are prone to measurement error and bias due to self-reporting. Wearable accelerometers offer a non-invasive and objective measure of one’s physical activity and are now widely used in observational studies. Accelerometers record high frequency data and each produce an unlabeled time series at the sub-second level. An important activity to identify from the data collected is walking, since it is often the only form of activity for certain populations. Currently, most methods use an activity summary which ignores the nuances of walking data. We propose methodology to model specific continuous responses with a functional linear model utilizing spectra obtained from the local fast Fourier transform (FFT) of walking as a predictor. Utilizing prior knowledge of the mechanics of walking, we incorporate this as additional information for the structure of our transformed walking spectra. The methods were applied to the in-the-laboratory data obtained from the Developmental Epidemiologic Cohort Study (DECOS). Full article
Show Figures

Figure 1

21 pages, 1318 KiB  
Perspective
Implementation of Wearable Sensing Technology for Movement: Pushing Forward into the Routine Physical Rehabilitation Care Field
by Catherine E. Lang, Jessica Barth, Carey L. Holleran, Jeff D. Konrad and Marghuretta D. Bland
Sensors 2020, 20(20), 5744; https://doi.org/10.3390/s20205744 - 10 Oct 2020
Cited by 43 | Viewed by 4582
Abstract
While the promise of wearable sensor technology to transform physical rehabilitation has been around for a number of years, the reality is that wearable sensor technology for the measurement of human movement has remained largely confined to rehabilitation research labs with limited ventures [...] Read more.
While the promise of wearable sensor technology to transform physical rehabilitation has been around for a number of years, the reality is that wearable sensor technology for the measurement of human movement has remained largely confined to rehabilitation research labs with limited ventures into clinical practice. The purposes of this paper are to: (1) discuss the major barriers in clinical practice and available wearable sensing technology; (2) propose benchmarks for wearable device systems that would make it feasible to implement them in clinical practice across the world and (3) evaluate a current wearable device system against the benchmarks as an example. If we can overcome the barriers and achieve the benchmarks collectively, the field of rehabilitation will move forward towards better movement interventions that produce improved function not just in the clinic or lab, but out in peoples’ homes and communities. Full article
Show Figures

Figure 1

14 pages, 2251 KiB  
Letter
Validity of a New 3-D Motion Analysis Tool for the Assessment of Knee, Hip and Spine Joint Angles during the Single Leg Squat
by Igor Tak, Willem-Paul Wiertz, Maarten Barendrecht and Rob Langhout
Sensors 2020, 20(16), 4539; https://doi.org/10.3390/s20164539 - 13 Aug 2020
Cited by 12 | Viewed by 4902
Abstract
Aim: Study concurrent validity of a new sensor-based 3D motion capture (MoCap) tool to register knee, hip and spine joint angles during the single leg squat. Design: Cross-sectional. Setting: University laboratory. Participants: Forty-four physically active (Tegner ≥ 5) subjects (age 22.8 (±3.3)) Main [...] Read more.
Aim: Study concurrent validity of a new sensor-based 3D motion capture (MoCap) tool to register knee, hip and spine joint angles during the single leg squat. Design: Cross-sectional. Setting: University laboratory. Participants: Forty-four physically active (Tegner ≥ 5) subjects (age 22.8 (±3.3)) Main outcome measures: Sagittal and frontal plane trunk, hip and knee angles at peak knee flexion. The sensor-based system consisted of 4 active (triaxial accelerometric, gyroscopic and geomagnetic) sensors wirelessly connected with an iPad. A conventional passive tracking 3D MoCap (OptiTrack) system served as gold standard. Results: All sagittal plane measurement correlations observed were very strong for the knee and hip (r = 0.929–0.988, p < 0.001). For sagittal plane spine assessment, the correlations were moderate (r = 0.708–0.728, p < 0.001). Frontal plane measurement correlations were moderate in size for the hip (ρ = 0.646–0.818, p < 0.001) and spine (ρ = 0.613–0.827, p < 0.001). Conclusions: The 3-D MoCap tool has good to excellent criterion validity for sagittal and frontal plane angles occurring in the knee, hip and spine during the single leg squat. This allows bringing this type of easily accessible MoCap technology outside laboratory settings. Full article
Show Figures

Figure 1

17 pages, 8481 KiB  
Letter
Estimation of Relative Hand-Finger Orientation Using a Small IMU Configuration
by Zhicheng Yang, Bert-Jan F. van Beijnum, Bin Li, Shenggang Yan and Peter H. Veltink
Sensors 2020, 20(14), 4008; https://doi.org/10.3390/s20144008 - 19 Jul 2020
Cited by 6 | Viewed by 3183
Abstract
Relative orientation estimation between the hand and its fingers is important in many applications, such as virtual reality (VR), augmented reality (AR) and rehabilitation. It is still quite a big challenge to do the estimation by only exploiting inertial measurement units (IMUs) because [...] Read more.
Relative orientation estimation between the hand and its fingers is important in many applications, such as virtual reality (VR), augmented reality (AR) and rehabilitation. It is still quite a big challenge to do the estimation by only exploiting inertial measurement units (IMUs) because of the integration drift that occurs in most approaches. When the hand is functionally used, there are many instances in which hand and finger tips move together, experiencing almost the same angular velocities, and in some of these cases, almost the same accelerations are measured in different 3D coordinate systems. Therefore, we hypothesize that relative orientations between the hand and the finger tips can be adequately estimated using 3D IMUs during such designated events (DEs) and in between these events. We fused this extra information from the DEs and IMU data with an extended Kalman filter (EKF). Our results show that errors in relative orientation can be smaller than five degrees if DEs are constantly present and the linear and angular movements of the whole hand are adequately rich. When the DEs are partially available in a functional water-drinking task, the orientation error is smaller than 10 degrees. Full article
Show Figures

Figure 1

Back to TopTop