Fall risk assessment in the wild: A critical examination of wearable sensor use in free-living conditions

BackgroundDespite advances in laboratory-based supervised fall risk assessment methods (FRAs), falls still remain a major public health problem. This can be due to the alteration of behavior in laboratory due to the awareness of being observed (i.e., Hawthorne effect), the multifactorial complex etiology of falls, and our limited understanding of human behaviour in natural environments, or in the’ wild’. To address these imitations, a growing body of literature has focused on free-living wearable-sensor-based FRAs. The objective of this narrative literature review is to discuss papers investigating natural data collected by wearable sensors for a duration of at least 24 h to identify fall-prone older adults.MethodsDatabases (Scopus, PubMed and Google Scholar) were searched for studies based on a rigorous search strategy.ResultsTwenty-four journal papers were selected, in which inertial sensors were the only wearable system employed for FRA in the wild. Gait was the most-investigated activity; but sitting, standing, lying, transitions and gait events, such as turns and missteps, were also explored. A multitude of free-living fall predictors (FLFPs), e.g., the quantity of daily steps, were extracted from activity bouts and events. FLFPs were further categorized into discrete domains (e.g., pace, complexity) defined by conceptual or data-driven models. Heterogeneity was found within the reviewed studies, which includes variance in: terminology (e.g., quantity vs macro), hyperparameters to define/estimate FLFPs, models and domains, and data processing approaches (e.g., the cut-off thresholds to define an ambulatory bout). These inconsistencies led to different results for similar FLFPs, limiting the ability to interpret and compare the evidence.ConclusionFree-living FRA is a promising avenue for fall prevention. Achieving a harmonized model is necessary to systematically address the inconsistencies in the field and identify FLFPs with the highest predictive values for falls to eventually address intervention programs and fall prevention.     

Normative database of postural sway measures using inertial sensors in typically developing children and young adults

ObjectiveReference values utilizing the APDM MobilityLab® inertial sensor system have not been established in children and young adults ages 5−30. These values are necessary for clinicians and researchers to compare to children with balance impairments.MethodsA group of 144 typically developing children and young adults from age 5–30 years completed the instrumented SWAY test during 6 test conditions: normal stance, firm surface, eyes open (EO) and closed (EC); normal stance, foam surface, EO and EC; and tandem stance, firm surface, EO and EC. Selected variables for normative outcomes included total sway area, and the mean, sagittal and coronal values for RMS sway, jerk, sway velocity and path length. Sex differences were examined within age groups via t tests. The effect of age on postural sway variables was analyzed using a one-way ANOVA for the mean values of total sway area, RMS sway, velocity and jerk, followed by post-hoc pairwise comparisons.ResultsAll sway parameters decreased significantly with age (p < 0.0001). Adult-like total sway area and jerk were achieved by ages 9−10 except for jerk during EC on foam. RMS sway and sway velocity reached adult levels by ages 11−13 during all EO and tandem stance conditions, and 14−21 with EC during normal stance on firm and foam surfaces for RMS sway and EC on firm surfaces for velocity. Females ages 5−6 performed more poorly during EO firm and EC foam for certain variables, but better during EO tandem and females ages 7–13 outperformed males when sex differences were found.SignificanceThese reference values can now be used by clinicians and researchers to evaluate abnormal postural sway and response to interventions in children and young adults.     

Swaying to the complex motion of a visual target affects postural sway variability

BackgroundVoluntary shifting body weight in the anteroposterior direction is an important element of daily life activities, such as rising from a chair or initiating a step. In order to accommodate the daily-life challenges of such tasks, voluntary postural sway needs to be flexible and variable.Research questionIn this study we asked how whole-body tracking of a complex visual target motion with the concurrent provision of feedback modulates the variability of voluntary sway.MethodsTwenty young adults (age: 27.10 ± 9.15years, height: 170.73 ± 9.40 cm, mass: 62.84 ± 11.48 kg) performed 132 cycles of voluntary antero-posterior sway, on a force platform, under two conditions: a) self-paced sway and b) swaying while tracking the complex motion of a visual target. Magnitude and temporal structure of variability of postural sway were investigated with the Coefficient of Variance (CoV) and the fractal exponent α, respectively. This analysis was performed for sway cycle duration, amplitude and velocity. The cross-correlation function between the target and sway cycle parameters was computed as a measure of visuo-postural coupling.ResultsThe CoV of sway cycle amplitude, duration and velocity increased during active tracking of the complex target. Fractal exponent α increased for sway cycle amplitude but decreased for cycle duration and remained unchanged for sway velocity. The cross-correlation function revealed a consistent peak at lag+1 indicating an asynchrony between the target and sway cycle duration, while the peak cross-correlation for cycle amplitude was noted at lag 0.SignificanceSwaying to the complex motion of a visual target improves the variability of sway cycle amplitude, at the cost of cycle duration. This is associated with a more synchronous spatial than temporal coupling to the visual target motion. This knowledge could inform the design of postural tracking paradigms as appropriate exercise interventions, for improving voluntary sway in populations with reduced limits of stability (i.e. older adults).     

Measuring gait kinematics in patients with severe hip osteoarthritis using wearable sensors

BackgroundThe popularity of inertial sensors in gait analysis is steadily rising. To date, an application of a wearable inertial sensor system for assessing gait in hip osteoarthritis (OA) has not been reported.Research question: Can the known kinematic differences between patients with hip OA and asymptomatic control subjects be measured using the inertial sensor system RehaGait®?MethodsThe patients group consisted of 22 patients with unilateral hip OA scheduled for total hip replacement. Forty-five age matched healthy control subjects served as control group. All subjects walked for a distance of 20 m at their self-selected speed. Spatiotemporal parameters and sagittal kinematics at the hip, knee, and ankle including range of motion (ROM) were measured using the RehaGait® system.ResultsPatients with hip OA walked at a slower walking speed (−0.18 m/s, P < 0.001) and with shorter stride length (−0.16 m, P < 0.001), smaller hip ROM during stance (−11.6°, P < 0.001) and swing (−11.3°, P < 0.001) and smaller knee ROM during terminal stance and swing (−9.0° and−11.5°, P < 0.001). Patients had a smaller hip ROM during stance and swing and smaller knee ROM during terminal stance and swing in the affected compared to the unaffected side (P < 0.001).SignificanceThe differences in spatiotemporal and kinematic gait parameters between patients with hip OA and age matched control subjects assessed using the inertial sensor system agree with those documented for camera-based systems. Hence, the RehaGait® system can measure gait kinematics characteristic for hip OA, and its use in daily clinical practice is feasible.     

Time dependent structure of postural sway in individuals with multiple sclerosis

Non-linear analyses, which examine the time dependent structure of physiological output have been found to be able to detect subtle differences in postural control between pathological groups and healthy controls while traditional linear parameters do not. This investigation examines whether a specific non-linear metric, approximate entropy, may provide a novel biomarker for balance impairment in individuals with multiple sclerosis (MS) who have normal sway. This analysis included a sample of 30 individuals with MS with normal postural sway and 36 controls. Participants stood on a force platform for two trials of 30 s with eyes open. Postural control was indexed by sway area, mean velocity along the antero-posterior (AP) and mediolateral (ML) axis. The time dependent structure of the COP along the AP and ML axes was indexed with approximate entropy (ApEn_AP; ApEn_ML). T-tests and Mann–Whitney U tests were utilized to analyze differences between groups. Per design there were no differences in sway area between the MS and control groups. Additionally, there were no differences in sway velocity. The MS group had lower ApEn_ML values compared to the control group (U = 376, p = .026). The results indicate that individuals with MS who have normal sway area had greater time dependent structure in ML sway. This investigation highlights the utility of non-linear analyses when assessing balance impairment in MS samples that present with minimal sway area.     

Trunk sway in patients with and without, mild traumatic brain injury after whiplash injury

Objective

This study assessed the addition effect of mild traumatic brain injury (MTBI) on the balance control of patients who simultaneously suffered a whiplash associated disorder (WAD).

Background

Dizziness is common in patients suffering from whiplash injury with or without a MTBI, but data is lacking about the additional balance problems and dizziness caused by MTBI.

Methods

44 patients with WAD and MTBI and 36 WAD patients without MTBI participated in the study. A dizziness handicap index (DHI) was used to quantify self-perceived handicap. Balance control was assessed using measures of trunk sway for a battery of stance and gait tests.

Results

Patients with WAD and MTBI perceived significantly higher dizziness and unsteadiness (higher score in DHI Emotional category) and had greater trunk sway than WAD patients without MTBI for stance tasks and complex gait tasks (e.g. walking up and down stairs). Both groups had greater sway than controls for these tasks. Both groups of patients showed equal reductions in trunk sway with respect to controls for simple gait tasks (e.g. walking while rotating the head).

Conclusions

A similar pattern of balance impairment was present in patients with whiplash injury with and without MTBI. However, the impairment was greater for stance and complex gait tasks in WAD patients with MTBI.     

Immediate effects of real-time postural biofeedback on spinal posture,muscle activity,and perceived pain severity in adults with neck pain

BackgroundPrevious studies have investigated various types of postural biofeedback devices on different body regions to improve posture; however, they focused only on healthy adults without a history of chronic musculoskeletal disorders. In addition, those postural biofeedback devices used in previous studies are often designed for experimental research. The designs are usually bulky with many wires, which is not practical for everyday use.Research questionThe aim of this study was to determine the immediate effect of a commercially available real-time postural biofeedback device on spinal posture, muscle activity, and perceived pain severity in adults with neck pain.Methods21 adults who had chronic or recurrent nonspecific neck pain for more than 3 months and whose pain was induced or aggravated by prolonged computer work were enrolled in this study. Spinal posture (head tilt, neck flexion, cervical and thoracic angles), muscle activity (cervical erector spinae, upper trapezius, and thoracic erector spinae), and self-reported neck and shoulder pain were measured during computer typing tasks, with and without biofeedback.ResultsCompared with the non-biofeedback condition, the biofeedback condition significantly decreased neck flexion, upper cervical, and lower thoracic angles and lowered the activity of the cervical erector spinae. Self-reported neck pain was not influenced by the application of biofeedback, but significantly increased over the 1-hour typing task.SignificanceThe application of a commercially available wearable real-time biofeedback device improves sitting posture and reduces muscular activity in adults with nonspecific neck pain during computer work. Future studies should examine the long-term effects of wearable real-time postural biofeedback devices for prevention and management of neck pain.     

Smartphone accelerometry to assess postural control in individuals with multiple sclerosis

BackgroundFalls are a major health concern for people with Multiple Sclerosis (pwMS), and impaired postural control is an important predictor of falls. Lab-based technology to measure posture is precise but expensive, and clinical tests may not capture underlying impairments. An alternative solution is to leverage smartphone accelerometry as it is affordable, ubiquitous, and portable.Research question: Can smartphone accelerometry measure postural control compared to a force plate and research grade accelerometer in pwMS, and can smartphone accelerometry discriminate between assisted device and non-assisted device users?Methods27 pwMS (12 assisted device users, 15 non-assisted device users) stood on a force plate while holding a smartphone with an attached research grade accelerometer against their chest. Participants performed two, 30 s trials of: eyes open, eyes closed, semi-tandem, tandem, and single leg. Acceleration and center of pressure were extracted, and Root Mean Square (RMS) and 95 % confidence ellipse were calculated. Spearman’s correlations were performed, and receiving operating characteristic (ROC) curves and the Area Under the Curve (AUC) were calculated.ResultsThere were moderate to high correlations between the smartphone and accelerometer for RMS (ρ = 0.85 – 1.0; p = 0.001 – <0.001) and 95 % area ellipse (ρ = 0.92 – 0.99; p = <0.001). There were weak to moderate correlations between the smartphone and force plate for RMS (ρ = 0.38 – 0.92; p = 0.06 – <0.001) and 95 % area ellipse (ρ = 0.69 – 0.90 p = 0.002 – <0.001). To discriminate between assisted device usage, ROC curves for smartphone outputs were constructed, the AUC was high and statistically significant (p < 0.001 – 0.02).SignificanceThere is potential to leverage smartphone accelerometery to measure postural control in pwMS. These finding provide preliminary results to support the development of a mobile health application to measure fall risk for pwMS.     

Validity and repeatability of inertial measurement units for measuring gait parameters

Inertial measurement units (IMUs) are small wearable sensors that have tremendous potential to be applied to clinical gait analysis. They allow objective evaluation of gait and movement disorders outside the clinic and research laboratory, and permit evaluation on large numbers of steps. However, repeatability and validity data of these systems are sparse for gait metrics. The purpose of this study was to determine the validity and between-day repeatability of spatiotemporal metrics (gait speed, stance percent, swing percent, gait cycle time, stride length, cadence, and step duration) as measured with the APDM Opal IMUs and Mobility Lab system. We collected data on 39 healthy subjects. Subjects were tested over two days while walking on a standard treadmill, split-belt treadmill, or overground, with IMUs placed in two locations: both feet and both ankles. The spatiotemporal measurements taken with the IMU system were validated against data from an instrumented treadmill, or using standard clinical procedures. Repeatability and minimally detectable change (MDC) of the system was calculated between days. IMUs displayed high to moderate validity when measuring most of the gait metrics tested. Additionally, these measurements appear to be repeatable when used on the treadmill and overground. The foot configuration of the IMUs appeared to better measure gait parameters; however, both the foot and ankle configurations demonstrated good repeatability. In conclusion, the IMU system in this study appears to be both accurate and repeatable for measuring spatiotemporal gait parameters in healthy young adults.     

Individuals with chronic ankle instability exhibit decreased postural sway while kicking in a single-leg stance

Individuals with chronic ankle instability (CAI) usually experience deficits in balance control, which increase displacement in the body's center of pressure (COP) when they balance on a single leg. Little is known, however, about whether or not these individuals use the strategies of postural adjustment properly, especially during functional tasks that may predispose them to ankle sprain. The aim of this study was to investigate anticipatory (APA) and compensatory (CPA) postural adjustments in individuals with and without CAI as they kick a ball while standing in a single-leg stance with their ankle in neutral and supinated positions. COP displacements were calculated and their magnitudes (range) analyzed during APA and CPA intervals and over the duration of the whole task, represented by the COP area of sway and mean velocity. The CAI group exhibited a significant decrease in CPA and area of sway over the whole task, relative to controls. These results suggest that the decreased balance sway could be caused by the need for further stabilization of the ankle in more unstable postures to prevent recurrent sprain. Our findings could help clinicians to better understand the strategies of postural adjustments in individuals with CAI, and may assist and motivate new investigations into balance control interventions in such individuals, as well as proactively address recurrent ankle sprain conditions.     

Lower limb joint-specific contributions to standing postural sway in persons with unilateral lower limb loss

BackgroundIndividuals with lower limb loss are at an increased risk for falls, likely due to impaired balance control. Standing balance is typically explained by double- or single-inverted pendulum models of the hip and/or ankle, neglecting the knee joint. However, recent work suggests knee joint motion contributes toward stabilizing center-of-mass kinematics during standing balance.Research QuestionTo what extent do hip, knee, and ankle joint motions contribute to postural sway in standing among individuals with lower limb loss?MethodsForty-two individuals (25 m/17f) with unilateral lower limb loss (30 transtibial, 12 transfemoral) stood quietly with eyes open and eyes closed, for 30 s each, while wearing accelerometers on the pelvis, thigh, shank, and foot. Triaxial inertial measurement units were transformed to inertial anterior-posterior components and sway parameters were computed: ellipse area, root-mean-square, and jerk. A state-space model with a Kalman filter calculated hip, knee, and ankle joint flexion-extension angles and ranges of motion. Multiple linear regression predicted postural sway parameters from intact limb joint ranges of motion, with BMI as a covariate (p < 0.05).ResultsWith eyes open, intact limb hip flexion predicted larger sway ellipse area, whereas hip flexion and knee extension predicted larger sway root-mean-square, and hip flexion, knee extension, and ankle plantarflexion predicted larger sway jerk. With eyes closed, intact limb hip flexion remained the predictor of sway ellipse area; no other joint motions influenced sway parameters in this condition.SignificanceHip, knee, and ankle motions influence postural sway during standing balance among individuals with lower limb loss. Specifically, increasing intact-side hip flexion, knee extension, and ankle plantarflexion motion increased postural sway. With vision removed, a re-weighting of lower limb joint sensory mechanisms may control postural sway, such that increasing sway may be regulated by proximal coordination strategies and vestibular responses, with implications for fall risk.     

Ambulatory running speed estimation using an inertial sensor

Techniques have been developed to analyze walking gait using accelerometer and gyroscope data from miniature inertial measurement units (IMU), but few attempts have been made to use similar approaches for running gait. The purpose of this study was to develop an algorithm capable of estimating running speed using a single shank-mounted IMU. Raw acceleration and angular velocity were recorded from an IMU sensor attached on the lateral side of the shank in the sagittal plane and a method of reliably detecting the shank vertical and the minimal shank velocity gait event was used to segment a running sequence into individual strides. Through integration, the orientation of the shank segment was determined and an estimate of stride-by-stride running speed was calculated by integrating the acceleration data. The algorithm was verified using data collected from a group of seven volunteers running on a treadmill at speeds between 2.50 m/s and 3.50 m/s. Over the entire speed range, the estimation results gave a percentage root mean square error (%RMSE) of approximately 4.10%. With the accurate estimation capability and portability, the use of the proposed system in outdoor running gait analysis is promising.     

Lower body kinematics estimation from wearable sensors for walking and running: A deep learning approach

Background:Inertial measurement units (IMUs) are promising tools for collecting human movement data. Model-based filtering approaches (e.g. Extended Kalman Filter) have been proposed to estimate joint angles from IMUs data but little is known about the potential of data-driven approaches.Research question:Can deep learning models accurately predict lower limb joint angles from IMU data during gait?Methods:Lower-limb kinematic data were simultaneously measured with a marker-based motion capture system and running leggings with 5 integrated IMUs measuring acceleration and angular velocity at the pelvis, thighs and tibias. Data acquisition was performed on 27 participants (26.5 (3.9) years, 1.75 (0.07) m, 68.3 (10.0) kg) while walking at 4 and 6 km/h and running at 8, 10, 12 and 14 km/h on a treadmill. The model input consists of raw IMU data, while the output estimates the joint angles of the lower body. The model was trained with a nested $k$-fold cross-validation and tested considering a user-independent approach. Mean error (ME), mean absolute error (MAE) and Pearson correlation coefficient ($r$) were computed between the ground truth and predicted joint angles.Results:MAE for the DOFs ranged from 2.2(0.9) to 5.1(2.7)$°$ with an average of 3.6(2.1)$°$. $r$ ranged from 0.67(0.23) to 0.99(0.01) with moderate correlation ($0.4\le r<0.7$) was found for the hip right rotation and lumbar extension, strong correlation ($0.7\le r<0.9$) was found for the hip left rotation and ankle right/left inversion while all other DOFs showed very strong correlation ($r\ge 0.9$).Significance:The proposed model can reliably predict joint kinematics for walking, running and gait transitions without specific knowledge about the body characteristics of the wearer, or the position and orientation of the IMU relative to the attached segment. These results have been validated with treadmill gait, and have not yet been confirmed for gait in other settings.     

Effects of age,sex and task on postural sway during quiet stance

BackgroundPostural sway during quiet standing has been shown as a useful task to assess risk of falling in older adults. While the risk of falling is consistently reported to be higher in older females than males, the sex-related differences in postural sway are not consistent across the studies.Research questionWhat are the effects of age and sex on postural sway during quiet standing during different stance conditions?MethodsWe examined the effects of age (40 young and 34 older adults), sex (37 males and 37 females), and their interaction on the postural sway during different stance conditions. We compared the center of pressure (CoP) velocity, amplitude and frequency during parallel (eyes open and eyes closed) and semi-tandem (eyes open) stances.ResultsOur results suggest that postural sway is similar between sexes in young participants, while older males exhibit larger postural sway than older female participants (10/21 outcomes). Older female participants exhibited lower CoP amplitude (but larger total and anterior-posterior CoP velocity) compared to young female participants. We also found that the increase in the postural sway with increasing task difficulty is more pronounced in older vs. young adults.SignificanceThis study shows that ageing-related changes in postural sway are sex- and task-specific. Researchers and clinicians need to be aware of these effects when comparing groups or monitoring changes in time.     

Effects of orthopedic insoles on postural balance in patients with chronic stroke: A randomized crossover study

BackgroundOrthopedic insoles (OIs) with medial arch support and heel cushion are widely used to manage lower extremity injuries, but their effects on postural balance in patients with chronic stroke have not been adequately explored.MethodsDesign: Double-blinded, sham-controlled, randomized crossover trial. Participants: A total of 32 ambulatory patients (20 men and 12 women, aged between 30 and 76 years) with more than 6 months since stroke onset. Interventions: All participants received one assessment session wearing OIs and one session wearing sham insole (SI) in a random order with a 1-day interval. Outcomes: Our primary outcome was the Berg Balance Scale score. Secondary outcomes included the Functional Reach Test, Timed Up and Go test, and computerized posturography. All were performed in both sessions. Subgroup analyses regarding demographic and functional variables were conducted to identify potential responders.ResultsSignificant between-insole differences favoring OIs were seen in all clinical tests (P < 0.05), but were seen only in the static medial-lateral sway in computerized posturography assessment (P = 0.04). An approximate 2-point difference in the BBS score favoring OIs was observed in all subgroups, not reaching the minimal clinically important difference.ConclusionThe use of OIs generated small but significant positive effects on improving postural balance among patients with chronic stroke. Additional biomechanical and clinical studies are required to evaluate their potential for routine clinical use.Trial registrationNCT03194282     

Multiscale entropy identifies differences in complexity in postural control in women with multiple sclerosis

Loss of postural center-of-pressure complexity (COP complexity) has been associated with reduced adaptability that accompanies disease and aging. The aim of this study was to identify if COP complexity is reduced: (1) in those with Multiple Sclerosis (MS) compared to controls; (2) when vision is limited compared to remaining intact; and (3) during more demanding postural conditions compared to quiet standing. Additionally, we explored the relationship between the COP complexity and disease severity, fatigue, cutaneous sensation and central motor drive. Twelve women with MS and 12 age-matched controls were tested under quiet standing and postural maximal lean conditions with normal and limited vision. The key dependent variable was the complexity index (C_I) of the center of pressure. We observed a lower C_I in the MS group compared to controls in both anterior-posterior (AP) and medio-lateral (ML) directions (p's < 0.002), during the performance of maximal self-regulated leans (AP: p < 0.001; ML: p = 0.018), and under limited vision (AP: p = 0.001; ML: p = 0.006). No group-by-vision interaction (p > 0.05) was observed, indicating that limiting vision did not impact COP complexity differently in the two groups. Decreased cutaneous sensitivity was associated with lower C_I values in the AP direction among those with MS (r² = 0.57); all other measures did not exhibit significant relationships. The findings reported here suggest that (1) MS is associated with diminished COP complexity under both normal and challenging postures, and (2) complexity is strongly correlated with cutaneous sensitivity, suggesting the unique contribution of impaired somatosensation on postural control deficits in persons with MS.     

Increased postural sway in persons with multiple sclerosis during short-term exposure to warm ambient temperatures

BackgroundMultiple sclerosis (MS) is a neurological disease marked by demyelination and axonal loss. Individuals with MS experience increases in clinical signs and symptoms during heat exposure.ObjectiveTo test the hypothesis that moderate heat exposure adversely affects postural sway in individuals with MS.MethodsTen individuals with relapsing-remitting MS (50 ± 8y) and nine controls (47 ± 10y) were examined under a Thermal and a Time Control trial. Following a 30 min thermoneutral baseline (25 °C, 30% relative humidity (RH)), stand tests randomized with eyes open and closed, were performed. For Thermal, subjects were first exposed to 60 min of heating (40 °C, 30%RH) followed by 60 min of cooling (20 °C, 30%RH). For Time Control, subjects remained in a thermoneutral environment throughout. Stand tests were repeated at consistent times in both trials.ResultsNo difference in skin and core temperatures between groups were observed for any trial (P > 0.05). During heating, postural sway was higher in MS relative to control subjects (eyes open, P = 0.03; eyes closed, P = 0.011). No differences in postural sway, regardless of eye status, were observed during the Time Control trial for either group (P > 0.05).ConclusionThese data demonstrate that exposure to a moderate heating environment increases postural sway in patients with MS.     

Objective assessment of movement competence in children using wearable sensors: An instrumented version of the TGMD-2 locomotor subtest

Movement competence (MC) is defined as the development of sufficient skill to assure successful performance in different physical activities. Monitoring children MC during maturation is fundamental to detect early minor delays and define effective intervention. To this purpose, several MC assessment batteries are available. When evaluating movement strategies, with the aim of identifying specific skill components that may need improving, widespread MC assessment is limited by high time consumption for scoring and the need for trained operators to ensure reliability. This work aims to facilitate and support the assessment by designing, implementing and validating an instrumented version of the TGMD-2 locomotor subtest based on Inertial Measurement Units (IMUs) to quantify MC in children rapidly and objectively. 45 typically developing children, aged 6–10, performed the TGMD-2 locomotor subtest (six skills). During the tests, children wore five IMUs mounted on lower back, on ankles and on wrists. Sensor and video recordings of the tests were collected. Three expert evaluators performed the standard assessment of TGMD-2. Using theoretical and modelling approaches, algorithms were implemented to automatically score children tests based on IMUs’ data. The automatic assessment, compared to the standard one, showed an agreement higher than 87% on average on the entire group for each skill and a reduction of time for scoring from 15 to 2 min per participant. Results support the use of IMUs for MC assessment: this approach will allow improving the usability of MC assessment, supporting objectively evaluator decisions and reducing time requirement for the evaluation of large groups.     

Measurement of uni-planar and sport specific trunk motion using magneto-inertial measurement units: The concurrent validity of Noraxon and Xsens systems relative to a retro-reflective system

BackgroundThere is a range of magneto-inertial measurement unit (MIMU) systems commercially available, however sensor specifications and fusion methods vary considerably between manufacturers. Such variability can influence the concurrent validity of MIMUs relative to reference standard measurement devices. Different MIMUs have been compared during static or low-velocity conditions, with higher-velocity movements assessed in robotic-based studies. However, there is a need for the concurrent validity of higher-velocity movements to be established in human-based studies.Research questionThis study aimed to assess the concurrent validity of two commercial MIMU systems (Noraxon and Xsens), relative to a ‘gold-standard’ retro-reflective motion capture system, when measuring trunk angles during uni-planar range of motion (ROM) and cricket bowling, which involves high-speed, multi-planar movements.MethodsFor this criterion-based validity study, both MIMU systems incorporated comparable sensor specifications and employed Kalman filter sensor fusion algorithms. The MIMU based angles were compared with angles derived from concurrently captured three-dimensional retro-reflective data for 10 fast-medium bowlers. Statistical parametric mapping and root mean squared differences (RMSD) were computed for both MIMU systems.ResultsOne-dimensional statistical parametric mapping showed no significant differences for angles from both MIMU systems when compared with retro-reflective based angle outputs. The MIMU systems produced ROM RMSDs between 1.4 ± 1.0° and 2.6 ± 1.5°. One system displayed RMSDs between 4.6 ± 1.4° and 7.4 ± 1.9° during bowling, indicating functionally relevant differences to retro-reflective derived angles. There were some small but statistically significant differences in RMSDs between the MIMU systems.SignificanceMIMU-based angle accuracy is poorer during high-speed, multi-planar movement than uni-planar tasks. Comparable MIMU systems can produce varying measurements during ROM and bowling tasks. It is likely that varying sample rates and sensor fusion algorithm parameters contributed to the differences.