Assessment and validation of a simple automated method for the detection of gait events and intervals

Use of a Triaxial Accelerometer to Measure Changes in Gait Sway and Related Motor Function after Corrective Spinal Fusion Surgery for Adult Spinal Deformity

Background: Adult spinal deformity is a complex condition that causes lower back pain, causing spinal imbalance and discomfort in activities of daily life. After corrective spinal surgery, patients' gait and balance abilities might not revert to normalcy and they might be at increased risk of falling. Therefore, early evaluation of such a risk is imperative to prevent further complications such as a fall, or even worse, fractures in post-surgery ASD patients. However, there has been no report of an investigation of such early changes in gait sway before and after ASD surgery. This is a prospective to investigate changes in gait sway before and following ASD surgery, using accelerometers, and also to examine motor function related to postoperative gait sway. Methods: Twenty patients were included who underwent corrective surgery as treatment for ASD, from October 2019 to January 2023. Measurement parameters included a 10 m walking test and the timed up-and-go test (TUG), gait sway was evaluated using accelerometers (root mean square; RMS), and hip flexion and knee extension muscle strength were tested. RMS included RMS vertical: RMSV; RMS anterior posterior: RMSAP; RMS medial lateral: RMSML. The radiographic spinopelvic parameters were also evaluated preoperatively and postoperatively. p < 0.05 was noted as remarkably significant. Results: Preoperative and postoperative RMSV were 1.07 ± 0.6 and 1.31 ± 0.8, respectively (p < 0.05). RMSML significantly decreased from 0.33 ± 0.2 to 0.19 ± 0.1 postoperatively (p < 0.01). However, RMSAP did not change postoperatively (0.20 ± 0.2 vs. 0.14 ± 0.1, p > 0.05). Patients' one-month postoperative hip flexor muscle strength became significantly weaker (0.16 ± 0.04 vs. 0.10 ± 0.03 kgf/kg, p = 0.002), but TUG was maintained (11.6 ± 4.2 vs. 11.7 s, p = 0.305). RMSV was negatively correlated with quadriceps muscle strength and positively with TUG. RMSAP was negatively correlated with quadriceps muscle strength. All spinopelvic parameters became normal range after surgery. Conclusions: After corrective spinal fusion for ASD patients, the gait pattern improved significantly. Iliopsoas (hip flexor) and quadriceps femoris (knee extensor) muscles may play important roles for gait anterolateral and vertical swing, respectively.

Spatiotemporal walking performance in different settings: effects of walking speed and sex

Background

Understanding the factors that influence walking is important as quantitative walking assessments have potential to inform health risk assessments. Wearable technology innovation has enabled quantitative walking assessments to be conducted in different settings. Understanding how different settings influence quantitative walking performance is required to better utilize the health-related potential of quantitative walking assessments.

Research question

How does spatiotemporal walking performance differ during walking in different settings at different speeds for young adults?

Methods

Forty-two young adults [21 male (23 ± 4 years), 21 female (24 ± 5 years)] walked in two laboratory settings (overground, treadmill) and three non-laboratory settings (hallway, indoor open, outdoor pathway) at three self-selected speeds (slow, preferred, fast) following verbal instructions. Six walking trials of each condition (10 m in laboratory overground, 20 m in other settings) were completed. Participants wore 17 inertial sensors (Xsens Awinda, Movella, Henderson, NV) and spatiotemporal parameters were computed from sensor-derived kinematics. Setting × speed × sex repeated measures analysis of variance were used for statistical analysis.

Results

Regardless of the speed condition, participants walked faster overground when compared to while on the treadmill and walked faster in the indoor open and outdoor pathway settings when compared to the laboratory overground setting. At slow speeds, participants also walked faster in the hallway when compared to the laboratory overground setting. Females had greater cadence when compared to males, independent of settings and speed conditions.

Significance

Particularly at slow speeds, spatiotemporal walking performance was different between the settings, suggesting that setting characteristics such as walkway boundary definition may significantly influence spatiotemporal walking performance.

Accuracy validation of a wearable IMU-based gait analysis in healthy female

OBJECTIVE

The aim of this study was to assess the accuracy and test-retest reliability of a wearable inertial measurement unit (IMU) system for gait analysis in healthy female compared to a gold-standard optoelectronic motion capture (OMC) system.

METHODS

In our study, we collected data from 5 healthy young females. Participants were attached with markers from both the OMC system and the IMU system simultaneously. Data was collected when participants walked on a 7 m walking path. Each participant performed 50 repetitions of walking on the path. To ensure the collection of complete gait cycle data, a gait cycle was considered valid only if the participant passed through the center of the walking path at the same time that the OMC system detected a valid marker signal. As a result, 5 gait cycles that met the standards of the OMC system were included in the final analysis. The stride length, cadence, velocity, stance phase and swing phase of the spatio-temporal parameters were included in the analysis. A generalized linear mixture model was used to assess the repeatability of the two systems. The Wilcoxon rank-sum test for continuous variables was used to compare the mean differences between the two systems. For evaluating the reliability of the IMU system, we calculated the Intra-class Correlation Coefficient (ICC). Additionally, Bland-Altman plots were used to compare the levels of agreement between the two systems.

RESULTS

The measurements of Spatio-temporal parameters, including the stance phase (P = 0.78, 0.13, L-R), swing phase (P = 0.78, 0.13, L-R), velocity (P = 0.14, 0.13, L-R), cadence (P = 0.53, 0.22, L-R), stride length (P = 0.05, 0.19, L-R), by the IMU system and OMC system were similar. Which suggested that IMU and OMC systems could be used interchangeably for gait measurements. The intra-rater reliability showed an excellent correlation for the stance phase, swing phase, velocity and cadence (Intraclass Correlation Coefficient, ICC > 0.9) for both systems. However, the correlation of stride length was poor (ICC = 0.36, P = 0.34, L) to medium (ICC = 0.56, P = 0.22, R). Additionally, the measurements of IMU systems were repeatable.

CONCLUSIONS

The results of IMU system and OMC system shown good repeatability. Wearable IMU system could analyze gait data accurately. In particular, the measurement of stance phase, swing phase, velocity and cadence showed excellent reliability. IMU system provided an alternative measurement to OMC for gait analysis. However, the measurement of stride length by IMU needs further consideration.

Comparison of Two Contact Detection Methods for Ground Reaction Forces and Moment Estimation During Sidestep Cuts, Runs, and Walks

Force platforms often limit the analysis of human movement to the laboratory. Promising methods for estimating ground reaction forces and moments (GRF&M) can overcome this limitation. The most effective family of methods consists of minimizing a cost, constrained by the subject's dynamic equilibrium, for distributing the force over the contact surface on the ground. The detection of contact surfaces over time is dependent on numerous parameters. This study proposes to evaluate two contact detection methods: the first based on foot kinematics and the second based on pressure sole data. Optimal parameters for these two methods were identified for walking, running, and sidestep cut tasks. The results show that a single threshold in position or velocity is sufficient to guarantee a good estimate. Using pressure sole data to detect contact improves the estimation of the position of the center of pressure (CoP). Both methods demonstrated a similar level of accuracy in estimating ground reaction forces.

Recurrent Neural Network Methods for Extracting Dynamic Balance Variables during Gait from a Single Inertial Measurement Unit

Monitoring dynamic balance during gait is critical for fall prevention in the elderly. The current study aimed to develop recurrent neural network models for extracting balance variables from a single inertial measurement unit (IMU) placed on the sacrum during walking. Thirteen healthy young and thirteen healthy older adults wore the IMU during walking and the ground truth of the inclination angles (IA) of the center of pressure to the center of mass vector and their rates of changes (RCIA) were measured simultaneously. The IA, RCIA, and IMU data were used to train four models (uni-LSTM, bi-LSTM, uni-GRU, and bi-GRU), with 10% of the data reserved to evaluate the model errors in terms of the root-mean-squared errors (RMSEs) and percentage relative RMSEs (rRMSEs). Independent t-tests were used for between-group comparisons. The sensitivity, specificity, and Pearson's r for the effect sizes between the model-predicted data and experimental ground truth were also obtained. The bi-GRU with the weighted MSE model was found to have the highest prediction accuracy, computational efficiency, and the best ability in identifying statistical between-group differences when compared with the ground truth, which would be the best choice for the prolonged real-life monitoring of gait balance for fall risk management in the elderly.

Robust deep learning-based gait event detection across various pathologies

The correct estimation of gait events is essential for the interpretation and calculation of 3D gait analysis (3DGA) data. Depending on the severity of the underlying pathology and the availability of force plates, gait events can be set either manually by trained clinicians or detected by automated event detection algorithms. The downside of manually estimated events is the tedious and time-intensive work which leads to subjective assessments. For automated event detection algorithms, the drawback is, that there is no standardized method available. Algorithms show varying robustness and accuracy on different pathologies and are often dependent on setup or pathology-specific thresholds. In this paper, we aim at closing this gap by introducing a novel deep learning-based gait event detection algorithm called IntellEvent, which shows to be accurate and robust across multiple pathologies. For this study, we utilized a retrospective clinical 3DGA dataset of 1211 patients with four different pathologies (malrotation deformities of the lower limbs, club foot, infantile cerebral palsy (ICP), and ICP with only drop foot characteristics) and 61 healthy controls. We propose a recurrent neural network architecture based on long-short term memory (LSTM) and trained it with 3D position and velocity information to predict initial contact (IC) and foot off (FO) events. We compared IntellEvent to a state-of-the-art heuristic approach and a machine learning method called DeepEvent. IntellEvent outperforms both methods and detects IC events on average within 5.4 ms and FO events within 11.3 ms with a detection rate of ≥ 99% and ≥ 95%, respectively. Our investigation on generalizability across laboratories suggests that models trained on data from a different laboratory need to be applied with care due to setup variations or differences in capturing frequencies.

Crossover gait in running and measuring foot inversion angle at initial foot strike: a front-view video analysis approach

Introduction: Foot inversion angle at initial foot strike is associated with various running-related injuries. Traditionally, video analysis of foot inversion angle has been accomplished by positioning a camera to record from the back view, but complications arise when a crossover gait obscures the area of measurement. This study aims to investigate the viability of measuring foot inversion angles at initial foot strike of running from the front view as an alternative to using the back view in 2D video analysis. Methods: Forty-four healthy runners (20 females, 24 males) ran at their self-selected speeds on a treadmill with their gait recorded from front and back camera views. Foot inversion angles at initial foot strike were analyzed using Kinovea. A 2 × 2 (Camera × Foot) ANOVA with repeated measures was performed on the foot inversion angle data. Subsequently, correlation and linear regression were performed to determine the relationship between the back and front-view measurements. Results: Thirteen runners (29.5%) displayed crossover gait within 18 gait cycles. ANOVA revealed a significant main effect on Camera (p < .001) only, where foot inversion angle was greater from the front camera view. Correlation analysis showed a significant positive correlation between the front and back camera views (r = 0.388, p < .001). Regression analyses yielded an equation, y = 0.42 + 0.53 x, where y and x were the foot inversion angle measured from the back and front camera views, respectively. Discussion: With a linear regression conversion equation, front-view foot inversion angles at initial foot strike can be used to determine rearfoot inversion angles when crossover gait obstructs the back camera view.

Comparison of Three Motion Capture-Based Algorithms for Spatiotemporal Gait Characteristics: How Do Algorithms Affect Accuracy and Precision of Clinical Outcomes?

Gait assessment is of interest to clinicians and researchers because it provides information about patients' functional mobility. Optoelectronic camera-based systems with gait event detection algorithms are considered the gold standard for gait assessment. Yet, the choice of the algorithm used to process data and extract the desired parameters from those detected gait events has an impact on the validity and reliability of the gait parameters computed. There are multiple techniques documented in the literature for computing gait events, including the analysis of the minimal position of the heel and toe markers, the computation of the relative distance between sacrum and foot markers, and the assessment of the smallest distance between the heel and toe markers. Validation studies conducted on these algorithms report variations in accuracy. Yet, these studies were conducted in different conditions, at varying gait velocities, and on different populations. The purpose of this study is to compare accuracy, precision, and robustness of three algorithms using motion capture data obtained from 25 healthy persons and 21 psoriatic arthritic patients walking at three distinct speeds on an instrumented treadmill. Errors in gait events recognition (heel strike-HS and toe-off-TO) and their impact on gait metrics (stance phase and stride length) are reported and compared to ground reaction force events measured with force plates. Over the 9114 collected steps across all walking speeds, more than 99% of gait events were recognized by all algorithms. On average, HS events were detected within 1.2 ms of the reference for two algorithms, while the third one detected HS late, with an average detection error of 40.7 ms. Yet, significant variations in accuracy were noted with gait speed; the performance decreased for all algorithms at slow speed. TO events were identified early by all algorithms, with an average error ranging from 16.0 to 100.0 ms. These gait events errors lead to 2-15% inaccuracies in stance phase assessment, while the impact on stride length remains below 0.3 cm. Overall, the algorithm based on the relative distance between the sacral and foot markers stood out for its accuracy, precision, and robustness at all walking speeds.

Advancing age is associated with more impaired mediolateral balance control after step down task

BACKGROUND

To prevent the risk of stair descent falls and associated injuries in the older adults, it is important to understand the factors that affect this frequent locomotion of daily living. The fact that falls are in most cases the result of the interaction between intrinsic and extrinsic factors is very often underestimated when designing test protocols.

RESEARCH QUESTION

This study aimed at evaluating balance control during and immediately after step down onto an unstable surface.

METHODS

Physically active men and women aged 60-69 years (n = 28) and 70-79 years (n = 18) were asked to perform a step down onto a foam pad and stand still for 30 s (restabilization phase). Centre of pressure (CoP) velocity and standard deviation of CoP sway in anteroposterior and mediolateral direction were evaluated during the step down (CoP V_AP, CoP V_ML, CoP SD_AP, CoP SD_ML) and in the first 5 s of restabilization (CoP V_AP5, CoP V_ML5, CoP SD_AP5, CoP SD_ML5). In addition, time to complete step was investigated.

RESULTS

Participants aged 70-79 years presented worse ML balance control after step down onto an unstable surface than their younger counterparts. This was represented by the significantly higher values of CoP SD_ML5 and CoP V_ML5 (p = 0.022 and p = 0.017). No other significant differences were detected.

SIGNIFICANCE

Age is associated with a more significant ML center of pressure velocity and sway after step down onto a foam pad in physically active older adults. Exercises aimed at improving ML balance control in unstable conditions should be the subject of physical interventions even in older adults with overall good state of health and physical fitness.

Silk Fibroin-Based Piezoelectric Sensor with Carbon Nanofibers for Wearable Health Monitoring Applications

The continuous real-time monitoring of human health using biomedical sensing devices has recently become a promising approach to the realization of distant health monitoring. In this paper, the piezoelectric characteristics of the silk fibroin (SF) natural polymer were analyzed as the material used for obtaining sensing information in the application of distance health monitoring. To enhance the SF piezoelectricity, this paper presents the development of a novel SF-based sensor realized by combining SF with different carbon nanofiber (CNF) densities, and for such newly developed SF-based sensors comprehensive performance analyses have been performed. Versatile methods including the scanning electron microscope, Fourier transform infrared spectroscopy, Raman and X-ray diffraction measurements and impedance analysis were used to study the morphologic, mechanical and electrical properties of the developed SF-based sensor. The SF with CNF samples was analyzed for three different pressure loads (40 N, 60 N and 80 N) in 500 compression test cycles. The analyses thoroughly describe how combining natural polymer SF with different CNF densities impacts the piezoelectricity and mechanical strength of the proposed SF-based sensor. The developed piezoelectric SF-based sensors were further tested on humans in real medical applications to detect generated piezoelectric voltage in versatile body movements. The maximum piezoelectricity equal to 2.95 ± 0.03 V was achieved for the jumping movement, and the SF sample with a CNF density equal to 0.4% was tested. Obtained results also show that the proposed SF-based sensor has an appropriate piezoelectric sensitivity for each of the analyzed body movement types, and that the proposed SF-based sensor can be applied in real medical applications as a biomedical sensing device. The proposed SF-based sensor's practical implementation is further confirmed by the results of cytotoxicity analyses, which show that the developed sensor has a non-toxic and biocompatible nature and can be efficiently used in skin contact for biomedical wearable health monitoring applications.

Impact of Gait Events Identification through Wearable Inertial Sensors on Clinical Gait Analysis of Children with Idiopathic Toe Walking

Idiopathic toe walking (ITW) is a gait deviation characterized by forefoot contact with the ground and excessive ankle plantarflexion over the entire gait cycle observed in otherwise-typical developing children. The clinical evaluation of ITW is usually performed using optoelectronic systems analyzing the sagittal component of ankle kinematics and kinetics. However, in standardized laboratory contexts, these children can adopt a typical walking pattern instead of a toe walk, thus hindering the laboratory-based clinical evaluation. With these premises, measuring gait in a more ecological environment may be crucial in this population. As a first step towards adopting wearable clinical protocols embedding magneto-inertial sensors and pressure insoles, this study analyzed the performance of three algorithms for gait events identification based on shank and/or foot sensors. Foot strike and foot off were estimated from gait measurements taken from children with ITW walking barefoot and while wearing a foot orthosis. Although no single algorithm stands out as best from all perspectives, preferable algorithms were devised for event identification, temporal parameters estimate and heel and forefoot rocker identification, depending on the barefoot/shoed condition. Errors more often led to an erroneous characterization of the heel rocker, especially in shoed condition. The ITW gait specificity may cause errors in the identification of the foot strike which, in turn, influences the characterization of the heel rocker and, therefore, of the pathologic ITW behavior.

Neurological scoring and gait kinematics to assess functional outcome in an ovine model of ischaemic stroke

Background

Assessment of functional impairment following ischaemic stroke is essential to determine outcome and efficacy of intervention in both clinical patients and pre-clinical models. Although paradigms are well described for rodents, comparable methods for large animals, such as sheep, remain limited. This study aimed to develop methods to assess function in an ovine model of ischaemic stroke using composite neurological scoring and gait kinematics from motion capture.

Methods

Merino sheep (n = 26) were anaesthetised and subjected to 2 hours middle cerebral artery occlusion. Animals underwent functional assessment at baseline (8-, 5-, and 1-day pre-stroke), and 3 days post-stroke. Neurological scoring was carried out to determine changes in neurological status. Ten infrared cameras measured the trajectories of 42 retro-reflective markers for calculation of gait kinematics. Magnetic resonance imaging (MRI) was performed at 3 days post-stroke to determine infarct volume. Intraclass Correlation Coefficients (ICC's) were used to assess the repeatability of neurological scoring and gait kinematics across baseline trials. The average of all baselines was used to compare changes in neurological scoring and kinematics at 3 days post-stroke. A principal component analysis (PCA) was performed to determine the relationship between neurological score, gait kinematics, and infarct volume post-stroke.

Results

Neurological scoring was moderately repeatable across baseline trials (ICC > 0.50) and detected marked impairment post-stroke (p < 0.05). Baseline gait measures showed moderate to good repeatability for the majority of assessed variables (ICC > 0.50). Following stroke, kinematic measures indicative of stroke deficit were detected including an increase in stance and stride duration (p < 0.05). MRI demonstrated infarction involving the cortex and/or thalamus (median 2.7 cm³, IQR 1.4 to 11.9). PCA produced two components, although association between variables was inconclusive.

Conclusion

This study developed repeatable methods to assess function in sheep using composite scoring and gait kinematics, allowing for the evaluation of deficit 3 days post-stroke. Despite utility of each method independently, there was poor association observed between gait kinematics, composite scoring, and infarct volume on PCA. This suggests that each of these measures has discreet utility for the assessment of stroke deficit, and that multimodal approaches are necessary to comprehensively characterise functional impairment.

Kinematic strategies for obstacle-crossing in older adults with mild cognitive impairment

Introduction

Mild cognitive impairment (MCI) is considered a transitional stage between soundness of mind and dementia, often involving problems with memory, which may lead to abnormal postural control and altered end-point control when dealing with neuromechanical challenges during obstacle-crossing. The study aimed to identify the end-point control and angular kinematics of the pelvis-leg apparatus while crossing obstacles for both leading and trailing limbs.

Methods

12 patients with MCI (age: 66.7 ± 4.2 y/o; height: 161.3 ± 7.3 cm; mass: 62.0 ± 13.6 kg) and 12 healthy adults (age: 67.7 ± 2.9 y/o; height: 159.3 ± 6.1 cm; mass: 61.2 ± 12.0 kg) each walked and crossed obstacles of three different heights (10, 20, and 30% of leg length). Angular motions of the pelvis and lower limbs and toe-obstacle clearances during leading- and trailing-limb crossings were calculated. Two-way analyses of variance were used to study between-subject (group) and within-subject (obstacle height) effects on the variables. Whenever a height effect was found, a polynomial test was used to determine the trend. A significance level of α = 0.05 was set for all tests.

Results

Patients with MCI significantly increased pelvic anterior tilt, hip abduction, and knee adduction in the swing limb during leading-limb crossing when compared to controls (p < 0.05). During trailing-limb crossing, the MCI group showed significantly decreased pelvic posterior tilt, as well as ankle dorsiflexion in the trailing swing limb (p < 0.05).

Conclusion

Patients with MCI adopt altered kinematic strategies for successful obstacle-crossing. The patients were able to maintain normal leading and trailing toe-obstacle clearances for all tested obstacle heights with a specific kinematic strategy, namely increased pelvic anterior tilt, swing hip abduction, and knee adduction during leading-limb crossing, and decreased pelvic posterior tilt and swing ankle dorsiflexion during trailing-limb crossing. The current results suggest that regular monitoring of obstacle-crossing kinematics for reduced toe-obstacle clearance or any signs of changes in crossing strategy may be helpful for early detection of compromised obstacle-crossing ability in patients with single-domain amnestic MCI. Further studies using a motor/cognitive dual-task approach on the kinematic strategies adopted by multiple-domain MCI will be needed for a complete picture of the functional adaptations in such a patient group.

Highly Sensitive Self-Powered Biomedical Applications Using Triboelectric Nanogenerator

The triboelectric nanogenerator (TENG) is a promising research topic for the conversion of mechanical to electrical energy and its application in different fields. Among the various applications, self-powered bio-medical sensing application has become popular. The selection of a wide variety of materials and the simple design of devices has made it attractive for the applications of real-time self-powered healthcare sensing systems. Human activity is the source of mechanical energy which gets converted to electrical energy by TENG fitted to different body parts for the powering up of the biomedical sensing and detection systems. Among the various techniques, wearable sensing systems developed by TENG have shown their merit in the application of healthcare sensing and detection systems. Some key studies on wearable self-powered biomedical sensing systems based on TENG which have been carried out in the last seven years are summarized here. Furthermore, the key features responsible for the highly sensitive output of the self-powered sensors have been briefed. On the other hand, the challenges that need to be addressed for the commercialization of TENG-based biomedical sensors have been raised in order to develop versatile sensitive sensors, user-friendly devices, and to ensure the stability of the device over changing environments.

Predicting vertical ground reaction forces from 3D accelerometry using reservoir computers leads to accurate gait event detection

Accelerometers are low-cost measurement devices that can readily be used outside the lab. However, determining isolated gait events from accelerometer signals, especially foot-off events during running, is an open problem. We outline a two-step approach where machine learning serves to predict vertical ground reaction forces from accelerometer signals, followed by force-based event detection. We collected shank accelerometer signals and ground reaction forces from 21 adults during comfortable walking and running on an instrumented treadmill. We trained one common reservoir computer using segmented data using both walking and running data. Despite being trained on just a small number of strides, this reservoir computer predicted vertical ground reaction forces in continuous gait with high quality. The subsequent foot contact and foot off event detection proved highly accurate when compared to the gold standard based on co-registered ground reaction forces. Our proof-of-concept illustrates the capacity of combining accelerometry with machine learning for detecting isolated gait events irrespective of mode of locomotion.

A deep-learning approach for automatically detecting gait-events based on foot-marker kinematics in children with cerebral palsy-Which markers work best for which gait patterns?

Neuromotor pathologies often cause motor deficits and deviations from typical locomotion, reducing the quality of life. Clinical gait analysis is used to effectively classify these motor deficits to gain deeper insights into resulting walking behaviours. To allow the ensemble averaging of spatio-temporal metrics across individuals during walking, gait events, such as initial contact (IC) or toe-off (TO), are extracted through either manual annotation based on video data, or through force thresholds using force plates. This study developed a deep-learning long short-term memory (LSTM) approach to detect IC and TO automatically based on foot-marker kinematics of 363 cerebral palsy subjects (age: 11.8 ± 3.2). These foot-marker kinematics, including 3D positions and velocities of the markers located on the hallux (HLX), calcaneus (HEE), distal second metatarsal (TOE), and proximal fifth metatarsal (PMT5), were extracted retrospectively from standard barefoot gait analysis sessions. Different input combinations of these four foot-markers were evaluated across three gait subgroups (IC with the heel, midfoot, or forefoot). For the overall group, our approach detected 89.7% of ICs within 16ms of the true event with a 18.5% false alarm rate. For TOs, only 71.6% of events were detected with a 33.8% false alarm rate. While the TOE|HEE marker combination performed well across all subgroups for IC detection, optimal performance for TO detection required different input markers per subgroup with performance differences of 5-10%. Thus, deep-learning LSTM based detection of IC events using the TOE|HEE markers offers an automated alternative to avoid operator-dependent and laborious manual annotation, as well as the limited step coverage and inability to measure assisted walking for force plate-based detection of IC events.

One size does not fit all: Influence of sex and maturation on temporal-spatial parameters for adolescent long-distance runners

Runners and coaches are often interested in identifying the "ideal" running form to reduce the risk of injury and improve performance. While differences in pelvis and hip motion have been reported among adolescent female and male long-distance runners of different stages of physical maturation, the influence of sex and/or maturation on temporal-spatial parameters is unknown for adolescent runners. Adolescent runners of different stages of physical maturation (pre-, mid-, post-pubertal) completed an overground running analysis at a self-selected speed. We performed 2 × 3 ANCOVAs (covariate = running speed) to compare temporal-spatial parameters among sex and maturation groups. Pre-adolescents ran with higher cadences and shorter step lengths than mid- (p ≤ .01) and post-pubertal adolescents (p ≤ .01), respectively. Mid-pubertal males and post-pubertal females also ran with higher cadences and shorter step lengths than post-pubertal males (p ≤ .01). When step length was normalized to leg length, less physically mature runners demonstrated longer normalized step lengths (p ≤ .01). Caution is advised when using a "one-size-fits-all" approach for recommending an "ideal" cadence and/or step length for adolescent long-distance runners. A runner's sex, stage of physical maturation and leg length should be considered when assessing and prescribing cadence and/or step length.

Automatic gait event detection in pathologic gait using an auto-selection approach among concurrent methods

BACKGROUND

Accurate gait event detection is crucial to analyze pathological gait data. Existing methods relying on marker trajectories were reported to be sensitive to different gait patterns, which is an inherent characteristic of pathologic gait.

RESEARCH QUESTION

We propose a new approach based on auto-selection among different methods, original and taken from the literature.

METHODS

The auto-selection approach evaluates the accuracy of the implemented methods for both foot-strike and foot-off on all available events detected by the force platforms, independently, and automatically selects the most accurate one to be used on the whole gait session. Pathological gait data from 272 patients with cerebral palsy and idiopathic toe walking were used retrospectively to evaluate the accuracy of this approach. Three methods previously reported in literature together with original methods developed based on auto-correlation were implemented and constituted our auto-selection approach. The accuracy and precision were compared to a recently reported method based on deep events as it is the method that showed the best performance in literature.

RESULTS

Results showed that the proposed approach outperformed all implemented methods used alone, with an accuracy of - 2.0 ms and - 0.9 ms for foot strike and foot-off, respectively. Additionally, more than 99% and 93% of events detected were detected within 20 ms and 10 ms of accuracy, respectively.

SIGNIFICANCE

The proposed methodology has demonstrated to improve the accuracy and precision of gait event detection in gait analysis.

The placement of foot-mounted IMU sensors does affect the accuracy of spatial parameters during regular walking

Gait analysis using foot-worn inertial measurement units has proven to be a reliable tool to diagnose and monitor many neurological and musculoskeletal indications. However, only few studies have investigated the robustness of such systems to changes in the sensor attachment and no consensus for suitable sensor positions exists in the research community. Specifically for unsupervised real-world measurements, understanding how the reliability of the monitoring system changes when the sensor is attached differently is from high importance. In these scenarios, placement variations are expected because of user error or personal preferences. In this manuscript, we present the largest study to date comparing different sensor positions and attachments. We recorded 9000 strides with motion-capture reference from 14 healthy participants with six synchronized sensors attached at each foot. Spatial gait parameters were calculated using a double-integration method and compared to the reference system. The results indicate that relevant differences in the accuracy of the stride length exists between the sensor positions. While the average error over multiple strides is comparable, single stride errors and variability parameters differ greatly. We further present a physics model and an analysis of the raw sensor data to understand the origin of the observed differences. This analysis indicates that a variety of attachment parameters can influence the systems’ performance. While this is only the starting point to understand and mitigate these types of errors, we conclude that sensor systems and algorithms must be reevaluated when the sensor position or attachment changes.

An Algorithm for Accurate Marker-Based Gait Event Detection in Healthy and Pathological Populations During Complex Motor Tasks

There is growing interest in the quantification of gait as part of complex motor tasks. This requires gait events (GEs) to be detected under conditions different from straight walking. This study aimed to propose and validate a new marker-based GE detection method, which is also suitable for curvilinear walking and step negotiation. The method was first tested against existing algorithms using data from healthy young adults (YA, n = 20) and then assessed in data from 10 individuals from the following five cohorts: older adults, chronic obstructive pulmonary disease, multiple sclerosis, Parkinson’s disease, and proximal femur fracture. The propagation of the errors associated with GE detection on the calculation of stride length, duration, speed, and stance/swing durations was investigated. All participants performed a variety of motor tasks including curvilinear walking and step negotiation, while reference GEs were identified using a validated methodology exploiting pressure insole signals. Sensitivity, positive predictive values (PPV), F1-score, bias, precision, and accuracy were calculated. Absolute agreement [intraclass correlation coefficient (ICC2,1)] between marker-based and pressure insole stride parameters was also tested. In the YA cohort, the proposed method outperformed the existing ones, with sensitivity, PPV, and F1 scores ≥ 99% for both GEs and conditions, with a virtually null bias (&lt;10 ms). Overall, temporal inaccuracies minimally impacted stride duration, length, and speed (median absolute errors ≤1%). Similar algorithm performances were obtained for all the other five cohorts in GE detection and propagation to the stride parameters, where an excellent absolute agreement with the pressure insoles was also found (ICC2,1=0.817− 0.999). In conclusion, the proposed method accurately detects GE from marker data under different walking conditions and for a variety of gait impairments.

Effects of Tai-Chi Chuan Practice on Patterns and Stability of Lower Limb Inter-Joint Coordination During Obstructed Gait in the Elderly

Losing balance or tripping during obstacle-crossing is one of the most frequent causes of falls in the elderly. As a low speed, low impact exercise, Tai Chi Chuan (TCC) can be promising in helping the elderly develop strategies for improved balance, inter-joint coordination, and end-point control during obstacle-crossing. This study investigates the effects of TCC training on the patterns and variability of the lower-limb inter-joint coordination during obstacle-crossing in the elderly. Fifteen older TCC practitioners and 15 healthy controls crossed obstacles of three different heights, while sagittal angles (x) and angular velocities (x′) of the hips, knees and ankles were measured and their phase angles obtained. The continuous relative phases (CRP) of the hip-knee and knee-ankle coordination were also calculated. The standard deviations of the CRP curve points were averaged to obtain deviation phase (DP) values for the stance and swing phases. The TCC group was found to cross obstacles with increased leading and trailing toe-clearances with unaltered CRP values when the swing toe was above the obstacle. Long-term TCC training altered the patterns and magnitudes of the CRPs primarily over double-limb support and significantly reduced the variabilities of leading knee-ankle and trailing hip-knee and knee-ankle CRP curves over the crossing cycle, regardless of obstacle height. The current results suggest that long-term TCC practice was helpful for a crossing strategy with significantly increased foot-obstacle clearances and reduced variability of the way the motions of the lower limb joints are coordinated during obstacle-crossing. These benefits may be explained by the long-lasting effects of continuous practice of the slow movement patterns emphasizing between-limb transfer of body weight in TCC.

Comparison between the Rizzoli and Oxford foot models with independent and clustered tracking markers

BACKGROUND

The Rizzoli Foot Model (RFM) and Oxford Foot Model (OFM) are used to analyze segmented foot kinematics with independent tracking markers. Alternatively, rigid marker clusters can be used to improve markers' visualization and facilitate analyzing shod gait.

RESEARCH QUESTION

Are there differences in angles from the RFM and OFM, obtained with independent and clustered tracking markers, during the stance phase of walking?

METHODS

Walking kinematics of 14 non-disabled participants (25.2 years (SD 2.8)) were measured at self-selected speed. Rearfoot-shank and forefoot-rearfoot angles were measured from two models with two tracking methods: RFM, OFM, RFM-cluster, and OFM-cluster. In RFM-cluster and OFM-cluster, the rearfoot and forefoot tracking markers were rigidly clustered, fixed on rods' tips attached to a metallic base. Statistical Parametric Mapping (SPM) One-Way Repeated Measures ANOVAs and SPM Paired t-tests were used to compare waveforms. Coefficients of Multiple Correlation (CMC) quantified the similarity between waveforms. One-way Repeated Measures ANOVAs were conducted to compare the ranges of motion (ROMs), and pre-planned contrasts investigated differences between the models and tracking methods. Intraclass Correlation Coefficients (ICC) were computed to verify the similarity between ROMs.

RESULTS

Differences occurred mostly in small parts of the stance phase for the cluster vs. non-cluster comparisons and the RFM vs. OFM comparisons. ROMs were slightly different between the models and tracking methods in most comparisons. The curves (CMC ≥ 0.71) were highly similar between the models and tracking methods. The ROMs (ICC ≥ 0.67) were moderatetly to highly similar in most comparisons. RFM vs. RFM-cluster (forefoot-rearfoot angle - transverse plane), OFM vs. OFM-cluster and RFM vs. OFM (forefoot-rearfoot angle - frontal plane) were not similar (non-significant).

SIGNIFICANCE

Rigid clusters are an alternative for tracking rearfoot-shank and forefoot-rearfoot angles during the stance phase of walking. However, specific differences should be considered to contrast results from different models and tracking methods.

The Impact of Load Style Variation on Gait Recognition Based on sEMG Images Using a Convolutional Neural Network

Surface electromyogram (sEMG) signals are widely employed as a neural control source for lower-limb exoskeletons, in which gait recognition based on sEMG is particularly important. Many scholars have taken measures to improve the accuracy of gait recognition, but several real-time limitations affect its applicability, of which variation in the load styles is obvious. The purposes of this study are to (1) investigate the impact of different load styles on gait recognition; (2) study whether good gait recognition performance can be obtained when a convolutional neural network (CNN) is used to deal with the sEMG image from sparse multichannel sEMG (SMC-sEMG); and (3) explore whether the control system of the lower-limb exoskeleton trained by sEMG from part of the load styles still works efficiently in a real-time environment where multiload styles are required. In addition, we discuss an effective method to improve gait recognition at the levels of the load styles. In our experiment, fifteen able-bodied male graduate students with load (20% of body weight) and using three load styles (SBP = backpack, SCS = cross shoulder, SSS = straight shoulder) were asked to walk uniformly on a treadmill. Each subject performed 50 continuous gait cycles under three speeds (V3 = 3 km/h, V5 = 5 km/h, and V7 = 7 km/h). A CNN was employed to deal with sEMG images from sEMG signals for gait recognition, and back propagation neural networks (BPNNs) and support vector machines (SVMs) were used for comparison by dealing with the same sEMG signal. The results indicated that (1) different load styles had remarkable impact on the gait recognition at three speeds under three load styles (p < 0.001); (2) the performance of gait recognition from the CNN was better than that from the SVM and BPNN at each speed (84.83%, 81.63%, and 83.76% at V3; 93.40%, 88.48%, and 92.36% at V5; and 90.1%, 86.32%, and 85.42% at V7, respectively); and (3) when all the data from three load styles were pooled as testing sets at each speed, more load styles were included in the training set, better performance was obtained, and the statistical analysis suggested that the kinds of load styles included in training set had a significant effect on gait recognition (p = 0.002), from which it can be concluded that the control system of a lower-limb exoskeleton trained by sEMG using only some load styles is not sufficient in a real-time environment.

Gait event detection using kinematic data in children with bilateral spastic cerebral palsy

BACKGROUND

Ground reaction forces are the gold standard for detecting gait events, but they are not always applicable in cerebral palsy. Ghoussayni's algorithm is an event detection method based on the sagittal plane velocity of heel and toe markers. We aimed to evaluate whether Ghoussayni's algorithm, using two different thresholds, was a valid event detection method in children with bilateral spastic cerebral palsy. We also aimed to define a new adaptation of Ghoussayni's algorithm for detecting foot strike in cerebral palsy, and study the effect of event detection methods on spatiotemporal parameters.

METHODS

Synchronized kinematic and kinetic data were collected retrospectively from 16 children with bilateral spastic cerebral palsy (7 males and 9 females; age 8.9 ± 2.7 years) walking barefoot at self-selected speed. Gait events were detected using methods: 1) ground reaction forces, 2) Ghoussayni's algorithm with a threshold of 0.5 m/s, and 3) Ghoussayni's algorithm with a walking speed dependent threshold. The new adaptation distinguished how foot strikes were performed (heel and/or toe) comparing the timing when the foot markers velocities fell below the threshold. Differences between the three methods, and between spatiotemporal parameters calculated from the two Ghoussayni's thresholds were analyzed.

FINDINGS

There were statistically significant (P < 0.05) differences between methods 1 and 3, and between some spatiotemporal parameters calculated from methods 2 and 3. Ghoussayni's algorithm showed better performance for foot strike than for toe off.

INTERPRETATION

Ghoussayni's algorithm using 0.5 m/s is valid in children with bilateral spastic cerebral palsy. Event detection methods affect spatiotemporal parameters.

Impact of the Marker Set Configuration on the Accuracy of Gait Event Detection in Healthy and Pathological Subjects

For interpreting outcomes of clinical gait analysis, an accurate estimation of gait events, such as initial contact (IC) and toe-off (TO), is essential. Numerous algorithms to automatically identify timing of gait events have been developed based on various marker set configurations as input. However, a systematic overview of the effect of the marker selection on the accuracy of estimating gait event timing is lacking. Therefore, we aim to evaluate (1) if the marker selection influences the accuracy of kinematic algorithms for estimating gait event timings and (2) what the best marker location is to ensure the highest event timing accuracy across various gait patterns. 104 individuals with cerebral palsy (16.0 ± 8.6 years) and 31 typically developing controls (age 20.6 ± 7.8) performed clinical gait analysis, and were divided into two out of eight groups based on the orientation of their foot, in sagittal and frontal plane at mid-stance. 3D marker trajectories of 11 foot/ankle markers were used to estimate the gait event timings (IC, TO) using five commonly used kinematic algorithms. Heatmaps, for IC and TO timing per group were created showing the median detection error, compared to detection using vertical ground reaction forces, for each marker. Our findings indicate that median detection errors can be kept within 7 ms for IC and 13 ms for TO when optimizing the choice of marker and detection algorithm toward foot orientation in midstance. Our results highlight that the use of markers located on the midfoot is robust for detecting gait events across different gait patterns.

MATLAB-based tools for automated processing of motion tracking data provided by the GRAIL

BACKGROUND

The ability for independent bipedal locomotion is an important prerequisite for autonomous mobility and participation in everyday life. Walking requires not only a functional musculoskeletal unit but relies on coordinated activation of muscles and may even require cognitive resources. The time-resolved monitoring of the position of joints, feet, legs and other body segments relative to each other alone or in combination with simultaneous recording of ground reaction forces and concurrent measurement of electrical muscle activity, using surface electromyography, are well-established tools for the objective assessment of gait.

RESEARCH QUESTION

The Gait Real-time Analysis Interactive Lab (GRAIL) has been introduced for gait analysis in a highly standardized and well-controlled virtual environment. However, apart from high computing capacity and sophisticated software required to run the system, handling of GRAIL data is challenging due to the utilization of different software packages resulting in a huge amount of data stored using different file formats and different sampling rates. These issues make gait analysis even with such a sophisticated instrument rather tedious, especially within the frame of an experimental or clinical study.

METHODS

A user-friendly Matlab based toolset for automated processing of motion capturing data recorded using the GRAIL, with the inherent option for batch analysis was developed.

RESULTS

The toolset allows the reading, resampling, filtering and synchronization of data stored in different input files recorded with the GRAIL. It includes a coordinate-based algorithm for the detection of initial contact and toe-off events to split and normalize data relative to gait cycles. Batch processing of multiple measurements and automatic detection of outliers is possible.

SIGNIFICANCE

The authors hope that the toolset will be useful to the research community and invite everyone to use, modify or implement it in their own work.

Two-dimensional dynamic walking stability of elderly females with a history of falls

Injuries related with falls are a major health risk for the elderly. Accurate evaluation of the dynamic walking stability of elderly people is the key to fall prevention. A two-dimensional (2-D) model is proposed in this study given that the custom method is mainly focused on the dynamic walking stability along the antero-posterior axis. An inverted pendulum model was utilised to calculate the region of stability at toe-off, and stability conditions were evaluated first along the antero-posterior and medio-lateral axes. The analysis was then extended to the 2-D plane. In the 2-D case, the region of stability was determined based on the use of the information of the envelope of the foot. Twenty-four female participants, categorised as healthy young, healthy elderly, and elderly with a history of falls, were examined. Significant differences among the three groups were demonstrated with the 2-D analysis method, but not in the antero-posterior or medio-lateral analyses. The centre-of-masses of elderly fallers were significantly closer to the foot-supporting boundary compared with that of healthy young and elderly adults at toe-off. A 2-D analysis method using the envelope-of-foot could evaluate the dynamic stability of elderly females based on a more accurate scale.

Foot pronation affects pelvic motion during the loading response phase of gait

BACKGROUND

Increased foot pronation during walking has been associated with low back pain. This association may be due to the impact of increased pronation on pelvic motion.

OBJECTIVE

To investigate the effects of increased bilateral foot pronation on pelvic kinematic in frontal and transverse planes during the loading response phase of gait.

METHODS

Pelvic, hip, and foot angular positions of 20 participants were collected while they walked at fast speed wearing flat and medially inclined insoles inserted in the shoes. Pelvic motion in frontal and transverse planes was analyzed during the loading response phase. Foot eversion-inversion was analyzed during the complete stance phase to verify the insoles effectiveness in inducing increased pronation and to exclude excessive pronators.

RESULTS

Inclined insoles were effective in inducing increased foot pronation. Pelvic and hip motion were altered in the increased pronation condition compared to the control condition. In the frontal plane, mean pelvic position was more inclined to the contralateral side (mean difference [MD]: 0.54°; 95%CI: 0.23, 0.86) and its range of motion (ROM) was reduced (MD: 0.50°; 95%CI: 0.20, 0.79). In the transverse plane, mean pelvic position was less rotated toward the contralateral leg (MD: 1.03°; 95%CI: 0.65, 1.60) without changes in ROM (MD: 0.04°; 95%CI: -0.17, 0.25). The hip was more internally rotated (MD: 1.37°; 95%CI: 0.76, 1.98) without changes in ROM (MD: 0.10°; 95%CI: -1.02, 1.23).

CONCLUSION

Increased bilateral foot pronation changes pelvic motion during walking and should be assessed, as a contributing factor to possible pelvic and lower back disorders.

Towards validation and standardization of automatic gait event identification algorithms for use in paediatric pathological populations

BACKGROUND

To analyse and interpret gait patterns in pathological paediatric populations, accurate determination of the timing of specific gait events (e.g. initial contract - IC, or toe-off - TO) is essential. As currently used clinical identification methods are generally subjective, time-consuming, or limited to steps with force platform data, several techniques have been proposed based on processing of marker kinematics. However, until now, validation and standardization of these methods for use in diverse gait patterns remains lacking.

RESEARCH QUESTIONS

1) What is the accuracy of available kinematics-based identification algorithms in determining the timing of IC and TO for diverse gait signatures? 2) Does automatic identification affect interpretation of spatio-temporal parameters?.

METHODS

3D kinematic and kinetic data of 90 children were retrospectively analysed from a clinical gait database. Participants were classified into 3 gait categories: group A (toe-walkers), B (flat IC) and C (heel IC). Five kinematic algorithms (one modified) were implemented for two different foot marker configurations for both IC and TO and compared with clinical (visual and force-plate) identification using Bland-Altman analysis. The best-performing algorithm-marker configuration was used to compute spatio-temporal parameters (STP) of all gait trials. To establish whether the error associated with this configuration would affect clinical interpretation, the bias and limits of agreement were determined and compared against inter-trial variability established using visual identification.

RESULTS

Sagittal velocity of the heel (Group C) or toe marker configurations (Group A and B) was the most reliable indicator of IC, while the sagittal velocity of the hallux marker configuration performed best for TO. Biases for walking speed, stride time and stride length were within the respective inter-trial variability values.

SIGNIFICANCE

Automatic identification of gait events was dependent on algorithm-marker configuration, and best results were obtained when optimized towards specific gait patterns. Our data suggest that correct selection of automatic gait event detection approach will ensure that misinterpretation of STPs is avoided.

How does wearing slippers affect the movement strategy while crossing over an obstacle?

BACKGROUND

Wearing slippers may increase the risk of tripping while crossing obstacles, regardless of age. This is because slippers are more likely to leave the feet and come into contact with obstacles. However, how wearing slippers affects the movement strategy while crossing over an obstacle has not been clarified.

RESEARCH QUESTION

How does wearing slippers affect the movement strategy while crossing over an obstacle?

METHODS

Thirty healthy young adults crossed over an obstacle using a comfortable speed under two conditions: barefoot and wearing slippers. The moment when the leading or trailing limb was crossing the obstacle was defined as obstacle lead (OL) or obstacle trail (OT), respectively. The margin of stability (MoS) as a measure of stability was measured at OL and OT. Toe clearance and lower limb joint angles of the leading limb were measured at OL, and those of the trailing limb were measured at OT.

RESULTS

Wearing slippers increased toe clearance and flexed the hip and knee joints, regardless of the crossing event (OL, OT). However, the ankle joint angle did not change between footwear conditions at OL and was significantly dorsiflexed in the slippered condition at OT. In addition, the MoS did not change between footwear conditions at OL and increased significantly in the slippered condition at OT.

SIGNIFICANCE

The increase of the MoS at OT when wearing slippers is probably a conservative strategy to reduce the risk of falling forward when a trip occurs. More attention to avoiding tripping likely caused the ankle to dorsiflex and increased toe clearance. This conservative strategy and increased attention are most likely due to the slippers coming off easily. Since these strategies are probably safety measures to prevent tripping, slippers may not be appropriate footwear in terms of tripping risk.

Estimation of ground reaction forces during stair climbing in patients with ACL reconstruction using a depth sensor-driven musculoskeletal model

BACKGROUND

Although stair ambulation should be included in the rehabilitation of the long-term effects of ACL injury on knee function, the assessment of kinetic parameter in the situation where stair gait can only be established using costly and cumbersome force platforms via conventional inverse dynamic analysis. Therefore, there is a need to develop a practical laboratory setup as an assessment tool of the stair gait abnormalities in lower extremity that arise from an ACL deficiency.

RESEARCH QUESTION

Can the use of a single depth sensor-driven full-body musculoskeletal gait model be considered an accurate assessment tool of the ground reaction forces (GRFs) during stair climbing for patients following ACL reconstruction (ACLR) surgery?

METHODS

A total of 15 patients who underwent ACLR participated in this study. GRFs data during stair climbing was collected using a custom-built 3-step staircase with two embedded force platforms. A single depth sensor, commercially available and cost effective, was used to obtain participants' depth map information to extract the full-body skeleton information. The AnyBody TM GaitFullBody model was utilized to estimate GRFs attained by 25 artificial muscle-like actuators placed under each foot. Mean differences between the measured and estimated GRFs were compared using paired samples t-tests. The ensemble curves of the GRFs were compared between both approaches during stance phase of the gait cycle.

RESULTS

The findings of this study showed that the estimation of the GRFs produced during staircase gait using a depth sensor-driven musculoskeletal model can produce acceptable results when compared to the traditional inverse dynamics modelling approach as an alternative tool in clinical settings for individuals who had undergone ACLR.

SIGNIFICANCE

The introduced approach of full-body musculoskeletal modelling driven by a single depth sensor has the potential to be a cost-effective stair gait analysis tool for patients with ACL injury.

Initial center of pressure position prior to anticipatory postural adjustments during gait initiation in people with Parkinson's disease with freezing of gait

INTRODUCTION

Freezing of gait (FOG) in Parkinson's disease (PD) is associated with an altered posture during quiet stance as well as an impaired preparation and execution of the gait initiation process. We aimed to investigate whether an altered initial posture impacts anticipatory postural adjustments (APAs) and first-step execution during gait initiation in people with PD with FOG (PD + FOG).

METHODS

Twenty-seven PD+FOG, 30 PD patients without FOG and 27 age-matched healthy controls performed self-generated gait initiation. Initial mean center of pressure (COP) position prior to APA onset, characteristics of APAs and features of first-step execution were investigated.

RESULTS

Contrarily to controls, PD patients showed a COP that was initially positioned more towards the stance leg (p = 0.007). Moreover, significantly smaller backward COP shift, longer duration of swing-foot unloading phase, and lower first-step length and velocity characterized PD+FOG compared to controls. While size and duration of backward COP shift during APA and lateral COP shift during the unloading phase were main predictors of first-step length and velocity in all groups, the medio-lateral shift of the initial COP position in PD+FOG was a main predictor of first-step execution (β = -0.191, p = 0.001 for velocity).

CONCLUSION

In PD+FOG, the more the COP was initially positioned towards the stance foot, the slower and shorter the first step. The initial medio-lateral COP position may be a compensatory strategy to address postural instability of PD+FOG. A specific training regarding postural control prior to gait preparation and execution could improve functional mobility in PD+FOG.

Age-related differences in stair descent balance control: Are women more prone to falls than men?

Stair descent is one of the most common forms of daily locomotion and concurrently one of the most challenging and hazardous daily activities performed by older adults. Thus, sufficient attention should be devoted to this locomotion and to the factors that affect it. This study investigates gender and age-related differences in balance control during and after stair descent on a foam mat. Forty-seven older adults (70% women) and 38 young adults (58% women) performed a descent from one step onto a foam mat. Anteroposterior (AP) and mediolateral (ML) centre of pressure velocity (CoP) and standard deviation of the CoP sway were investigated during stair descent and restabilization. A two-way analysis of variance (ANOVA) revealed the main effects of age for the first 5 s of restabilization. Older women exhibited significantly higher values of CoP sway and velocity in both directions compared to the younger individuals (CoP SD_AP5, 55%; CoP SD_ML5, 30%; CoP V_AP5, 106%; CoP V_ML5, 75%). Men achieved significantly higher values of CoP sway and velocity only in the AP direction compared to their younger counterparts (CoP SD_AP5, 50% and CoP V_AP5, 79%). These findings suggest that with advancing age, men are at higher risk of forward falls, whereas women are at higher risk of forward and sideways falls.

Finite Class Bayesian Inference System for Circle and Linear Walking Gait Event Recognition Using Inertial Measurement Units

Accurate and fast human motion pattern recognition is the key to ensuring lower limb assistive devices' appropriate assistance. The research on human motion pattern recognition of lower limb assistive devices mainly focuses on sagittal gait. The motion pattern such as circular walking (CW) is asymmetric about the sagittal plane of the body. CW is common in daily living. However, the recognition algorithm of CW is rarely reported. Since lower limb assistive devices interact with humans, lacking the capability of recognizing CW is dangerous. Thus, to realize the accurate and fast recognition of CW, this article proposed a finite class Bayesian interference system (FC-BesIS). FC-BesIS is designed to recognize walking activities (linear walking and CW) and gait events (heel contact, load response, mid stance, terminal stance, pre-swing, initial swing, mid swing, and terminal swing). A finite class method which reduces the number of potential classes according to elimination rules before decision-making is introduced. Elimination rules are designed based on likelihood estimation and sensor information. The experiments show that walking activities and gait events can be accurately and fastly recognized by FC-BesIS. The experiments also show that the performance of FC-BesIS in mean recognition accuracy (MRA) and mean decision time (MDT) is improved compared with BesIS. The MRA of walking activities and gait events are 100% and 97.38%, respectively. The MDT of walking activities and gait events are 28.19 ms and 33.94 ms, respectively. Overall, FC-BesIS has been proved to be an accurate and fast recognition algorithm for human motion patterns using wearable sensors.

Balance control during stair descent on compliant surface is associated with knee flexor and ankle plantar flexor strength in older adults

Balance and lower limb strength deficits are associated with a high incidence of falls in older adults. This study investigated the association between balance control during and after stair descent onto a compliant surface and lower limb strength. Thirty-five women and 14 men participated in this study. Stair descent time, mean center of pressure velocity in anteroposterior and mediolateral direction during stair descent (CoP V_AP and CoP V_ML), and CoP velocity in the first 5 s of restabilization phase (CoP V₅) were evaluated. Bilateral strength of the knee flexors and extensors, and ankle plantar and dorsal flexors was evaluated. Spearman correlation analysis with Bonferroni correction yielded a significant association between the strength of the knee flexors on the trailing limb and stair descent time in women (r = 0.502, p = 0.002, R² = 0.246). The same analysis in men revealed a significant association between the strength of the knee flexors on the trailing limb and CoP V_AP (r = -0.820, p < 0.001, R² = 0.280) and CoP V_ML (r = -0.697, p = 0.006, R² = 0.359). The strength of the ankle plantar flexors on the trailing limb was significantly associated with stair descent time (r = 0.684, p = 0.007, R² = 0.429) and CoP V_AP (r = -0.723, p = 0.003, R² = 0.408) in men. Stair descent balance control is associated with knee flexion strength on trailing limb in women, and with ankle plantar flexion and knee flexion strength on the same limb in men.

Reactive balance responses to an unexpected slip perturbation in individuals with incomplete spinal cord injury

BACKGROUND

Frequent falls while walking among individuals with incomplete spinal cord injury may suggest impairments in reactive balance control; however, reactive balance control during walking has not been studied in this population. The objective was to compare reactive balance control with respect to changes in margin of stability, onset of arm and heel responses, and onset and magnitude of muscle activity following an unexpected slip perturbation in individuals with incomplete spinal cord injury and able-bodied individuals.

METHODS

Kinematic and electromyography data were obtained during normal walking and one unexpected slip. Changes in margin of stability following a compensatory or aborted step, onset of arms and trail heel responses, and onset and magnitude of activation of the tibialis anterior, soleus and gluteus medius were calculated. Multivariate analyses compared responses between incomplete spinal cord injury and able-bodied groups.

FINDINGS

Data from 16 participants with incomplete spinal cord injury (all American Spinal Injury Association Impairment Scale Grade D, 8 with tetraplegia) and 13 age-and-sex matched able-bodied individuals were included. Individuals with incomplete spinal cord injury demonstrated limited ability to increase margin of stability in the lateral direction during a compensatory or aborted step, and a smaller magnitude of soleus activity compared to able-bodied individuals.

INTERPRETATION

There are limitations in reactive balance control of individuals with incomplete spinal cord injury, which may be a reason for the high frequency of falls in this population. Reactive balance assessment should be included as a component of routine balance assessment and fall avoidance strategies in this population.

Is There an Optimal Recovery Step Landing Zone Against Slip-Induced Backward Falls During Walking?

Recovery stepping in response to forward slips has the potential to not only rebuild the base of support to prevent backward falling, but also provide extra limb support to prevent downward falling. Hence, recovery stepping is often necessary for fall prevention following an unexpected slip. However, less is known about whether recovery foot placement could affect the likelihood of recovery following a slip. The purpose of this study was to determine whether there is an optimal recovery landing zone within which older adults have a higher likelihood of recovery. 195 participants experienced a novel, unannounced forward slip while walking on a 7-m walkway. The center of mass (COM) stability (computed from its position and velocity), vertical limb support (computed from change in hip kinematics), and recovery limb joint moments (computed from joint kinematics and ground reaction force) in the sagittal plane were analyzed. The results showed that a longer distance between recovery foot landing position and the projected COM position at recovery foot touchdown (relative recovery step placement) was conducive to stability improvement but adverse to limb support enhancement, and vice versa for a shorter distance. Relative recovery step placement could predict the recovery likelihood with an accuracy of 67.3%, and the recovery rate was greater than 50% when the distance between recovery foot and COM is less than 0.3 × foot length. This study also found more posterior stepping could be attributed to insufficient ankle plantar flexor and hip flexor moments in the pre-swing phase, while more anterior stepping was induced by insufficient hip and knee extensor moments in the following swing phase.

Comparison of three kinematic gait event detection methods during overground and treadmill walking for individuals post stroke

Detecting gait events using ground reaction forces (i.e. kinetic detection) is the gold standard, but it is not always possible. Kinematic methods exist; however, accuracy of these methods in stroke survivors during treadmill and overground walking is unknown. Thus, this study compared the accuracy of three kinematic methods during overground and treadmill walking in stroke survivors. Heel strike and toe off were calculated bilaterally using three kinematic methods (horizontal sacral-heel distance, horizontal ankle-heel distance, and horizontal velocity) and a kinetic method for ten stroke survivors. We calculated true and absolute error for each kinematic method relative to the kinetic method to evaluate accuracy. Repeated-measures ANOVAs compared the absolute error between the different methods for each condition. There was a significant effect of method for all conditions except heel strike during treadmill walking. Post hoc tests showed ankle-heel distance detected heel strike with significantly less error than the other methods during overground walking (p < 0.05). Ankle-heel distance identified 93.0% and 77.8% of gait events within 50 ms of the kinetic event for overground and treadmill walking, respectively. Sacral-heel distance detected toe-off with significantly less error than the other methods during overground and treadmill walking (p < 0.05) and identified 87.2% and 90.3% of gait events within 50 ms of the kinetic event for overground and treadmill walking, respectively. Results suggest that ankle-heel distance and sacral-heel distance accurately detect heel strike and toe-off, respectively, in stroke survivors.

Deep Learning based Gait Abnormality Detection using Wearable Sensor System

Gait is an extraordinary complex function of human body that involves the activation of entire visceral nervous system, making human gait definite to various functional abnormalities. Diagnosis and treatment of such disorders prior to their development can be achieved through integration of modern technologies with state-of-the-art developed methods. Modern machine learning techniques have outperformed and complemented the use of conventional statistical methods in bio-medical systems. In this research a wearable sensor system is presented, which combines plantar pressure measurement unit and Inertial Measurement Units (IMU's) integrated with a stacked Long short-term memory (LSTM) model to detect human gait abnormalities that are prone to the risk of fall. The computed metrics and gait parameters show significant differences between normal and abnormal gait patterns. Three specific abnormalities involving Hemiplegic, Parkinsonian and Sensory-Ataxic gaits are simulated to validate the proposed model and show promising results. The proposed research aims to demonstrate how advanced technologies can be used in gait diagnosis and treatment assistant systems.

A new deep learning-based method for the detection of gait events in children with gait disorders: Proof-of-concept and concurrent validity

The stance and swing phases of the gait cycle are defined by foot strike (FS) and foot off (FO). Accurate determination of these events is thus an essential component of 3D motion recordings processing. Several methods have been developed for the automatic detection of these events (based on the heuristics of 3D marker position, velocity and acceleration), however the results may be inaccurate due to the high variability that is intrinsic to pathological gait. For this reason, gait events are still commonly determined manually, which is a tedious process. Here we propose a new application (DeepEvent) of a long short term memory recurrent neural network for the automatic detection of gait events. The 3D position and velocity of the markers on the heel, toe and lateral malleolus were used by the network to determine FS and FO. The method was developed from 10526 FS and 9375 FO from 226 children. DeepEvent predicted FS within 5.5 ms and FO within 10.7 ms of the gold standard (automatic determination using force platform data) and was more accurate than common heuristic marker trajectory-based methods proposed in the literature and another deep learning method. A sensitivity analysis showed that DeepEvent mainly used the toe and heel markers (z-axis (longitudinal) position and velocity) at the beginning and end of gait cycle to predict FS, and the toe marker (x-axis (anterior/posterior) velocity and z-axis position and velocity) at around 60% of the gait cycle to predict FO.

Does the margin of stability measure predict medio-lateral stability of gait with a constrained-width base of support?

This study aimed to determine the validity of the centre of mass position (COM) position and extrapolated COM (XCOM), relative to the base of support, for predicting medio-lateral stability during a walking task where the base of support width is limited. Nine young healthy participants walked on a narrow beam. Three-dimensional motion capture was used to calculate the COM and XCOM relative to the base of support. Steps were classified as having either the COM or XCOM inside or outside the base of support, and were classified as successful (stable - foot placed on the beam) or failed (unstable - foot stepped off the beam). If the COM or XCOM are valid measures of stability, they should be within the base of support for successful steps and outside the base of support for failed steps. Classifying the COM and XCOM inside or outside the base of support correctly predicted successful or failed steps in 69% and 58% of cases, respectively. When the COM or XCOM were outside the base of support, walking faster seemed to help participants to maintain stability. The further the COM or XCOM were outside the base of support during a successful step, the more likely participants were to fail on a subsequent step. The results of this study suggest that both COM and XCOM are valid measures of stability during a beam walking task, but that classifying COM and XCOM as inside or outside the base of support may be over-simplistic.

A Direct Comparison of Biplanar Videoradiography and Optical Motion Capture for Foot and Ankle Kinematics

Measuring motion of the human foot presents a unique challenge due to the large number of closely packed bones with congruent articulating surfaces. Optical motion capture (OMC) and multi-segment models can be used to infer foot motion, but might be affected by soft tissue artifact (STA). Biplanar videoradiography (BVR) is a relatively new tool that allows direct, non-invasive measurement of bone motion using high-speed, dynamic x-ray images to track individual bones. It is unknown whether OMC and BVR can be used interchangeably to analyse multi-segment foot motion. Therefore, the aim of this study was to determine the agreement in kinematic measures of dynamic activities. Nine healthy participants performed three walking and three running trials while BVR was recorded with synchronous OMC. Bone position and orientation was determined through manual scientific-rotoscoping. The OMC and BVR kinematics were co-registered to the same coordinate system, and BVR tracking was used to create virtual markers for comparison to OMC during dynamic trials. Root mean square (RMS) differences in marker positions and joint angles as well as a linear fit method (LFM) was used to compare the outputs of both methods. When comparing BVR and OMC, sagittal plane angles were in good agreement (ankle: R² = 0.947, 0.939; Medial Longitudinal Arch (MLA) Angle: R² = 0.713, 0.703, walking and running, respectively). When examining the ankle, there was a moderate agreement between the systems in the frontal plane (R² = 0.322, 0.452, walking and running, respectively), with a weak to moderate correlation for the transverse plane (R² = 0.178, 0.326, walking and running, respectively). However, root mean squared error (RMSE) showed angular errors ranging from 1.06 to 8.31° across the planes (frontal: 3.57°, 3.67°, transverse: 4.28°, 4.70°, sagittal: 2.45°, 2.67°, walking and running, respectively). Root mean square (RMS) differences between OMC and BVR marker trajectories were task dependent with the largest differences in the shank (6.0 ± 2.01 mm) for running, and metatarsals (3.97 ± 0.81 mm) for walking. Based on the results, we suggest BVR and OMC provide comparable solutions to foot motion in the sagittal plane, however, interpretations of out-of-plane movement should be made carefully.

Pelvic Drop Changes due to Proximal Muscle Strengthening Depend on Foot-Ankle Varus Alignment

Background

Strengthening of hip and trunk muscles can modify pelvis and hip movements. However, the varus alignment of the foot-ankle complex (FAC) may influence the effects of muscle strengthening, due to the relationship of FAC alignment with pelvic and hip kinematics. This study evaluated the effects of hip and trunk muscle strengthening on pelvis and hip kinematics during walking, in subgroups with larger and smaller values of FAC varus alignment. In addition, this study evaluated the effects of hip and trunk muscle strengthening on hip passive and active properties, in the same subgroups.

Methods

Fifty-three women, who were divided into intervention and control groups, participated in this nonrandomized controlled trial. Each group was split into two subgroups with larger and smaller values of FAC varus alignment. Hip and trunk muscle strengthening was performed three times a week for two months, with a load of 70% to 80% of one repetition maximum. Before and after strengthening, we evaluated (1) pelvis and hip excursions in the frontal and transverse planes during walking, (2) isokinetic hip passive external rotator torque, and (3) isokinetic concentric and eccentric peak torques of the hip external rotator muscles. Mixed analyses of variance (ANOVAs) were carried out for each dependent variable related to walking kinematics and isokinetic measurements (α = 0.05).

Results

The subgroup with smaller varus alignment, of the intervention group, presented a reduction in pelvic drop after strengthening (P = 0.03). The subgroup with larger varus alignment increased pelvic drop after strengthening, with a marginal significance (P = 0.06). The other kinematic excursions did not change (pelvic anterior rotation P = 0.30, hip internal rotation P = 0.54, and hip adduction P = 0.43). The intervention group showed increases in passive torque (P = 0.002), peak concentric torque (P < 0.001), and peak eccentric torque (P < 0.001), independently of FAC alignment. These results suggest that FAC varus alignment influences the effects of strengthening and should be considered when hip and trunk muscle strengthening is used to reduce pelvic drop during walking.

Identification of gait events in children with spastic cerebral palsy: comparison between the force plate and algorithms

OBJECTIVE

To compare the gait event identification of five algorithms recommended in the literature with those provided by force plate (gold standard) in children with unilateral or bilateral spastic cerebral palsy (SCP).

METHODS

This was a cross-sectional study of the gait of three girls and four boys with a mean age of 8.6±4.7 years. Four children had unilateral SCP with an equinus gait pattern, and the remaining three children exhibited bilateral SCP with a slide/drag gait pattern. Kinematic and kinetic gait data were collected during barefoot walking at a comfortable speed. From a total of 202 steps, the detection of 202 foot-strike (FS) and 194 toe-off (TO) events by each algorithm was compared with the detection of these same events by the force plate. The error between the events detected by the algorithms and those detected by the force plate was determined in milliseconds. Repeated measures ANOVA was used to compare the errors among the algorithms.

RESULTS

The algorithm reported by Ghoussayni et al. showed the best performance in all situations, except for the identification of FS events on the unaffected side in children with unilateral SCP. For these events, the algorithms reported by Desailly et al. and Zeni et al. showed the best performance.

CONCLUSION

Ghoussayni et al.'s algorithm can be used to detect gait events in children with SCP when a force plate is not available.