Walking and running cadence estimation using a single trunk-fixed accelerometer for daily physical activities assessment.

Accurate assessment of the type, duration, and intensity of physical activity (PA) in daily life is considered very important because of the close relationship between PA level, health, and well-being. Therefore, the assessment of PA using lightweight wearable sensors has gained interest in recent years. In particular, the use of activity monitors could help to measure the health-related effects of specific PA interventions. Our study, named as Run4Vit, focuses on evaluating the acute and longterm effects of an eight-week running intervention on PA behaviour and vitality. To achieve this goal, we developed an algorithm to detect running and estimate instantaneous cadence using a single trunk-fixed accelerometer. Cadence was computed using time and frequency domain approaches. Validation was performed over a wide range of locomotion speeds using an open-source gait database. Across all subjects, the cadence estimation algorithms achieved a mean bias and precision of - 0.01 ± 0.69 steps/min for the temporal method and 0.02 ± 1.33 steps/min for the frequency method. The running detection algorithm demonstrated very good performance, with an accuracy of 98% and a precision superior to 99%. These algorithms could be used to extract metrics related to the multiple dimensions of PA, and provide reliable outcome measures for the Run4Vit longitudinal running intervention program. Clinical Relevance- This work aims at validating a multi-dimensional physical activity (PA) classification algorithm for assessing the acute and long-term effects of eight weeks running intervention on PA behaviours and vitality.

Algorithms for Walking Speed Estimation Using a Lower-Back-Worn Inertial Sensor: A Cross-Validation on Speed Ranges.

Walking/gait speed is a key measure for daily mobility characterization. To date, various studies have attempted to design algorithms to estimate walking speed using an inertial sensor worn on the lower back, which is considered as a proper location for activity monitoring in daily life. However, these algorithms were rarely compared and validated on the same datasets, including people with different preferred walking speed. This study implemented several original, improved, and new algorithms for estimating cadence, step length and eventually speed. We designed comprehensive cross-validation to compare the algorithms for walking slow, normal, fast, and using walking aids. We used two datasets, including reference data for algorithm validation from an instrumented mat (40 subjects) and shanks-worn inertial sensors (88 subjects), with normal and impaired walking patterns. The results showed up to 50% performance improvements. Training of algorithms on data from people with different preferred speeds led to better performance. For the slow walkers, an average RMSE of 2.5 steps/min, 0.04 m, and 0.10 m/s were respectively achieved for cadence, step length, and speed estimation. For normal walkers, the errors were 3.5 steps/min, 0.08 m, and 0.12 m/s. An average RMSE of 1.3 steps/min, 0.05 m, and 0.10 m/s were also observed on fast walkers. For people using walking aids, the error significantly increased up to an RMSE of 14 steps/min, 0.18 m, and 0.27 m/s. The results demonstrated the robustness of the proposed combined speed estimation approach for different speed ranges. It achieved an RMSE of 0.10, 0.18, 0.15, and 0.32 m/s for slow, normal, fast, and using walking aids, respectively.

Clinical value of assessing motor performance in postacute stroke patients.

BACKGROUND: Rehabilitative treatment plans after stroke are based on clinical examinations of functional capacity and patient-reported outcomes. Objective information about daily life performance is usually not available, but it may improve therapy personalization. OBJECTIVE: To show that sensor-derived information about daily life performance is clinically valuable for counseling and the planning of rehabilitation programs for individual stroke patients who live at home. Performance information is clinically valuable if it can be used as a decision aid for the therapeutic management or counseling of individual patients. METHODS: This was an observational, cross-sectional case series including 15 ambulatory stroke patients. Motor performance in daily life was assessed with body-worn inertial sensors attached to the wrists, shanks and trunk that estimated basic physical activity and various measures of walking and arm activity in daily life. Stroke severity, motor function and activity, and degree of independence were quantified clinically by standard assessments and patient-reported outcomes. Motor performance was recorded for an average of 5.03 ± 1.1 h on the same day as the clinical assessment. The clinical value of performance information is explored in a narrative style by considering individual patient performance and capacity information. RESULTS: The patients were aged 59.9 ± 9.8 years (mean ± SD), were 6.5 ± 7.2 years post stroke, and had a National Institutes of Health Stroke Score of 4.0 ± 2.6. Capacity and performance measures showed high variability. There were substantial discrepancies between performance and capacity measures in some patients. CONCLUSIONS: This case series shows that information about motor performance in daily life can be valuable for tailoring rehabilitative therapy plans and counseling according to the needs of individual stroke patients. Although the short recording time (average of 5.03 h) limited the scope of the conclusions, this study highlights the usefulness of objective measures of daily life performance for the planning of rehabilitative therapies. Further research is required to investigate whether information about performance in daily life leads to improved rehabilitative therapy results.

Suitability of commercial barometric pressure sensors to distinguish sitting and standing activities for wearable monitoring.

Despite its medical relevance, accurate recognition of sedentary (sitting and lying) and dynamic activities (e.g. standing and walking) remains challenging using a single wearable device. Currently, trunk-worn wearable systems can differentiate sitting from standing with moderate success, as activity classifiers often rely on inertial signals at the transition period (e.g. from sitting to standing) which contains limited information. Discriminating sitting from standing thus requires additional sources of information such as elevation change. The aim of this study is to demonstrate the suitability of barometric pressure, providing an absolute estimate of elevation, for evaluating sitting and standing periods during daily activities. Three sensors were evaluated in both calm laboratory conditions and a pilot study involving seven healthy subjects performing 322 sitting and standing transitions, both indoor and outdoor, in real-world conditions. The MS5611-BA01 barometric pressure sensor (Measurement Specialties, USA) demonstrated superior performance to counterparts. It discriminates actual sitting and standing transitions from stationary postures with 99.5% accuracy and is also capable to completely dissociate Sit-to-Stand from Stand-to-Sit transitions.

Multi-parametric evaluation of sit-to-stand and stand-to-sit transitions in elderly people.

The aim of this study was to extract multi-parametric measures characterizing different features of sit-to-stand (Si-St) and stand-to-sit (St-Si) transitions in older persons, using a single inertial sensor attached to the chest. Investigated parameters were transition's duration, range of trunk tilt, smoothness of transition pattern assessed by its fractal dimension, and trunk movement's dynamic described by local wavelet energy. A measurement protocol with a Si-St followed by a St-Si postural transition was performed by two groups of participants: the first group (N=79) included Frail Elderly subjects admitted to a post-acute rehabilitation facility and the second group (N=27) were healthy community-dwelling elderly persons. Subjects were also evaluated with Tinetti's POMA scale. Compared to Healthy Elderly persons, frail group at baseline had significantly longer Si-St (3.85±1.04 vs. 2.60±0.32, p=0.001) and St-Si (4.08±1.21 vs. 2.81±0.36, p=0.001) transition's duration. Frail older persons also had significantly decreased smoothness of Si-St transition pattern (1.36±0.07 vs. 1.21±0.05, p=0.001) and dynamic of trunk movement. Measurements after three weeks of rehabilitation in frail older persons showed that smoothness of transition pattern had the highest improvement effect size (0.4) and discriminative performance. These results demonstrate the potential interest of such parameters to distinguish older subjects with different functional and health conditions.

Foot worn inertial sensors for gait assessment and rehabilitation based on motorized shoes.

Fall prevention in elderly subjects is often based on training and rehabilitation programs that include mostly traditional balance and strength exercises. By applying such conventional interventions to improve gait performance and decrease fall risk, some important factors are neglected such as the dynamics of the gait and the motor learning processes. The EU project "Self Mobility Improvement in the eLderly by counteractING falls" (SMILING project) aimed to improve age-related gait and balance performance by using unpredicted external perturbations during walking through motorized shoes that change insole inclination at each stance. This paper describes the shoe-worn inertial module and the gait analysis method needed to control in real-time the shoe insole inclination during training, as well as gait spatio-temporal parameters obtained during long distance walking before and after the 8-week training program that assessed the efficacy of training with these motorized shoes.

Nonlinear analysis of human physical activity patterns in health and disease.

The reliable and objective assessment of chronic disease state has been and still is a very significant challenge in clinical medicine. An essential feature of human behavior related to the health status, the functional capacity, and the quality of life is the physical activity during daily life. A common way to assess physical activity is to measure the quantity of body movement. Since human activity is controlled by various factors both extrinsic and intrinsic to the body, quantitative parameters only provide a partial assessment and do not allow for a clear distinction between normal and abnormal activity. In this paper, we propose a methodology for the analysis of human activity pattern based on the definition of different physical activity time series with the appropriate analysis methods. The temporal pattern of postures, movements, and transitions between postures was quantified using fractal analysis and symbolic dynamics statistics. The derived nonlinear metrics were able to discriminate patterns of daily activity generated from healthy and chronic pain states.

Kinematics and dynamic complexity of postural transitions in frail elderly subjects.

The aim of this study was to propose a methodology allowing a detailed characterization of body sit-to-stand/stand-to-sit postural transition. Parameters characterizing the kinematics of the trunk movement during sit-to-stand (Si-St) postural transition were calculated using one initial sensor system fixed on the trunk and a data logger. Dynamic complexity of these postural transitions was estimated by fractal dimension of acceleration-angular velocity plot. We concluded that this method provides a simple and accurate tool for monitoring frail elderly and to objectively evaluate the efficacy of a rehabilitation program.

Ambulatory system for the quantitative and qualitative analysis of gait and posture in chronic pain patients treated with spinal cord stimulation.

The physical activity in normal daily life is determined to a large extent by the functional ability of a subject. As a result, the measurement of the physical activity that a subject performs spontaneously could be a useful and objective measurement of disability, particularly in patients with disease-related functional impairment. The aim of this study is to provide an accurate method for the measurement and analysis of the physical activity under normal life conditions. Using three kinematical sensors strapped to the body, both the posture and the gait parameters can be assessed qualitatively and quantitatively. A detailed description of the algorithms used to analyse both the posture and the gait are presented in this paper. Two methods, based on different sensor configurations and signal processing, are proposed for the detection of sitting and standing postures (Methods P1 and P2). Two other methods are used for the quantitative assessment of walking (Methods W1 and W2). The performance of the algorithms (expressed in terms of sensitivity, specificity and error) is based on the comparison of data recorded simultaneously by a non-interfering observer (reference data) with the data provided by the recording system (21 patients, 61 h). Sensitivity and specificity are respectively 98.2% and 98.8% (P1), 97.8% and 98.1% (P2) for sitting; 98.0% and 98.5% (P1), 97.4% and 97.8% (P2) for standing; 97.1% and 97.9% (W1), 92.4% and 94.9% (W2) for walking; and finally, 99.2% and 98.6% for lying. Overall detection errors (as a percent of range) are as follows: 1.15% (P1) and 1.20% (P2) for sitting, 1.36% (P1) and 1.40% (P2) for standing, 1.20% (W1) and 1.60% (W2) for walking and 0.40% for lying. The error for the estimated walking distance and the speed is 6.8% and 9.6%, respectively. We conclude that both methods can be used for the accurate measurement of the basic physical activity in normal daily life. Measurements performed before and after the delivery of a treatment can therefore provide information of unprecedented accuracy and objectivity on the ability of a procedure, in this case spinal cord stimulation, to restore functional capabilities.