A wrist sensor and algorithm to determine instantaneous walking cadence and speed in daily life walking

Classification of human walking context using a single-point accelerometer

Real-world walking data offers rich insights into a person's mobility. Yet, daily life variations can alter these patterns, making the data challenging to interpret. As such, it is essential to integrate context for the extraction of meaningful information from real-world movement data. In this work, we leveraged the relationship between the characteristics of a walking bout and context to build a classification algorithm to distinguish between indoor and outdoor walks. We used data from 20 participants wearing an accelerometer on the thigh over a week. Their walking bouts were isolated and labeled using GPS and self-reporting data. We trained and validated two machine learning models, random forest and ensemble Support Vector Machine, using a leave-one-participant-out validation scheme on 15 subjects. The 5 remaining subjects were used as a testing set to choose a final model. The chosen model achieved an accuracy of 0.941, an F1-score of 0.963, and an AUROC of 0.931. This validated model was then used to label the walks from a different dataset with 15 participants wearing the same accelerometer. Finally, we characterized the differences between indoor and outdoor walks using the ensemble of the data. We found that participants walked significantly faster, longer, and more continuously when walking outdoors compared to indoors. These results demonstrate how movement data alone can be used to obtain accurate information on important contextual factors. These factors can then be leveraged to enhance our understanding and interpretation of real-world movement data, providing deeper insights into a person's health.

The Analytical and Clinical Validity of the pfSTEP Digital Biomarker of the Susceptibility/Risk of Declining Physical Function in Community-Dwelling Older Adults

Measures of stepping volume and rate are common outputs from wearable devices, such as accelerometers. It has been proposed that biomedical technologies, including accelerometers and their algorithms, should undergo rigorous verification as well as analytical and clinical validation to demonstrate that they are fit for purpose. The aim of this study was to use the V3 framework to assess the analytical and clinical validity of a wrist-worn measurement system of stepping volume and rate, formed by the GENEActiv accelerometer and GENEAcount step counting algorithm. The analytical validity was assessed by measuring the level of agreement between the wrist-worn system and a thigh-worn system (activPAL), the reference measure. The clinical validity was assessed by establishing the prospective association between the changes in stepping volume and rate with changes in physical function (SPPB score). The agreement of the thigh-worn reference system and the wrist-worn system was excellent for total daily steps (CCC = 0.88, 95% CI 0.83-0.91) and moderate for walking steps and faster-paced walking steps (CCC = 0.61, 95% CI 0.53-0.68 and 0.55, 95% CI 0.46-0.64, respectively). A higher number of total steps and faster paced-walking steps was consistently associated with better physical function. After 24 months, an increase of 1000 daily faster-paced walking steps was associated with a clinically meaningful increase in physical function (0.53 SPPB score, 95% CI 0.32-0.74). We have validated a digital susceptibility/risk biomarker-pfSTEP-that identifies an associated risk of low physical function in community-dwelling older adults using a wrist-worn accelerometer and its accompanying open-source step counting algorithm.

Free-Living Gait Cadence Measured by Wearable Accelerometer: A Promising Alternative to Traditional Measures of Mobility for Assessing Fall Risk

BACKGROUND

Wearable devices have become widespread in research applications, yet evidence on whether they are superior to structured clinic-based assessments is sparse. In this manuscript, we compare traditional, laboratory-based metrics of mobility with a novel accelerometry-based measure of free-living gait cadence for predicting fall rates.

METHODS

Using negative binomial regression, we compared traditional in-clinic measures of mobility (6-minute gait cadence, speed, and distance, and 4-m gait speed) with free-living gait cadence from wearable accelerometers in predicting fall rates. Accelerometry data were collected with wrist-worn Actigraphs (GT9X) over 7 days in 432 community-dwelling older adults (aged 77.29 ± 5.46 years, 59.1% men, 80.2% White) participating in the Study to Understand Fall Reduction and Vitamin D in You. Falls were ascertained using monthly calendars, quarterly contacts, and ad hoc telephone reports. Accelerometry-based free-living gait cadence was estimated with the Adaptive Empirical Pattern Transformation algorithm.

RESULTS

Across all participants, free-living cadence was significantly related to fall rates; every 10 steps per minute higher cadence was associated with a 13.2% lower fall rate (p = .036). Clinic-based measures of mobility were not related to falls (p > .05). Among higher-functioning participants (cadence ≥100 steps/minute), every 10 steps per minute higher free-living cadence was associated with a 27.7% lower fall rate (p = .01). In participants with slow baseline gait (gait speed <0.8 m/s), all metrics were significantly associated with fall rates.

CONCLUSION

Data collected from biosensors in the free-living environment may provide a more sensitive indicator of fall risk than in-clinic tests, especially among higher-functioning older adults who may be more responsive to intervention.

CLINICAL TRIAL REGISTRATION

NCT02166333.

Analytical and clinical validity of wearable, multi-sensor technology for assessment of motor function in patients with Parkinson's disease in Japan

Continuous, objective monitoring of motor signs and symptoms may help improve tracking of disease progression and treatment response in Parkinson's disease (PD). This study assessed the analytical and clinical validity of multi-sensor smartwatch measurements in hospitalized and home-based settings (96 patients with PD; mean wear time 19 h/day) using a twice-daily virtual motor examination (VME) at times representing medication OFF/ON states. Digital measurement performance was better during inpatient clinical assessments for composite V-scores than single-sensor-derived features for bradykinesia (Spearman |r|= 0.63, reliability = 0.72), tremor (|r|= 0.41, reliability = 0.65), and overall motor features (|r|= 0.70, reliability = 0.67). Composite levodopa effect sizes during hospitalization were 0.51-1.44 for clinical assessments and 0.56-1.37 for VMEs. Reliability of digital measurements during home-based VMEs was 0.62-0.80 for scores derived from weekly averages and 0.24-0.66 for daily measurements. These results show that unsupervised digital measurements of motor features with wrist-worn sensors are sensitive to medication state and are reliable in naturalistic settings.Trial Registration: Japan Pharmaceutical Information Center Clinical Trials Information (JAPIC-CTI): JapicCTI-194825; Registered June 25, 2019.

Steps parameters of elderly patients hospitalised for an acute medical illness in a Swiss University Hospital: a monocentric observational pilot-study

BACKGROUND

Objective mobility goals for elderly hospitalised medical patients remain debated. We therefore studied steps parameters of elderly patients hospitalised for an acute illness, to determine goals for future interventional trials and medical practice.

METHODS

Observational study conducted from February to November 2018 in a medical ward of the Lausanne University Hospital, Switzerland. We measured the step parameters of consecutive medical patients aged ≥65 years admitted for an acute medical illness using a wrist accelerometer (Geneactiv). We also collected demographic, somatic and functional factors.

RESULTS

Overall, 187 inpatients had their step parameters (daily step count, walking cadence and bout duration) measured with accelerometers worn for a mean of 3.6 days (standard deviation [SD] 3.2). Elderly inpatients (81.5 years, SD 8.5) walked a median of 603 steps daily (interquartile range [IQR] 456-809), at a median cadence of 100 steps/minute (IQR 99-101) with median walking bouts of 33 seconds (IQR 27-37) and with 70% of the walking bouts lasting less than 30 seconds. Patients walking ≥600 steps were younger (80.4 years, SD 8.9 vs 82.8 years,SD 7.9, p = 0.050) and had a longer length of stay (7.8 days, SD 5.1 vs 6.1 days, SD 4.1, p = 0.011) than those walking <600 steps. Patients at high risk of bed sores walked less (564 steps, IQR 394-814 vs 626, IQR 526-840) than those with a lower risk of sores.

CONCLUSION

During a hospitalisation for an acute medical illness, patients aged ≥65 years walk a mere 603 steps daily and most of the time for periods of less than 30 seconds. This information should be used to build up future interventional trials or to set mobility goals for patients hospitalised in Swiss hospitals.

Development and large-scale validation of the Watch Walk wrist-worn digital gait biomarkers

Digital gait biomarkers (including walking speed) indicate functional decline and predict hospitalization and mortality. However, waist or lower-limb devices often used are not designed for continuous life-long use. While wrist devices are ubiquitous and many large research repositories include wrist-sensor data, widely accepted and validated digital gait biomarkers derived from wrist-worn accelerometers are not available yet. Here we describe the development of advanced signal processing algorithms that extract digital gait biomarkers from wrist-worn devices and validation using 1-week data from 78,822 UK Biobank participants. Our gait biomarkers demonstrate good test-retest-reliability, strong agreement with electronic walkway measurements of gait speed and self-reported pace and significantly discriminate individuals with poor self-reported health. With the almost universal uptake of smart-watches, our algorithms offer a new approach to remotely monitor life-long population level walking speed, quality, quantity and distribution, evaluate disease progression, predict risk of adverse events and provide digital gait endpoints for clinical trials.

Detecting Gait Events from Accelerations Using Reservoir Computing

Segmenting the gait cycle into multiple phases using gait event detection (GED) is a well-researched subject with many accurate algorithms. However, the algorithms that are able to perform accurate and robust GED for real-life environments and physical diseases tend to be too complex for their implementation on simple hardware systems limited in computing power and memory, such as those used in wearable devices. This study focuses on a numerical implementation of a reservoir computing (RC) algorithm called the echo state network (ESN) that is based on simple computational steps that are easy to implement on portable hardware systems for real-time detection. RC is a neural network method that is widely used for signal processing applications and uses a fast-training method based on a ridge regression adapted to the large quantity and variety of IMU data needed to use RC in various real-life environment GED. In this study, an ESN was used to perform offline GED with gait data from IMU and ground force sensors retrieved from three databases for a total of 28 healthy adults and 15 walking conditions. Our main finding is that despite its low complexity, ESN is robust for GED, with performance comparable to other state-of-the-art algorithms. Our results show the ESN is robust enough to obtain good detection results in all conditions if the algorithm is trained with variable data that match those conditions. The distribution of the mean absolute errors (MAE) between the detection times from the ESN and the force sensors were between 40 and 120 ms for 6 defined gait events (95th percentile). We compared our ESN with four different state-of-the-art algorithms from the literature. The ESN obtained a MAE not more than 10 ms above three other reference algorithms for normal walking indoor and outdoor conditions and yielded the 2nd lowest MAE and the 2nd highest true positive rate and specificity when applied to outdoor walking and running conditions. Our work opens the door to using the ESN as a GED for applications in wearable sensors for long-term patient monitoring.

Gait Detection from a Wrist-Worn Sensor Using Machine Learning Methods: A Daily Living Study in Older Adults and People with Parkinson's Disease

Remote assessment of the gait of older adults (OAs) during daily living using wrist-worn sensors has the potential to augment clinical care and mobility research. However, hand movements can degrade gait detection from wrist-sensor recordings. To address this challenge, we developed an anomaly detection algorithm and compared its performance to four previously published gait detection algorithms. Multiday accelerometer recordings from a wrist-worn and lower-back sensor (i.e., the “gold-standard” reference) were obtained in 30 OAs, 60% with Parkinson’s disease (PD). The area under the receiver operator curve (AUC) and the area under the precision−recall curve (AUPRC) were used to evaluate the performance of the algorithms. The anomaly detection algorithm obtained AUCs of 0.80 and 0.74 for OAs and PD, respectively, but AUPRCs of 0.23 and 0.31 for OAs and PD, respectively. The best performing detection algorithm, a deep convolutional neural network (DCNN), exhibited high AUCs (i.e., 0.94 for OAs and 0.89 for PD) but lower AUPRCs (i.e., 0.66 for OAs and 0.60 for PD), indicating trade-offs between precision and recall. When choosing a classification threshold of 0.9 (i.e., opting for high precision) for the DCNN algorithm, strong correlations (r > 0.8) were observed between daily living walking time estimates based on the lower-back (reference) sensor and the wrist sensor. Further, gait quality measures were significantly different in OAs and PD compared to healthy adults. These results demonstrate that daily living gait can be quantified using a wrist-worn sensor.

Identifying and characterising sources of variability in digital outcome measures in Parkinson's disease

Smartphones and wearables are widely recognised as the foundation for novel Digital Health Technologies (DHTs) for the clinical assessment of Parkinson's disease. Yet, only limited progress has been made towards their regulatory acceptability as effective drug development tools. A key barrier in achieving this goal relates to the influence of a wide range of sources of variability (SoVs) introduced by measurement processes incorporating DHTs, on their ability to detect relevant changes to PD. This paper introduces a conceptual framework to assist clinical research teams investigating a specific Concept of Interest within a particular Context of Use, to identify, characterise, and when possible, mitigate the influence of SoVs. We illustrate how this conceptual framework can be applied in practice through specific examples, including two data-driven case studies.

Validation of Potential Reference Measure for Indoor Walking Distance to Evaluate Wearable Sensing Devices. ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. ANNUAL INTERNATIONAL CONFERENCE 2021;2021:7178-7181. [PMID: 34892756 DOI: 10.1109/embc46164.2021.9629854] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

The walking distance estimated from the coordinate position information of the center of mass obtained via Xsens MTw Awinda were validated from 5 adult volunteers and the accuracy was shown significantly high. (Average absolute error of -1.22% with a standard deviation of 2.26%).

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.

Assessment of Physical Activity in Adults using Wrist Accelerometers

The health benefits of physical activity have been widely recognized, yet traditional measures of physical activity including questionnaires and category-based assessments of volume and intensity provide only broad estimates of daily activities. Accelerometers have advanced epidemiologic research on physical activity by providing objective and continuous measurement of physical activity in free-living conditions. Wrist-worn accelerometers have become especially popular due to low participant burden. However, the validity and reliability of wrist-worn devices for adults have yet to be summarized. Moreover, accelerometer data provide rich information on how physical activity is accumulated throughout the day, but only a small portion of these rich data have been utilized by researchers. Lastly, new methodological developments that aim to overcome some of the limitations of accelerometers are emerging. The purpose of this review is to provide an overview of accelerometry research, with a special focus on wrist-worn accelerometers. We describe briefly how accelerometers work, summarize the validity and reliability of wrist-worn accelerometers, discuss the benefits of accelerometers including measuring light-intensity physical activity, and discuss pattern metrics of daily physical activity recently introduced in the literature. A summary of large-scale cohort studies and randomized trials that implemented wrist-worn accelerometry is provided. We conclude the review by discussing new developments and future directions of research using accelerometers, with a focus on wrist-worn accelerometers.

Toward a remote assessment of walking bout and speed: application in patients with multiple sclerosis

Gait speed as a powerful biomarker of mobility is mostly assessed in controlled environments, e.g. in the clinic. With wearable inertial sensors, gait speed can be estimated in an objective manner. However, most of the previous works have validated the gait speed estimation algorithms in clinical settings which can be different than the home assessments in which the patients demonstrate their actual performance. Moreover, to provide comfort for the users, devising an algorithm based on a single sensor setup is essential. To this end, the goal of this study was to develop and validate a new gait speed estimation method based on a machine learning approach to predict gait speed in both clinical and home assessments by a sensor on the lower back. Moreover, two methods were introduced to detect walking bouts during daily activities at home. We have validated the algorithms in 35 patients with multiple sclerosis as it often presents with mobility difficulties. Therefore, the robustness of the algorithm can be shown in an impaired or slow gait. Against silver standard multi-sensor references, we achieved a bias close to zero and a precision of 0.15 m/s for gait speed estimation. Furthermore, the proposed machine learning-based locomotion detection method had a median of 96.8% specificity, 93.0% sensitivity, 96.4% accuracy, and 78.6% F1-score in detecting walking bouts at home. The high performance of the proposed algorithm showed the feasibility of the unsupervised mobility assessment introduced in this study.

Running Speed Estimation Using Shoe-Worn Inertial Sensors: Direct Integration, Linear, and Personalized Model

The overground speed is a key component of running analysis. Today, most speed estimation wearable systems are based on GNSS technology. However, these devices can suffer from sparse communication with the satellites and have a high-power consumption. In this study, we propose three different approaches to estimate the overground speed in running based on foot-worn inertial sensors and compare the results against a reference GNSS system. First, a method is proposed by direct strapdown integration of the foot acceleration. Second, a feature-based linear model and finally a personalized online-model based on the recursive least squares' method were devised. We also evaluated the performance differences between two sets of features; one automatically selected set (i.e., optimized) and a set of features based on the existing literature. The data set of this study was recorded in a real-world setting, with 33 healthy individuals running at low, preferred, and high speed. The direct estimation of the running speed achieved an inter-subject mean ± STD accuracy of 0.08 ± 0.1 m/s and a precision of 0.16 ± 0.04 m/s. In comparison, the best feature-based linear model achieved 0.00 ± 0.11 m/s accuracy and 0.11 ± 0.05 m/s precision, while the personalized model obtained a 0.00 ± 0.01 m/s accuracy and 0.09 ± 0.06 m/s precision. The results of this study suggest that (1) the direct estimation of the velocity of the foot are biased, and the error is affected by the overground velocity and the slope; (2) the main limitation of a general linear model is the relatively high inter-subject variance of the bias, which reflects the intrinsic differences in gait patterns among individuals; (3) this inter-subject variance can be nulled using a personalized model.

Using Different Combinations of Body-Mounted IMU Sensors to Estimate Speed of Horses-A Machine Learning Approach

Speed is an essential parameter in biomechanical analysis and general locomotion research. It is possible to estimate the speed using global positioning systems (GPS) or inertial measurement units (IMUs). However, GPS requires a consistent signal connection to satellites, and errors accumulate during IMU signals integration. In an attempt to overcome these issues, we have investigated the possibility of estimating the horse speed by developing machine learning (ML) models using the signals from seven body-mounted IMUs. Since motion patterns extracted from IMU signals are different between breeds and gaits, we trained the models based on data from 40 Icelandic and Franches-Montagnes horses during walk, trot, tölt, pace, and canter. In addition, we studied the estimation accuracy between IMU locations on the body (sacrum, withers, head, and limbs). The models were evaluated per gait and were compared between ML algorithms and IMU location. The model yielded the highest estimation accuracy of speed (RMSE = 0.25 m/s) within equine and most of human speed estimation literature. In conclusion, highly accurate horse speed estimation models, independent of IMU(s) location on-body and gait, were developed using ML.

Indoor vs. Outdoor Walking: Does It Make Any Difference in Joint Angle Depending on Road Surface?

Measurement of the joint angle during walking in real-world environments facilitates comprehension of the adaptation strategy corresponding to road surfaces. This study investigated the differences between the joint angles in the lower limb when walking on flat road surfaces in indoor and outdoor environments. Ten healthy young males who walked on a carpet-lined corridor in the indoor environment and on an interlocking block pavement surface in the outdoor environment participated in the study. The joint angles of their lower limbs were measured using seven inertial measurement units, and the average and coefficient of variation (%CV) of the joint angular excursion in the two environments were evaluated. The %CVs of the ankle plantar flexion excursion in the early stance was 45% higher in the outdoor environment compared with that in the indoor, although the spatiotemporal parameters and joint angular excursion of the proximal joints showed no difference between the environments. Though the road surfaces were flat from a macroscopic point of view, the interlocking block pavement had stiffer and more irregular characteristics. The variability in the ankle plantar flexion motion in the early stance may be most likely affected by these surface characteristics in the real-world outdoor environment.

Robust Step Detection from Different Waist-Worn Sensor Positions: Implications for Clinical Studies

Analyzing human gait with inertial sensors provides valuable insights into a wide range of health impairments, including many musculoskeletal and neurological diseases. A representative and reliable assessment of gait requires continuous monitoring over long periods and ideally takes place in the subjects' habitual environment (real-world). An inconsistent sensor wearing position can affect gait characterization and influence clinical study results, thus clinical study protocols are typically highly proscriptive, instructing all participants to wear the sensor in a uniform manner. This restrictive approach improves data quality but reduces overall adherence. In this work, we analyze the impact of altering the sensor wearing position around the waist on sensor signal and step detection. We demonstrate that an asymmetrically worn sensor leads to additional odd-harmonic frequency components in the frequency spectrum. We propose a robust solution for step detection based on autocorrelation to overcome sensor position variation (sensitivity = 0.99, precision = 0.99). The proposed solution reduces the impact of inconsistent sensor positioning on gait characterization in clinical studies, thus providing more flexibility to protocol implementation and more freedom to participants to wear the sensor in the position most comfortable to them. This work is a first step towards truly position-agnostic gait assessment in clinical settings.

Decreasing the ambulatory knee adduction moment without increasing the knee flexion moment individually through modifications in footprint parameters: A feasibility study for a dual kinetic change in healthy subjects

Gait retraining is gaining in interest to reduce loading associated to knee osteoarthritis (OA) progression. So far, interventions focused on reducing the peak knee adduction moment (pKAM) and it remains unclear if this can be done individually without increasing the peak knee flexion moment (pKFM). Additionally, while modifying foot progression angle (FPA) and step width (SW) is common, little is known about modifications in stride length (SL). This study aimed at characterizing the feasibility of a dual kinetic change, consisting in reducing the pKAM by at least 10% without increasing the pKFM. It also aimed to evaluate the added value of SL modifications in achieving the dual kinetic change. Gait trials with modifications in FPA, SW and SL were recorded for 11 young healthy subjects in a laboratory equipped with an augmented-reality system displaying instruction footprints on the floor. All participants achieved the dual kinetic change with at least one of the modifications. Seven participants achieved it with FPA modification, three with SW modification, and seven with SL modification. In conclusion, this study showed that it is feasible to achieve the dual kinetic change individually through subject-specific modifications in footprint parameters, suggesting that, in the future, gait retraining could aim for more specific kinetic changes than simply pKAM reductions. Modifying SL allowed achieving the dual kinetic change, stressing out the value of this parameter for gait retraining, in addition to FPA and SW. Finally, an augmented-reality approach was introduced to help footprint parameter modifications in the framework of knee OA.

Real-world speed estimation using single trunk IMU: methodological challenges for impaired gait patterns

Walking speed (WS) is recognized as an important dimension of functional health and a candidate endpoint for clinical trials. To be adopted as a powerful outcome measure in clinical assessment, WS should be estimated pervasively and accurately in the real-life context. Although current state of the art points to possible solutions, e.g., by using pairing of wearable sensors with dedicated algorithms, the accuracy and robustness of existing algorithms in challenging situations should be carefully considered. This study highlights the main methodological issues for WS estimation using single inertial sensor fixed on trunk (chest/low back) and data recorded in a sample of stroke patients with impaired mobility.

Deep Learning in Gait Parameter Prediction for OA and TKA Patients Wearing IMU Sensors

Quantitative assessments of patient movement quality in osteoarthritis (OA), specifically spatiotemporal gait parameters (STGPs), can provide in-depth insight into gait patterns, activity types, and changes in mobility after total knee arthroplasty (TKA). A study was conducted to benchmark the ability of multiple deep neural network (DNN) architectures to predict 12 STGPs from inertial measurement unit (IMU) data and to identify an optimal sensor combination, which has yet to be studied for OA and TKA subjects. DNNs were trained using movement data from 29 subjects, walking at slow, normal, and fast paces and evaluated with cross-fold validation over the subjects. Optimal sensor locations were determined by comparing prediction accuracy with 15 IMU configurations (pelvis, thigh, shank, and feet). Percent error across the 12 STGPs ranged from 2.1% (stride time) to 73.7% (toe-out angle) and overall was more accurate in temporal parameters than spatial parameters. The most and least accurate sensor combinations were feet-thighs and singular pelvis, respectively. DNNs showed promising results in predicting STGPs for OA and TKA subjects based on signals from IMU sensors and overcomes the dependency on sensor locations that can hinder the design of patient monitoring systems for clinical application.

Validity of a Smart-Glasses-Based Step-Count Measure during Simulated Free-Living Conditions

Step counting represents a valuable approach to monitor the amount of daily physical activity. The feet, wrist and trunk have been demonstrated as the ideal locations to automatically detect the number of steps through body-worn devices (i.e., step counters). Key features of such devices are high usability, practicality and unobtrusiveness. Therefore, the opportunity to integrate step-counting functions in daily worn accessories represents one of the recent and most important challenges. In this context, the present study aimed to investigate the validity of a smart-glasses-based step-counter measure by comparing their performances against the most popular commercial step counters. To this purpose, smart glasses data from 26 healthy subjects performing simulated free-living walking conditions along a predefined path were collected. Reference measures from inertial sensors mounted on the subjects’ ankles and data from commercial (waist- and wrists-worn) step counters were acquired during the tests. The results showed an overall percentage error of 1%. In conclusion, the proposed smart glasses could be considered an accurate step counter, showing performances comparable to the most common commercial step counters.

Detection of Gait From Continuous Inertial Sensor Data Using Harmonic Frequencies

Mobile gait analysis using wearable inertial measurement units (IMUs) provides valuable insights for the assessment of movement impairments in different neurological and musculoskeletal diseases, for example Parkinson's disease (PD). The increase in data volume due to arising long-term monitoring requires valid, robust and efficient analysis pipelines. In many studies an upstream detection of gait is therefore applied. However, current methods do not provide a robust way to successfully reject non-gait signals. Therefore, we developed a novel algorithm for the detection of gait from continuous inertial data of sensors worn at the feet. The algorithm is focused not only on a high sensitivity but also a high specificity for gait. Sliding windows of IMU signals recorded from the feet of PD patients were processed in the frequency domain. Gait was detected if the frequency spectrum contained specific patterns of harmonic frequencies. The approach was trained and evaluated on 150 clinical measurements containing standardized gait and cyclic movement tests. The detection reached a sensitivity of 0.98 and a specificity of 0.96 for the best sensor configuration (angular rate around the medio-lateral axis). On an independent validation data set including 203 unsupervised, semi-standardized gait tests, the algorithm achieved a sensitivity of 0.97. Our algorithm for the detection of gait from continuous IMU signals works reliably and showed promising results for the application in the context of free-living and non-standardized monitoring scenarios.

Changes in spatio-temporal gait parameters and vertical speed during an extreme mountain ultra-marathon

The aim of the present study was to investigate the effects of altitude and distance on uphill vertical speed (VS) and the main spatio-temporal gait parameters during an extreme mountain ultra-marathon. The VS, stride height (SH) and stride frequency (SF) of 27 runners were measured with an inertial sensor at the shank for two different altitude ranges (low 1300-2000 m vs high 2400-3200 m) of 10 mountains passes distributed over a 220 km course. There was a significant interaction (F(4,52) = 4.04, p < 0.01) for the effect of altitude and distance on VS. During the first passes, the mean VS was faster at lower altitudes, but this difference disappeared at a quarter of the race length, suggesting that neuromuscular fatigue influenced the uphill velocity to a larger extent than the oxygen delivery. The average VS, SH and SF were 547 ± 135 m/h, 0.23 ± 0.05 m and 0.66 ± 0.09 Hz. The individual VS change for each uphill portions was more strongly correlated with the changes in SH (r = 0.80, P < 0.001, n = 321) than SF (r = 0.43, P < 0.001, n = 321). This suggests a large effect of the knee extensors strength loss on the diminution of VS.

Design a prototype for automated patient diagnosis in wireless sensor networks

It is indeed necessary to design of an elderly support mobile healthcare and monitoring system on wireless sensor network (WSN) for dynamic monitoring. It comes from the need for maintenance of healthcare among patients and elderly people that leads to the demand on change in traditional monitoring approaches among chronic disease patients and alert on acute events. In this paper, we propose a new automated patient diagnosis called automated patient diagnosis (AUPA) using ATmega microcontrollers over environmental sensors. AUPA monitors and aggregates data from patients through network connected over web server and mobile network. The scheme supports variable data management and route establishment. Data transfer is established using adaptive route discovery and management approaches. AUPA supports minimizing packet loss and delay, handling erroneous data, and providing optimized decision-making for healthcare support. The performance of AUPA's QoS approach is tested using a set of health-related sensors which gather the patient's data over variable period of time and send from a source to destination AUPA node. Experimental results show that AUPA outperforms the existing schemes, namely SPIN and LEACH, with minimal signal loss rate and a better neighborhood node selection and link selection. It diminishes the jitter compared to the related algorithms. Graphical abstract Stack architecture of AUPA.

Real-World Gait Speed Estimation Using Wrist Sensor: A Personalized Approach

Gait speed is an important parameter to characterize people's daily mobility. For real-world speed measurement, inertial sensors or global navigation satellite system (GNSS) can be used on wrist, possibly integrated in a wristwatch. However, power consumption of GNSS is high and data are only available outdoor. Gait speed estimation using wrist-mounted inertial sensors is generally based on machine learning and suffers from low accuracy because of the inadequacy of using limited training data to build a general speed model that would be accurate for the whole population. To overcome this issue, a personalized model was proposed, which took unique gait style of each subject into account. Cadence and other biomechanically derived gait features were extracted from a wrist-mounted accelerometer and barometer. Gait features were fused with few GNSS data (sporadically sampled during gait) to calibrate the step length model of each subject through online learning. The proposed method was validated on 30 healthy subjects where it has achieved a median [Interquartile Range] of root mean square error of 0.05 [0.04-0.06] (m/s) and 0.14 [0.11-0.17] (m/s) for walking and running, respectively. Results demonstrated that the personalized model provided similar performance as GNSS. It used 50 times less training GNSS data than nonpersonalized method and achieved even better results. This parsimonious GNSS usage allowed extending battery life. The proposed algorithm met requirements for applications which need accurate, long, real-time, low-power, and indoor/outdoor speed estimation in daily life.

Locomotion and cadence detection using a single trunk-fixed accelerometer: validity for children with cerebral palsy in daily life-like conditions

BACKGROUND

Physical therapy interventions for ambulatory youth with cerebral palsy (CP) often focus on activity-based strategies to promote functional mobility and participation in physical activity. The use of activity monitors validated for this population could help to design effective personalized interventions by providing reliable outcome measures. The objective of this study was to devise a single-sensor based algorithm for locomotion and cadence detection, robust to atypical gait patterns of children with CP in the real-life like monitoring conditions.

METHODS

Study included 15 children with CP, classified according to Gross Motor Function Classification System (GMFCS) between levels I and III, and 11 age-matched typically developing (TD). Six IMU devices were fixed on participant's trunk (chest and low back/L5), thighs, and shanks. IMUs on trunk were independently used for development of algorithm, whereas the ensemble of devices on lower limbs were used as reference system. Data was collected according to a semi-structured protocol, and included typical daily-life activities performed indoor and outdoor. The algorithm was based on detection of peaks associated to heel-strike events, identified from the norm of trunk acceleration signals, and included several processing stages such as peak enhancement and selection of the steps-related peaks using heuristic decision rules. Cadence was estimated using time- and frequency-domain approaches. Performance metrics were sensitivity, specificity, precision, error, intra-class correlation coefficient, and Bland-Altman analysis.

RESULTS

According to GMFCS, CP children were classified as GMFCS I (n = 7), GMFCS II (n = 3) and GMFCS III (n = 5). Mean values of sensitivity, specificity and precision for locomotion detection ranged between 0.93-0.98, 0.92-0.97 and 0.86-0.98 for TD, CP-GMFCS I and CP-GMFCS II-III groups, respectively. Mean values of absolute error for cadence estimation (steps/min) were similar for both methods, and ranged between 0.51-0.88, 1.18-1.33 and 1.94-2.3 for TD, CP-GMFCS I and CP-GMFCS II-III groups, respectively. The standard deviation was higher in CP-GMFCS II-III group, the lower performances being explained by the high variability of atypical gait patterns.

CONCLUSIONS

The algorithm demonstrated good performance when applied to a wide range of gait patterns, from normal to the pathological gait of highly affected children with CP using walking aids.

What is the Best Configuration of Wearable Sensors to Measure Spatiotemporal Gait Parameters in Children with Cerebral Palsy?

Wearable inertial devices have recently been used to evaluate spatiotemporal parameters of gait in daily life situations. Given the heterogeneity of gait patterns in children with cerebral palsy (CP), the sensor placement and analysis algorithm may influence the validity of the results. This study aimed at comparing the spatiotemporal measurement performances of three wearable configurations defined by different sensor positioning on the lower limbs: (1) shanks and thighs, (2) shanks, and (3) feet. The three configurations were selected based on their potential to be used in daily life for children with CP and typically developing (TD) controls. For each configuration, dedicated gait analysis algorithms were used to detect gait events and compute spatiotemporal parameters. Fifteen children with CP and 11 TD controls were included. Accuracy, precision, and agreement of the three configurations were determined in comparison with an optoelectronic system as a reference. The three configurations were comparable for the evaluation of TD children and children with a low level of disability (CP-GMFCS I) whereas the shank-and-thigh-based configuration was more robust regarding children with a higher level of disability (CP-GMFCS II–III).

Detecting Steps Walking at very Low Speeds Combining Outlier Detection, Transition Matrices and Autoencoders from Acceleration Patterns

In this paper, we develop and validate a new algorithm to detect steps while walking at speeds between 30 and 40 steps per minute based on the data sensed from a single tri-axial accelerometer. The algorithm concatenates three consecutive phases. First, an outlier detection is performed on the sensed data based on the Mahalanobis distance to pre-detect candidate points in the acceleration time series that may contain a ground contact segment of data while walking. Second, the acceleration segment around the pre-detected point is used to calculate the transition matrix in order to capture the time dependencies. Finally, autoencoders, trained with data segments containing ground contact transition matrices from acceleration series from labeled steps are used to reconstruct the computed transition matrices at each pre-detected point. A similarity index is used to assess if the pre-selected point contains a true step in the 30–40 steps per minute speed range. Our experimental results, based on a database from three different participants performing similar activities to the target one, are able to achieve a recall = 0.88 with precision = 0.50 improving the results when directly applying the autoencoders to acceleration patterns (recall = 0.77 with precision = 0.50).