Search

Mobile health technologies enable continuous, quantitative assessment of mobility and gait in real-world environments, facilitating early diagnoses of gait disorders, disease progression monitoring, and prediction of adverse events like falls. Traditionally, mobile gait assessment predominantly relied on body-fixed sensors positioned at the feet or lower trunk. Here, we investigate the potential of an algorithm utilizing an ear-worn motion sensor for spatiotemporal segmentation of gait patterns. We collected 3D acceleration profiles from the ear-worn sensor during varied walking speeds in 53 healthy adults. Temporal convolutional networks were trained to detect stepping sequences and predict spatial relations between steps. The resulting algorithm, mEar, accurately detects initial and final ground contacts (F1 score of 99% and 91%, respectively). It enables the determination of temporal and spatial gait cycle characteristics (among others, stride time and stride length) with good to excellent validity at a precision sufficient to monitor clinically relevant changes in walking speed, stride-to-stride variability, and side asymmetry. This study highlights the ear as a viable site for monitoring gait and proposes its potential integration with in-ear vital-sign monitoring. Such integration offers a practical approach to comprehensive health monitoring and telemedical applications, by integrating multiple sensors in a single anatomical location. [ABSTRACT FROM AUTHOR]

This paper presents the development of an analytical model of an internal vibrating ring gyroscope in a Microelectromechanical System (MEMS). The internal ring structure consists of eight semicircular beams that are attached to the externally placed anchors. This research work analyzes the vibrating ring gyroscope's in-plane displacement behavior and the resulting elliptical vibrational modes. The elliptical vibrational modes appear as pairs with the same resonance frequency due to the symmetric structure of the design. The analysis commences by conceptualizing the ring as a geometric structure with a circular shape possessing specific dimensions such as thickness, height, and radius. We construct a linear model that characterizes the vibrational dynamics of the internal vibrating ring. The analysis develops a comprehensive mathematical formulation for the radial and tangential displacements in local polar coordinates by considering the inextensional displacement of the ring structure. By utilizing the derived motion equations, we highlight the underlying relationships driving the vibrational characteristics of the MEMS' vibrating ring gyroscope. These dynamic vibrational relationships are essential in enabling the vibrating ring gyroscope's future utilization in accurate navigation and motion sensing technologies. [ABSTRACT FROM AUTHOR]

Background: In this study, we hypothesized that safety footwear (SF) impacts gait patterns, potentially contributing to the podiatric symptoms reported by workers. The purpose of this work was to compare the gait analyses of workers wearing SF and sneakers using inertial sensors while also examining the occurrence of foot problems. Methods: A consecutive cohort of workers from different occupational sectors who wore SF during their work shifts were prospectively assessed through a gait analysis. The gait analysis was conducted under two conditions: first, while wearing SF, and second, while wearing sneakers. In both conditions, inertial sensors were used (Wiva® MOB). Participants also underwent a podiatric physical examination to evaluate foot problems. Results: This study shows that SF resulted in a worsening gait pattern compared to sneakers in both genders. The impact was particularly pronounced in female participants, resulting in a significant decline in walking speed and cadence. Discomfort was reported by 83.3% of participants, with a higher prevalence in females (46.6% vs. 36.6%). The SF group exhibited an elevated prevalence of foot problems, with no significant gender variations. It seems that foot problems are more likely to occur when a foot deformity, such as flat or cavus foot or hallux valgus, is present. Conclusions: This study suggests that SF may contribute to the reported podiatric symptoms among workers. Certain footwear characteristics, including weight, mis-fit, and inadequate design, may be factors associated with footwear discomfort and adverse gait patterns, potentially leading to increased foot problems among workers. [ABSTRACT FROM AUTHOR]

Hoof care providers are pivotal for implementing biomechanical optimizations of the musculoskeletal system in the horse. Regular visits allow for the collection of longitudinal, quantitative information ("normal ranges"). Changes in movement symmetry, e.g., after shoeing, are indicative of alterations in weight-bearing and push-off force production. Ten Warmblood show jumping horses (7–13 years; 7 geldings, 3 mares) underwent forelimb re-shoeing with rolled rocker shoes, one limb at a time ("limb-by-limb"). Movement symmetry was measured with inertial sensors attached to the head, withers, and pelvis during straight-line trot and lunging. Normalized differences pre/post re-shoeing were compared to published test–retest repeatability values. Mixed-model analysis with random factors horse and limb within horse and fixed factors surface and exercise direction evaluated movement symmetry changes (p < 0.05, Bonferroni correction). Withers movement indicated increased forelimb push-off with the re-shod limb on the inside of the circle and reduced weight-bearing with the re-shod limb and the ipsilateral hind limb on hard ground compared to soft ground. Movement symmetry measurements indicate that a rolled rocker shoe allows for increased push-off on soft ground in trot in a circle. Similar studies should study different types of shoes for improved practically relevant knowledge about shoeing mechanics, working towards evidence-based preventative shoeing. [ABSTRACT FROM AUTHOR]

Upper limb rehabilitation robots have important practical significance in helping patients recover their motor function, but traditional training methods often lack real-time and accurate evaluation of patient movement status and intention and have low interactive participation. Therefore, the study proposes a training system for upper limb rehabilitation robots based on multi-sensor trajectory data and human-computer interaction. The system is designed from three aspects: inertial sensor trajectory tracking, Kinect sensor image restoration, and interaction system design. A polynomial joint zero value constraint algorithm is introduced to correct errors, and variable parameter pixel filtering combined with a weighted average moving algorithm is used to improve image quality. The training results showed that the robot trajectory tracking accuracy significantly improved, and the improvement in motion trajectory error was greater than 20%. Compared with traditional training methods, interactive training had a recognition accuracy of over 85% in rehabilitation actions, and the system had better stability. This rehabilitation robot can effectively meet the feasibility and effectiveness of improving the interaction system, providing new technological means for intelligent medical services. [ABSTRACT FROM AUTHOR]

Background: In this study, we hypothesized that safety footwear (SF) impacts gait patterns, potentially contributing to the podiatric symptoms reported by workers. The purpose of this work was to compare the gait analyses of workers wearing SF and sneakers using inertial sensors while also examining the occurrence of foot problems. Methods: A consecutive cohort of workers from different occupational sectors who wore SF during their work shifts were prospectively assessed through a gait analysis. The gait analysis was conducted under two conditions: first, while wearing SF, and second, while wearing sneakers. In both conditions, inertial sensors were used (Wiva® MOB). Participants also underwent a podiatric physical examination to evaluate foot problems. Results: This study shows that SF resulted in a worsening gait pattern compared to sneakers in both genders. The impact was particularly pronounced in female participants, resulting in a significant decline in walking speed and cadence. Discomfort was reported by 83.3% of participants, with a higher prevalence in females (46.6% vs. 36.6%). The SF group exhibited an elevated prevalence of foot problems, with no significant gender variations. It seems that foot problems are more likely to occur when a foot deformity, such as flat or cavus foot or hallux valgus, is present. Conclusions: This study suggests that SF may contribute to the reported podiatric symptoms among workers. Certain footwear characteristics, including weight, mis-fit, and inadequate design, may be factors associated with footwear discomfort and adverse gait patterns, potentially leading to increased foot problems among workers.

Although estimating the azimuth using a geomagnetic sensor is very useful, the estimation error may be very large due to the surrounding geomagnetic disturbance. We proposed a novel method for preprocessing appropriately for geomagnetic and inertial sensor data to be suitable for the proposed Artificial Neural Network model and training method for the model. As a result, the probability of azimuth estimation error within 1 degree is 96.4% with regression estimation. For classification estimation, when the azimuth estimation probability is 90% or more, the probability that the azimuth estimation error is within 1 degree is 100%.

Manual wheelchair users (MWUs) are prone to a sedentary life that can negatively affect their physical and cardiovascular health, making regular assessment important to identify appropriate interventions and lifestyle modifications. One mean of assessing MWUs' physical health is the 6 min push test (6MPT), where the user propels themselves as far as they can in six minutes. However, reliance on observer input introduces subjectivity, while limited quantitative data inhibit comprehensive assessment. Incorporating sensors into the 6MPT can address these limitations. Here, ten MWUs performed the 6MPT with additional sensors: two inertial measurement units (IMUs)—one on the wheelchair and one on the wrist together with a heart rate wristwatch. The conventional measurements of distance and laps were recorded by the observer, and the IMU data were used to calculate laps, distance, speed, and cadence. The results demonstrated that the IMU can provide the metrics of the traditional 6MPT with strong significant correlations between calculated laps and observer lap counts (r = 0.947, p < 0.001) and distances (r = 0.970, p < 0.001). Moreover, heart rate during the final minute was significantly correlated with calculated distance (r = 0.762, p = 0.017). Enhanced 6MPT assessment can provide objective, quantitative, and comprehensive data for clinicians to effectively inform interventions in rehabilitation. [ABSTRACT FROM AUTHOR]

Objectives Real-time measurement of running speed and stride length is of great significance in avoiding injury and improving exercise efficiency.Methods We propose a method using inertial measurement unit (IMU) to detect these two indicators. First, 3 IMU are placed on the foot, calf and thigh of the 10 runners which we recruited, and 5 137 data of gait cycles are collected. Second, principal component analysis is used to analyze the data, and Pearson correlation coefficient is used to discuss the relationship between the detection indicators of running and the sensor position and physical parameters. Then a multi-modal attention-long short-term memory (M-Att-LSTM ) is proposed for feature extraction, two long short-term memory (LSTM) modules with attention mechanism are used to extract features of acceleration and angle, and regression fitting is carried out.Results The experiment result shows that M-Att-LSTM has errors of 0.058 m/s in speed and 0.023 m in stride, the standard deviation is 0.013 m/s and 0.022 m, respectively．Both indicators are better than pure Att-LSTM.Conclusions The studies show that multi-modal network can improve network processing capabilities, compared with relevant researches in recent years, our study has obvious advantages in error control． [ABSTRACT FROM AUTHOR]

Recently, VR-based training applications have become popular and promising, as they can simulate real-world situations in a safe, repeatable, and cost-effective way. For immersive simulations, various input devices have been designed and proposed to increase the effectiveness of training. In this study, we developed a novel device that generates 3D hand motion data and provides haptic force feedback for VR interactions. The proposed device can track 3D hand positions using a combination of the global position estimation of ultrasonic sensors and the hand pose estimation of inertial sensors in real time. For haptic feedback, shape–memory alloy (SMA) actuators were designed to provide kinesthetic forces and an efficient power control without an overheat problem. Our device improves upon the shortcomings of existing commercial devices in tracking and haptic capabilities such that it can track global 3D positions and estimate hand poses in a VR space without using an external suit or tracker. For better flexibility in handling and feeling physical objects compared to exoskeleton-based devices, we introduced an SMA-based actuator to control haptic forces. Overall, our device was designed and implemented as a lighter and less bulky glove which provides comparable accuracy and performance in generating 3D hand motion data for a VR training application (i.e., the use of a fire extinguisher), as demonstrated in the experimental results. [ABSTRACT FROM AUTHOR]

Deep learning has proven to be highly effective for human activity recognition (HAR) when large amount of labelled data is available for the target task. However, training a deep learning model to generalize well on a new task with just-few observations of labelled data is an active area of research. In this paper, a novel few-shot transfer learning (FSTL) approach is proposed for classification of human activities using just few instances (shots) of the data obtained from a wearable system assembled to collect inertial sensor data for different human activities, performed by two users. First, a deep learning model is trained on a large publicly available HAR dataset. The model parameters of such a model are then fine-tuned using the Reptile algorithm to determine the optimal initial parameter set using which, the model will classify activities with just few-shots of data from the target task. The proposed FSTL approach yields an average classification accuracy of 74.86 ± 0.71% and 79.20 ± 1.05% for 3-way, 5-shot classification of new activities performed by a single user and same set of activities performed by a new user, respectively. When the pre-trained weights are used as the initial weights in the Reptile algorithm, the generalization ability of the model improves by about 10% for 3-way, 5-shot classification as compared to using few-shot learning without parameter transfer. [ABSTRACT FROM AUTHOR]

Caregivers experience lower back pain due to patient transfer. Foot position is an important and adjustable posture for reducing lumbar loads during patient transfer. Specifically, a suitable foot position provides the use of the lower limbs instead of the lumbar region in patient handling. Thus, we have developed a monitoring and feedback system for foot positioning using wearable sensors to instruct suitable foot positions. However, existing measurement methods require multiple specific wearable sensors. In addition, the existing method has not been evaluated in patient transfer, including twisting and lowering. Thus, the objective of this study was to develop and evaluate a measurement method using only a smartphone-installed inertial sensor for foot position during patient transfer, including twisting and lowering. The smartphone attached to the trunk measures the acceleration, angular velocity, and geomagnetic field. The proposed method recognizes anteroposterior and mediolateral foot positions by machine learning using inertial data. The proposed method was tested using simulated patient transfer motions, including horizontal rotation. The results showed that the proposed method could recognize the two foot positions with more than 90% accuracy. These results indicate that the proposed method can be applied to wearable monitoring and feedback systems to prevent lower back pain caused by patient transfer. [ABSTRACT FROM AUTHOR]

This study aimed to evaluate walking independence in acute-care hospital patients using neural networks based on acceleration and angular velocity from two walking tests. Forty patients underwent the 10-m walk test and the Timed Up-and-Go test at normal speed, with or without a cane. Physiotherapists divided the patients into two groups: 24 patients who were monitored or independent while walking with a cane or without aids in the ward, and 16 patients who were not. To classify these groups, the Transformer model analyzes the left gait cycle data from eight inertial sensors. The accuracy using all the sensor data was 0.836. When sensor data from the right ankle, right wrist, and left wrist were excluded, the accuracy decreased the most. When analyzing the data from these three sensors alone, the accuracy was 0.795. Further reducing the number of sensors to only the right ankle and wrist resulted in an accuracy of 0.736. This study demonstrates the potential of a neural network-based analysis of inertial sensor data for clinically assessing a patient's level of walking independence. [ABSTRACT FROM AUTHOR]

Human activity recognition (HAR) technology enables continuous behavior monitoring, which is particularly valuable in healthcare. This study investigates the viability of using an ear-worn motion sensor for classifying daily activities, including lying, sitting/standing, walking, ascending stairs, descending stairs, and running. Fifty healthy participants (between 20 and 47 years old) engaged in these activities while under monitoring. Various machine learning algorithms, ranging from interpretable shallow models to state-of-the-art deep learning approaches designed for HAR (i.e., DeepConvLSTM and ConvTransformer), were employed for classification. The results demonstrate the ear sensor's efficacy, with deep learning models achieving a 98% accuracy rate of classification. The obtained classification models are agnostic regarding which ear the sensor is worn and robust against moderate variations in sensor orientation (e.g., due to differences in auricle anatomy), meaning no initial calibration of the sensor orientation is required. The study underscores the ear's efficacy as a suitable site for monitoring human daily activity and suggests its potential for combining HAR with in-ear vital sign monitoring. This approach offers a practical method for comprehensive health monitoring by integrating sensors in a single anatomical location. This integration facilitates individualized health assessments, with potential applications in tele-monitoring, personalized health insights, and optimizing athletic training regimes. [ABSTRACT FROM AUTHOR]

Introduction: While meaningless gross motor imitation (GMI) is a common challenge for children diagnosed with autism spectrum disorder (ASD), this topic has not attracted much attention and few appropriate test paradigms have been developed. Methods: The current study proposed a wrist rotation imitation (WRI) task (a meaningless GMI assignment), and established a WRI ability evaluation system using low-cost wearable inertial sensors, which acquired the simultaneous data of acceleration and angular acceleration during the WRI task. Three metrics (i.e., total rotation time, rotation amplitude, and symmetry) were extracted from those data of acceleration and angular acceleration, and then were adopted to construct classifiers based on five machine learning (ML) algorithms, including k-nearest neighbors, linear discriminant analysis, naive Bayes, support vector machines, and random forests. To illustrate our technique, this study recruited 49 ASD children (aged 3.5-6.5 years) and 59 age-matched typically developing (TD) children. Results: Findings showed that compared with TD children, those with ASD may exhibit shorter total rotation time, lower rotation amplitude, and weaker symmetry. This implies that children with ASD might exhibit decreased WRI abilities. The classifier with the naive Bayes algorithm outperformed than other four algorithms, and achieved a maximal classification accuracy of 88% and a maximal AUC value of 0.91. Two metrics (i.e., rotation amplitude and symmetry) had high correlations with the gross and fine motor skills [evaluated by Gesell Developmental Schedules-Third Edition and Psychoeducational Profile-3 (PEP-3)]. While, the three metrics had no significant correlation with the visual-motor imitation abilities (evaluated by the subdomain of PEP-3) and the ASD symptom severity [evaluated by the Childhood Autism Rating Scale (CARS)] . Discussion: The strengths of this study are associated with the low-cost measurement system, correlation between the WRI metrics and clinical measures, decreased WRI abilities in ASD, and high classification accuracy. [ABSTRACT FROM AUTHOR]

A top-down design methodology and implementation of a time domain sensor is presented in this paper. The acceleration resolution of the time domain sensor is equal to the time-measurement accuracy divided by the sensor sensitivity. Combined with the sensitivity formula, the acceleration resolution is proportional to the vibration amplitude, the time-measurement accuracy, and the third power of the resonant frequency. According to the available time-measurement accuracy and the desired acceleration resolution, the parameters including the vibration amplitude and the resonant frequency were theoretically calculated. The geometrical configuration of the time domain sensor device was designed based on the calculated parameters. Then, the designed device was fabricated based on a standard silicon-on-insulator process and a matched interface circuit was developed for the fabricated device. Experimental results demonstrated that the design methodology is effective and feasible. Moreover, the implemented sensor works well. In addition, the acceleration resolution can be tuned by adjusting the time-measurement accuracy and the vibration amplitude. All the reported results of this work can be expanded to other time domain inertial sensors, e.g., a gyroscope or tilt sensor. [ABSTRACT FROM AUTHOR]

Sensors enable vehicles to perceive their surroundings even while parked. Camera‐based systems for vehicle surroundings perception are already available as of today. However, camera‐based systems result in excessive power consumption and data collection. In this article, a study on always‐on surroundings perception of parked vehicles is done using micro‐electromechanical systems (MEMS) inertial sensors as an alternative to camera‐based systems. The measurements of the MEMS inertial sensors are used to detect and classify 17 different types of events. These events include accidents, vandalism, rain, as well as normal use. A convolutional neural network (CNN) is used to classify the events. The data used to train the CNN is recorded using a test vehicle in a laboratory setting and on a test track. The CNN is quantized for deployment on a CNN hardware accelerator which is packaged with the sensor as a system‐in‐package (SiP). The novel sensor system achieves a classification accuracy of 88% at a total system power dissipation of 27 μW. [ABSTRACT FROM AUTHOR]

Caregivers experience lower back pain due to patient transfer. Foot position is an important and adjustable posture for reducing lumbar loads during patient transfer. Specifically, a suitable foot position provides the use of the lower limbs instead of the lumbar region in patient handling. Thus, we have developed a monitoring and feedback system for foot positioning using wearable sensors to instruct suitable foot positions. However, existing measurement methods require multiple specific wearable sensors. In addition, the existing method has not been evaluated in patient transfer, including twisting and lowering. Thus, the objective of this study was to develop and evaluate a measurement method using only a smartphone-installed inertial sensor for foot position during patient transfer, including twisting and lowering. The smartphone attached to the trunk measures the acceleration, angular velocity, and geomagnetic field. The proposed method recognizes anteroposterior and mediolateral foot positions by machine learning using inertial data. The proposed method was tested using simulated patient transfer motions, including horizontal rotation. The results showed that the proposed method could recognize the two foot positions with more than 90% accuracy. These results indicate that the proposed method can be applied to wearable monitoring and feedback systems to prevent lower back pain caused by patient transfer.

Abstract Background Incorporating instrument measurements into clinical assessments can improve the accuracy of results when assessing mobility related to activities of daily living. This can assist clinicians in making evidence-based decisions. In this context, kinematic measures are considered essential for the assessment of sensorimotor recovery after stroke. The aim of this study was to assess the validity of using an Android device to evaluate kinematic data during the performance of a standardized mobility test in people with chronic stroke and hemiparesis. Methods This is a cross-sectional study including 36 individuals with chronic stroke and hemiparesis and 33 age-matched healthy subjects. A simple smartphone attached to the lumbar spine with an elastic band was used to measure participants’ kinematics during a standardized mobility test by using the inertial sensor embedded in it. This test includes postural control, walking, turning and sitting down, and standing up. Differences between stroke and non-stroke participants in the kinematic parameters obtained after data sensor processing were studied, as well as in the total execution and reaction times. Also, the relationship between the kinematic parameters and the community ambulation ability, degree of disability and functional mobility of individuals with stroke was studied. Results Compared to controls, participants with chronic stroke showed a larger medial-lateral displacement (p = 0.022) in bipedal stance, a higher medial-lateral range (p

With the advancement of semiconductor technology, pedestrian navigation and positioning technology based on smartphones is becoming increasingly important in people's travel. However, precise positioning is challenging due to the use of inertial measurement units in low-cost smartphones and the complex motion states of pedestrians. To navigate and locate pedestrians in complex motion states, a method for converting between smartphone coordinate systems and navigation coordinate systems was studied and designed, and the errors of the built-in sensors of smartphones were analyzed and calibrated. In addition, support vector machines were used to optimize pedestrian trajectory prediction algorithms, and a pedestrian motion state recognition algorithm was designed based on this. To solve the classification problem of multiple human motion states, a multi classification model was constructed and adjacent gait correlation constraints were introduced to correct the classification results. Experiments indicated that the sum of squared errors for traditional algorithms estimating pedestrian trajectories was 0.92, whereas the optimized algorithms produced an improved sum of squared errors of 0.26. Consequently, the average sum of squared errors was reduced by 71.74 %, and the convergence speed increased by 55.56 %. The pedestrian trajectory prediction algorithm optimized by support vector machine could significantly lift the positioning and navigation efficiency, with a correct recognition rate of over 93 % and a position recognition accuracy of 78.8 % - 88.4 %. By optimizing recognition of the motion state of pedestrians, more accurate determination of their position and motion state can be achieved.

Background: Incorporating instrument measurements into clinical assessments can improve the accuracy of results when assessing mobility related to activities of daily living. This can assist clinicians in making evidence-based decisions. In this context, kinematic measures are considered essential for the assessment of sensorimotor recovery after stroke. The aim of this study was to assess the validity of using an Android device to evaluate kinematic data during the performance of a standardized mobility test in people with chronic stroke and hemiparesis. Methods: This is a cross-sectional study including 36 individuals with chronic stroke and hemiparesis and 33 age-matched healthy subjects. A simple smartphone attached to the lumbar spine with an elastic band was used to measure participants' kinematics during a standardized mobility test by using the inertial sensor embedded in it. This test includes postural control, walking, turning and sitting down, and standing up. Differences between stroke and non-stroke participants in the kinematic parameters obtained after data sensor processing were studied, as well as in the total execution and reaction times. Also, the relationship between the kinematic parameters and the community ambulation ability, degree of disability and functional mobility of individuals with stroke was studied. Results: Compared to controls, participants with chronic stroke showed a larger medial-lateral displacement (p = 0.022) in bipedal stance, a higher medial-lateral range (p < 0.001) and a lower cranio-caudal range (p = 0.024) when walking, and lower turn-to-sit power (p = 0.001), turn-to-sit jerk (p = 0.026) and sit-to-stand jerk (p = 0.001) when assessing turn-to-sit-to-stand. Medial-lateral range and total execution time significantly correlated with all the clinical tests (p < 0.005), and resulted significantly different between independent and limited community ambulation patients (p = 0.042 and p = 0.006, respectively) as well as stroke participants with significant disability or slight/moderate disability (p = 0.024 and p = 0.041, respectively). Conclusion: This study reports a valid, single, quick and easy-to-use test for assessing kinematic parameters in chronic stroke survivors by using a standardized mobility test with a smartphone. This measurement could provide valid clinical information on reaction time and kinematic parameters of postural control and gait, which can help in planning better intervention approaches. [ABSTRACT FROM AUTHOR]

Low back pain is a painful disorder that prevents normal mobilization, increases muscle tension and whose first-line treatment is usually non-steroidal anti-inflammatory drugs, together with non-invasive manual therapies, such as deep oscillation therapy. This systematic review aims to investigate and examine the scientific evidence of the effectiveness of deep oscillation therapy in reducing pain and clinical symptomatology in patients with low back pain, through the use of motion capture technology. To carry out this systematic review, the guidelines of the PRISMA guide were followed. A literature search was performed from 2013 to March 2022 in the PubMed, Elsevier, Science Director, Cochrane Library, and Springer Link databases to collect information on low back pain, deep oscillation, and motion capture. The risk of bias of the articles was assessed using the Cochrane risk of bias tool. Finally, they were included 16 articles and 5 clinical trials which met the eligibility criteria. These articles discussed the effectiveness of deep oscillation therapy in reducing pain, eliminating inflammation, and increasing lumbar range of motion, as well as analyzing the use of motion capture systems in the analysis, diagnosis, and evaluation of a patient with low back pain before, during and after medical treatment. There is no strong scientific evidence that demonstrates the high effectiveness of deep oscillation therapy in patients with low back pain, using motion capture systems. This review outlines the background for future research directed at the use of deep oscillation therapy as a treatment for other types of musculoskeletal injuries. [ABSTRACT FROM AUTHOR]

An educational augmented reality auscultation system (EARS) is proposed to enhance the reality of auscultation training using a simulated patient. The conventional EARS cannot accurately reproduce breath sounds according to the breathing of a simulated patient because the system instructs the breathing rhythm. In this study, we propose breath measurement methods that can be integrated into the chest piece of a stethoscope. We investigate methods using the thoracic variations and frequency characteristics of breath sounds. An accelerometer, a magnetic sensor, a gyro sensor, a pressure sensor, and a microphone were selected as the sensors. For measurement with the magnetic sensor, we proposed a method by detecting the breathing waveform in terms of changes in the magnetic field accompanying the surface deformation of the stethoscope based on thoracic variations using a magnet. During breath sound measurement, the frequency spectra of the breath sounds acquired by the built-in microphone were calculated. The breathing waveforms were obtained from the difference in characteristics between the breath sounds during exhalation and inhalation. The result showed the average value of the correlation coefficient with the reference value reached 0.45, indicating the effectiveness of this method as a breath measurement method. And the evaluations suggest more accurate breathing waveforms can be obtained by selecting the measurement method according to breathing method and measurement point. [ABSTRACT FROM AUTHOR]

Considering that laser scan stations are expensive and heavy for the indoor environment reconstruction of unmanned systems, a low-cost and light reconstruction system is proposed in this research. The system consists of a cross-structured light visual (SLV) sensor and an inertial navigation system (INS). The cross-SLV sensor is used to scan the surroundings and to estimate the angle change between two adjacent measurements. To improve the robustness and accuracy of the angle measurement, a Kalman Filter (KF) with simple construction is designed to fuse the inertial information from the INS. The factors which influence ranging accuracy are analyzed and ranging experiments show that the SLV sensor has an accuracy of higher than 2% when the distance is less than 4 m. Then the reconstruction results of a kitchen and corridor show that the error of most points is less than 50 mm for either kitchen (94%) or corridor (85%), and the mean errors and standard deviations of kitchen and corridor are less than 20 mm and 30 mm, respectively. [ABSTRACT FROM AUTHOR]

Estimation of vivo muscle forces during human motion is important for understanding human motion control mechanisms and joint mechanics. This paper combined the advantages of the convolutional neural network (CNN) and long-short-term memory (LSTM) and proposed a novel muscle force estimation method based on CNN–LSTM. A wearable sensor system was also developed to collect the angles and angular velocities of the hip, knee, and ankle joints in the sagittal plane during walking, and the collected kinematic data were used as the input for the neural network model. In this paper, the muscle forces calculated using OpenSim based on the Static Optimization (SO) method were used as the standard value to train the neural network model. Four lower limb muscles of the left leg, including gluteus maximus (GM), rectus femoris (RF), gastrocnemius (GAST), and soleus (SOL), were selected as the studying objects in this paper. The experiment results showed that compared to the standard CNN and the standard LSTM, the CNN–LSTM performed better in muscle forces estimation under slow (1.2 m/s), medium (1.5 m/s), and fast walking speeds (1.8 m/s). The average correlation coefficients between true and estimated values of four muscle forces under slow, medium, and fast walking speeds were 0.9801, 0.9829, and 0.9809, respectively. The average correlation coefficients had smaller fluctuations under different walking speeds, which indicated that the model had good robustness. The external testing experiment showed that the CNN–LSTM also had good generalization. The model performed well when the estimated object was not included in the training sample. This article proposed a convenient method for estimating muscle forces, which could provide theoretical assistance for the quantitative analysis of human motion and muscle injury. The method has established the relationship between joint kinematic signals and muscle forces during walking based on a neural network model; compared to the SO method to calculate muscle forces in OpenSim, it is more convenient and efficient in clinical analysis or engineering applications. [ABSTRACT FROM AUTHOR]

Inertial sensors are the key payloads in space gravitational wave detection missions, and they need to ensure that the test mass (TM), which serves as the inertial reference, freely floats in the spacecraft without contact, so that the TM is not disturbed by the satellite platform and the cosmic environment. Space gravitational wave detection missions require that the residual acceleration of the TM should be less than 3 × 10 − 15 ms − 2 Hz − 1 / 2 . However, the TM with charges will interact with surrounding conductors and magnetic fields, introducing acceleration noise such as electrostatic force and Lorentz force. Therefore, it is necessary to carry out charge management on the TM, in which the high-precision measurement of charge is crucial. Space gravitational wave detection missions require a residual charge measurement accuracy of 3 × 10 − 13 C for the TM. In this paper, we design a high-precision inertial sensor charge measurement method based on phase-sensitive demodulation (PSD). By establishing a torsion pendulum rotation model based on the force modulation method, the characteristics of the TM torsion angle signal are analyzed. The PSD is used to extract the amplitude of the specific frequency signal component containing the charge information, and then to calculate the value of the accumulated charges. The method is compared with the Butterworth band-pass filtering method, and the simulation results show that the method has a higher measurement accuracy, shorter settling time, and stronger anti-interference ability, meeting the TM residual charge measurement accuracy index requirement. [ABSTRACT FROM AUTHOR]

Gait analysis using foot-mounted IMUs is a promising method to acquire gait parameters outside of laboratory settings and in everyday clinical practice. However, the need for precise sensor attachment or calibration, the requirement of environments with a homogeneous magnetic field, and the limited applicability to pathological gait patterns still pose challenges. Furthermore, in previously published work, the measurement accuracy of such systems is often only validated for specific points in time or in a single plane. This study investigates the measurement accuracy of a gait analysis method based on foot-mounted IMUs in the acquisition of the foot motion, i.e., position and angle trajectories of the foot in the sagittal, frontal, and transversal plane over the entire gait cycle. A comparison of the proposed method with an optical motion capture system showed an average RMSE of 0.67° for pitch, 0.63° for roll and 1.17° for yaw. For position trajectories, an average RMSE of 0.51 cm for vertical lift and 0.34 cm for lateral shift was found. The measurement error of the IMU-based method is found to be much smaller than the deviations caused by the shoes. The proposed method is found to be sufficiently accurate for clinical practice. It does not require precise mounting, special calibration movements, or magnetometer data, and shows no difference in measurement accuracy between normal and pathological gait. Therefore, it provides an easy-to-use alternative to optical motion capture and facilitates gait analysis independent of laboratory settings. • Easy-to-use gait analysis using one foot-mounted IMU for each foot. • Does not require precise IMU attachment, calibration movements, or magnetometer data. • Position and angle trajectories in all planes are determined with high accuracy. • Same measurement accuracy for normal walking and pathological gait patterns. • Enables non-restrictive, detailed gait analysis in everyday clinical practice. [ABSTRACT FROM AUTHOR]

Progressive balance exercises are critical to early functional rehabilitation after total hip arthroplasty (THA) but little is known regarding the challenge imposed by common balance devices. Do progressively unstable balance devices provide a graded challenge to bipedal stance during early functional rehabilitation in THA patients? Postural control was evaluated in 42 patients (age, 63.7 ± 9.6 years; height, 1.72 ± 0.08 m and body mass, 78.9 ± 14.6 kg) approximately 3 weeks (23 ± 6 days) following unilateral primary THA. Patients were divided into two groups, based on their ability to complete a 20-second unipedal stance test (UPST) on the operated limb. A lumbar mounted inertial sensor monitored center of mass (COM) displacement during bipedal balance conditions involving three balance pads of progressive stiffness and an oscillatory platform, used in isolation and in combination with the most stable balance pad. COM displacement was normalised to bipedal stance on a hard surface. Differences between conditions and patient groups were assessed using a mixed-model analysis of variance. Twenty patients (48%) were able to complete the UPST on their operated limb. There was a significant effect of balance condition on COM displacement during bipedal stance (F 4 , 160 = 82.6, p <.01). COM displacement was lowest for the oscillatory platform but increased non-linearly across the three balance pads (p <.05). There was no significant difference in COM displacement between THA patients able and unable to complete the UPST. Increasingly compliant balance pads provided a progressive, though nonlinear, challenge to bipedal balance control in THA patients that was greater than that of an oscillating platform and independent of the ability to stand independently on the operated limb. These findings serve as a guide for the design of progressive training programs that enhance balance in THA patients. • Balance pads provided a progressive nonlinear challenge to postural control in THA. • Change in surface compliance invoked greater challenge than the oscillating platform. • Balance performance was unrelated to the ability to stand only on the operated limb. [ABSTRACT FROM AUTHOR]

This paper characterizes the sensitivity of a time domain MEMS accelerometer. The sensitivity is defined by the increment in the measured time interval per gravitational acceleration. Two sensitivities exist, and they can be enhanced by decreasing the amplitude and frequency. The sensitivity with minor nonlinearity is chosen to evaluate the time domain sensor. The experimental results of the developed accelerometer demonstrate that the sensitivities span from −68.91 μs/g to −124.96 μs/g and the 1σ noises span from 8.59 mg to 6.2 mg (amplitude of 626 nm: −68.91 μs/g and 10.21 mg; amplitude of 455 nm: −94.51 μs/g and 7.76 mg; amplitude of 342 nm: −124.96 μs/g and 6.23 mg), which indicates the bigger the amplitude, the smaller the sensitivity and the bigger the 1σ noise. The adjustable sensitivity provides a theoretical foundation for range self-adaption, and all the results can be extended to other time domain inertial sensors, e.g., a gyroscope or an inclinometer. [ABSTRACT FROM AUTHOR]

Sensors enable vehicles to perceive their surroundings even while parked. Camera‐based systems for vehicle surroundings perception are already available as of today. However, camera‐based systems result in excessive power consumption and data collection. In this article, a study on always‐on surroundings perception of parked vehicles is done using micro‐electromechanical systems (MEMS) inertial sensors as an alternative to camera‐based systems. The measurements of the MEMS inertial sensors are used to detect and classify 17 different types of events. These events include accidents, vandalism, rain, as well as normal use. A convolutional neural network (CNN) is used to classify the events. The data used to train the CNN is recorded using a test vehicle in a laboratory setting and on a test track. The CNN is quantized for deployment on a CNN hardware accelerator which is packaged with the sensor as a system‐in‐package (SiP). The novel sensor system achieves a classification accuracy of 88% at a total system power dissipation of 27 μW.

Farmworkers are often at risk of musculoskeletal health problems, with low back pain being the most common, accounting for approximately over half of the population worldwide. Moreover, these musculoskeletal disorders (MSD) are prevalent in other body regions including shoulders, elbows, wrists or hands. As these issues were found related to inappropriate working postures, it is beneficial to quantify these postures in order to prevent work-related MSD. The primary focus of this study was to improve the estimation of the postures and the exposure to non-neutral postures among agricultural workers. Three inertial measurement units (IMU) were attached to the upper back, upper arm, and forearm of nine workers while they were performing their regular work activities. A posture characterization algorithm was developed to rely on the data from only accelerometers and gyroscopes while excluding magnetometer readings due to high magnetic disturbance. Despite these challenges, a specialized unscented Kalman filter (UKF) was developed to achieve a more precise posture estimation. The UKF effectively expanded the range of pitch angles from ±90 degrees to ±180 degrees, resulting in a substantial improvement in the assessment of the back inclination as well as shoulder and elbow angles. The study was carried out among workers in a large-scale plant nursery, revealing instances of extreme postures in the back, upper arms, and elbows during their work activities. The quantitative findings highlighted the high exposure to ergonomic risks faced by the workers. This emphasized the urgent need for implementing appropriate measures to mitigate these risks.

Existing visible light positioning (VLP) algorithms typically require multiple LEDs to achieve high-precision positioning. However, in practice, the number of available LEDs is influenced by various factors such as the layout of LEDs and the field-of-view of the receiver, and these factors can significantly affect the coverage of a VLP algorithm. To address this issue, this paper proposes an inertial sensor-assisted single view geometry (ISA-SVG) algorithm, which can achieve accurate positioning as little as one LED. In particular, we first establish the projection transformation model between the actual point and its projection on the image plane. Then, a pose fusion model that effectively fuses the inertial sensor information and VLP information is established. Finally, the position of the receiver based on the pose fusion model is estimated. IAS-SVG significantly improves the usability and coverage performance of visible light positioning algorithms. Simulation results show that the coverage rate of the ISA-SVG algorithm is over 95%. Experimental results show that the proposed method can achieve an average positioning error as low as 4.70 cm.

Inertial sensor has become a hot spot in the research and application of shearer attitude measurement and underground autonomous navigation due to its distinct advantages. The inertial sensor is disturbed by the strong vibration of the shearer in practical application, and has the problems of poor measurement stability and low accuracy. A metal spring-adjustable friction damping three-way vibration damping device is developed to solve the above problems. Firstly, the types of inertial sensors and the attitude measurement method of shearers based on inertial sensors are introduced. The data of vibration acceleration of MG500/1180WD drum shearer when cutting coal seam are analyzed, the main vibration frequency in X, Y and Z directions is 0-30 Hz, and the three-direction vibration acceleration ranges are 1-2 m/s2, 0.15-0.2 m/s2 and 2-4 m/s2, respectively. Secondly, based on the vibration characteristics of the shearer, the mechanical model of the metal spring-adjustable friction damping one-way vibration damper is proposed, and the model transmissibility equation is established. Combined with the linear guide mechanism, a metal spring-adjustable friction damping three-way vibration damping device is designed. SolidWorks Motion is used to establish the simulation model of the vibration reduction device, and the main parameters of the vibration reduction device are determined by the simulation analysis. Finally, the vibration transmittance simulation analysis and experimental test of the vibration isolation device are carried out, and the effect of vibration reduction is verified by using the inertial sensor. The simulation analysis shows that under the vibration conditions that conform to the vibration frequency range and vibration acceleration range of the shearer, the adjustable friction damping of the vibration damper is 0.25-0.5 times of the maximum reference value of the friction damping, and the lower resonance amplification times and the lower vibration transmission rate can be obtained. The experimental results show that the resonance amplification of the vibration damping device in three directions is less than 4 times. Under the excitation of 10-30Hz frequency range, the average vibration transmission rate remains below 45%. Adding vibration damping device can effectively reduce the fluctuation of the measured data of the inertial measuring unit. The average drop rate of attitude angle standard deviation of the three-direction sensors is 78.36%, 36.96% and 66.51% respectively, which has obvious vibration damping effect.

Despite recent advancements in prosthetic technology, lower-limb amputees often remain limited to passive prostheses, which leads to an asymmetric gait and increased energy expenditure. Developing active prostheses with effective control systems is important to improve mobility for these individuals. This study presents a machine-learning-based approach to classify five distinct locomotion tasks: ground-level walking (GWL), ramp ascent (RPA), ramp descent (RPD), stairs ascent (SSA), and stairs descent (SSD). The dataset comprises fused electromyographic (EMG) and inertial measurement unit (IMU) signals from twenty non-amputated and five transtibial amputated participants. EMG sensors were strategically positioned on the thigh muscles, while IMU sensors were placed on various leg segments. The performance of two classification algorithms, support vector machine (SVM) and long short-term memory (LSTM), were evaluated on segmented data. The results indicate that SVM models outperform LSTM models in accuracy, precision, and F1 score in the individual evaluation of amputee and non-amputee datasets for 80–20 and 50–50 data distributions. In the 80–20 distribution, an accuracy of 95.46% and 95.35% was obtained with SVM for non-amputees and amputees, respectively. An accuracy of 93.33% and 93.30% was obtained for non-amputees and amputees by using LSTM, respectively. LSTM models show more robustness and inter-population generalizability than SVM models when applying domain-adaptation techniques. Furthermore, the average classification latency for SVM and LSTM models was 19.84 ms and 37.07 ms, respectively, within acceptable limits for real-time applications. This study contributes to the field by comprehensively comparing SVM and LSTM classifiers for locomotion tasks, laying the foundation for the future development of real-time control systems for active transtibial prostheses.

Mobile health technologies enable continuous, quantitative assessment of mobility and gait in real-world environments, facilitating early diagnoses of gait disorders, disease progression monitoring, and prediction of adverse events like falls. Traditionally, mobile gait assessment predominantly relied on body-fixed sensors positioned at the feet or lower trunk. Here, we investigate the potential of an algorithm utilizing an ear-worn motion sensor for spatiotemporal segmentation of gait patterns. We collected 3D acceleration profiles from the ear-worn sensor during varied walking speeds in 53 healthy adults. Temporal convolutional networks were trained to detect stepping sequences and predict spatial relations between steps. The resulting algorithm, mEar, accurately detects initial and final ground contacts (F1 score of 99% and 91%, respectively). It enables the determination of temporal and spatial gait cycle characteristics (among others, stride time and stride length) with good to excellent validity at a precision sufficient to monitor clinically relevant changes in walking speed, stride-to-stride variability, and side asymmetry. This study highlights the ear as a viable site for monitoring gait and proposes its potential integration with in-ear vital-sign monitoring. Such integration offers a practical approach to comprehensive health monitoring and telemedical applications, by integrating multiple sensors in a single anatomical location.

Within rehabilitation, there is a great need for a simple method to monitor wheelchair use, especially whether it is active or passive. For this purpose, an existing measurement technique was extended with a method for detecting self- or attendant-pushed wheelchair propulsion. The aim of this study was to validate this new detection method by comparison with manual annotation of wheelchair use. Twenty-four amputation and stroke patients completed a semi-structured course of active and passive wheelchair use. Based on a machine learning approach, a method was developed that detected the type of movement. The machine learning method was trained based on the data of a single-wheel sensor as well as a setup using an additional sensor on the frame. The method showed high accuracy (F1 = 0.886, frame and wheel sensor) even if only a single wheel sensor was used (F1 = 0.827). The developed and validated measurement method is ideally suited to easily determine wheelchair use and the corresponding activity level of patients in rehabilitation. [ABSTRACT FROM AUTHOR]

Aims: This paper aims to assess kinematic parameters related to functional capacity, fatigue, and breathlessness during the 6-min walk test (6MWT) in patients with heart failure with preserved ejection fraction (HFpEF). Methods and results: A cross-sectional study was conducted in which adults 70 years or older with HFpEF were voluntarily recruited between April 2019 and March 2020. An inertial sensor was placed at the L3–L4 level and another on the sternum to assess kinematic parameters. The 6MWT was divided into two 3-min phases. Leg fatigue and breathlessness, assessed by the Borg scale, the heart rate (HR), and the oxygen saturation (SpO2), were measured at the beginning and the end of the 6MWT. The difference in kinematic parameters between the 6MWT two 3-min phases was also calculated. Bivariate Pearson correlations and subsequent multivariate linear regression analysis were performed. Seventy older adults with HFpEF (mean = 80.74 years old) were included. Kinematic parameters explained 81.00% of the functional capacity, 45.50% of the leg fatigue and 66.10% of the breathlessness variance. Moreover, kinematic parameters could explain 30.90% of the SpO2 variance at the end of the 6MWT. Kinematic parameters also explained 33.10% of the SpO2 difference between the beginning and end of 6MWT. Kinematic parameters explained neither the HR variance at the end of 6MWT nor the HR difference between the beginning and end. Conclusion: Gait kinematics from L3-L4 and sternum explain a part of the variance in subjective outcomes, assessed by the Borg scale, and objective outcomes such as functional capacity and SpO2. The kinematic assessment allows clinicians to quantify fatigue and breathlessness through objective parameters related to the patient's functional capacity. Registration: ClinicalTrials.gov NCT03909919. Graphical Abstract [ABSTRACT FROM AUTHOR]

With the widespread use of Internet of Things (IoT) technology in daily life and the considerable safety risks of falls for elderly individuals, research on IoT-based fall detection systems has gainedmuch attention. This paper proposes an IoT-based spatiotemporal data processing framework based on a depthwise separable convolution generative adversarial network using skip-connection (Skip-DSCGAN) for fall detection. The method uses spatiotemporal data from accelerometers and gyroscopes in inertial sensors as input data. A semisupervised learning approach is adopted to train the model using only activities of daily living (ADL) data, which can avoid data imbalance problems. Furthermore, a quantile-basedapproachis employedtodetermine the fall threshold,whichmakes the fall detection frameworkmore robust. This proposed fall detection framework is evaluated against four other generative adversarial network (GAN) models with superior anomaly detection performance using two fall public datasets (SisFall & MobiAct). The test results show that the proposed method achieves better results, reaching 96.93% and 92.75% accuracy on the above two test datasets, respectively. At the same time, the proposed method also achieves satisfactory results in terms of model size and inference delay time, making it suitable for deployment on wearable devices with limited resources. In addition, this paper also compares GAN-based semisupervised learning methods with supervised learning methods commonly used in fall detection. It clarifies the advantages of GANbased semisupervised learning methods in fall detection. [ABSTRACT FROM AUTHOR]

Copyright of Rigakuryoho Kagaku is the property of International Press Editing Center Incorporation and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In this study, we introduce a novel framework that combines human motion parameterization from a single inertial sensor, motion synthesis from these parameters, and biped robot motion control using the synthesized motion. This framework applies advanced deep learning methods to data obtained from an IMU attached to a human subject's pelvis. This minimalistic sensor setup simplifies the data collection process, overcoming price and complexity challenges related to multi-sensor systems. We employed a Bi-LSTM encoder to estimate key human motion parameters: walking velocity and gait phase from the IMU sensor. This step is followed by a feedforward motion generator-decoder network that accurately produces lower limb joint angles and displacement corresponding to these parameters. Additionally, our method also introduces a Fourier series-based approach to generate these key motion parameters solely from user commands, specifically walking speed and gait period. Hence, the decoder can receive inputs either from the encoder or directly from the Fourier series parameter generator. The output of the decoder network is then utilized as a reference motion for the walking control of a biped robot, employing a constraint-consistent inverse dynamics control algorithm. This framework facilitates biped robot motion planning based on data from either a single inertial sensor or two user commands. The proposed method was validated through robot simulations in the MuJoco physics engine environment. The motion controller achieved an error of ≤5° in tracking the joint angles demonstrating the effectiveness of the proposed framework. This was accomplished using minimal sensor data or few user commands, marking a promising foundation for robotic control and human–robot interaction. [ABSTRACT FROM AUTHOR]

Microelectromechanical system (MEMS) inertial sensors are integral components in a variety of smart electronic devices, most notably MEMS vibrating gyroscopes, which are rotational inertial sensors. The applications of MEMS vibrating gyroscopes range from household appliances to GPS and even to military applications. However, the stability and reliability of these MEMS inertial sensors in space applications still pose challenges. In this research study, we introduce a simple design for a vibrating ring gyroscope with eight semicircular support springs connected to outside-placed anchors. The symmetric design structure with semicircular support springs provides higher sensitivity while minimizing mode mismatch. The design and modelling analysis of the vibrating ring gyroscope was conducted using Ansys 2023 R1. The proposed vibrating ring gyroscope has a ring radius of 1000 µm, a 210 µm radius for the semicircular support springs, a ring and support spring thicknesses of 10 µm, and an area of 80 × 80 µm2 for the outside-placed anchors. The vibrating ring gyroscope operates at two identical elliptical-shape resonant modes, one for driving resonance frequency and the other for sensing resonance frequency. Both simulated resonance frequencies were measured at 48.78 kHz and 48.80 kHz. The modelled result achieved a mode mismatch of 0.02 kHz, which can be easily rectified with tuning electrodes. [ABSTRACT FROM AUTHOR]

Despite recent advancements in prosthetic technology, lower-limb amputees often remain limited to passive prostheses, which leads to an asymmetric gait and increased energy expenditure. Developing active prostheses with effective control systems is important to improve mobility for these individuals. This study presents a machine-learning-based approach to classify five distinct locomotion tasks: ground-level walking (GWL), ramp ascent (RPA), ramp descent (RPD), stairs ascent (SSA), and stairs descent (SSD). The dataset comprises fused electromyographic (EMG) and inertial measurement unit (IMU) signals from twenty non-amputated and five transtibial amputated participants. EMG sensors were strategically positioned on the thigh muscles, while IMU sensors were placed on various leg segments. The performance of two classification algorithms, support vector machine (SVM) and long short-term memory (LSTM), were evaluated on segmented data. The results indicate that SVM models outperform LSTM models in accuracy, precision, and F1 score in the individual evaluation of amputee and non-amputee datasets for 80–20 and 50–50 data distributions. In the 80–20 distribution, an accuracy of 95.46% and 95.35% was obtained with SVM for non-amputees and amputees, respectively. An accuracy of 93.33% and 93.30% was obtained for non-amputees and amputees by using LSTM, respectively. LSTM models show more robustness and inter-population generalizability than SVM models when applying domain-adaptation techniques. Furthermore, the average classification latency for SVM and LSTM models was 19.84 ms and 37.07 ms, respectively, within acceptable limits for real-time applications. This study contributes to the field by comprehensively comparing SVM and LSTM classifiers for locomotion tasks, laying the foundation for the future development of real-time control systems for active transtibial prostheses. [ABSTRACT FROM AUTHOR]