Human Movement Patterns Research Articles

Gait Analysis through Animated GIFs: A Novel Approach for Visualizing Step Dynamics

The field of prosthetic knee development has made great progress in recent years, notably with the use of machine learning (ML) approaches. Researchers made significant progress in enhancing the functionality, flexibility, and user experience of prosthetic knees using revolutionary machine learning techniques.Since gait analysis delivers part of the most important information about human motion patterns, it becomes critical in the development of a prosthetic knee. In this paper, some new techniques for gait analysis with feature extraction are proposed that include visualization using animated GIF images. This methodology extracts key parameters like step length and step width from the frames of a GIF image and can completely represent the dynamics of gait with respect to time.This methodology is particularly significant to design an efficient prosthetic knee since it shows how combining computer analysis and visual data may enhance gait analysis methods. For this study, brief walking footage of both healthy individuals and amputees are recorded, then turned into GIFs. Subsequently, these GIF data points are retrieved, making it possible to quantify step length and other essential metrics required to analyses gait patterns.Gait analysis is incorporated as part of the development process for these prosthetic knee joints due to the fact that it gives information on human motion patterns. The methods used in this paper provide novel ways of carrying out gait analysis through feature extraction and visualization using animated GIF images. This method would help in depicting the full picture of gait dynamics with time and assist in extracting important parameters such as step length and width from the frames of the GIF. The work, as described, falls squarely into the use of computer vision techniques for feature extraction and into time-series data analysis for trend identification within the broader application of machine learning in the development of the prosthetic knee. These techniques could be generalized toward designing better prosthetic knee design for more natural and fluent motion.

Kirigami-Inspired Stretchable Piezoelectret Sensor for Analysis and Assessment of Parkinson's Tremor.

Human muscle activity contains rich information that can reflect human movement patterns and conditions of diseases or physical abnormalities. Flexible pressure sensors enable the assessment of muscle tremors in Parkinson's disease (PD) through Force Myography (FMG). Here, an easily fabricated, ultra-sensitive, and stretchable piezoelectret pressure sensor is presented. Utilizing an effective integration of Kirigami structure and piezoelectret air gap, the sensor achieved a dynamic sensitivity of ≈725 pC/N (@5Hz), measurement repeatability of <2.5%, measurement hysteresis of <1%, a pressure detection limit of <15Pa, a response time of ≈2.5ms, stable output within ±3% over 40000 cycles, and output decay of <2.5% after 1000 cycles of complex deformation, meeting non-distorted measurement conditions up to 20Hz. Successful monitoring and assessment of hand muscle tremors are demonstrated. Furthermore, using a 1×3 sensor array enabled tremor localization, achieving a high accuracy rate of 99.5% with machine learning algorithms. Additionally, the sensor facilitated the experimental quantification and assisted scoring of the Movement Disorder Society Unified Parkinson's Disease Rating Scale (MDS-UPDRS), with an accuracy of ≈85%. The sensor demonstrates potential for assisting in the diagnosis and rehabilitation monitoring of Parkinson's disease.

Robust and low swelling polyacrylamide/sodium alginate dual-network hydrogels via multiple freezing strategy toward amphibious motion sensors

The swelling and disintegration behavior of conventional hydrogels in liquid environments significantly affects their mechanical and ionic conductive properties, thereby limiting their widespread application in underwater setting. Herein, we propose a multiple freezing (MF) strategy for the construction of polyacrylamide/sodium alginate (PAM/SA) dual-network hydrogels that exhibit excellent anti-swelling properties and high toughness. The MF strategy employs a combination of freeze casting and refreezing to promote a cascading enhancement of polymer network density, resulting in a highly dense polymer network. This dense network structure endows the hydrogel with impressive tensile strength (3.491 MPa), exceptional elongation at break (435.71 %), and high toughness (5.291 MJ m⁻3), along with remarkable ionic conductivity (142.75 mS m⁻1) and sensitivity (GF = 2.278). These properties enable the hydrogel to function as a flexible sensor capable of detecting various human movement patterns. Its excellent resistance to swelling makes it an ideal material for underwater sensors, effectively monitoring different swimming strokes. This strategy provides a straightforward and effective means to enhance the mechanical strength and adaptability of traditional hydrogels in aquatic environments, demonstrating significant potential for applications in underwater sensing.

The effects of seasonal human mobility and Aedes aegypti habitat suitability on Zika virus epidemic severity in Colombia.

The Zika virus epidemic of 2015-16, which caused over 1 million confirmed or suspected human cases in the Caribbean and Latin America, was driven by a combination of movement of infected humans and availability of suitable habitat for mosquito species that are key disease vectors. Both human mobility and mosquito vector abundances vary seasonally, and the goal of our research was to analyze the interacting effects of disease vector densities and human movement across metapopulations on disease transmission intensity and the probability of super-spreader events. Our research uses the novel approach of combining geographical modeling of mosquito presence with network modeling of human mobility to offer a comprehensive simulation environment for Zika virus epidemics that considers a substantial number of spatial and temporal factors compared to the literature. Specifically, we tested the hypotheses that 1) regions with the highest probability of mosquito presence will have more super-spreader events during dry months, when mosquitoes are predicted to be more abundant, 2) regions reliant on tourism industries will have more super-spreader events during wet months, when they are more likely to contribute to network-level pathogen spread due to increased travel. We used the case study of Colombia, a country with a population of about 50 million people, with an annual calendar that can be partitioned into overlapping cycles of wet and dry seasons and peak tourism and off tourism seasons that drive distinct cyclical patterns of mosquito abundance and human movement. Our results show that whether the first infected human was introduced to the network during the wet versus dry season and during the tourism versus off tourism season profoundly affects the severity and trajectory of the epidemic. For example, Zika virus was first detected in Colombia in October of 2015. Had it originated in January, a dry season month with high rates of tourism, it likely could have infected up to 60% more individuals and up to 40% more super-spreader events may have occurred. In addition, popular tourism destinations such as Barranquilla and Cartagena have the highest risk of super-spreader events during the winter, whereas densely populated areas such as Medellín and Bogotá are at higher risk of sustained transmission during dry months in the summer. Our research demonstrates that public health planning and response to vector-borne disease outbreaks requires a thorough understanding of how vector and host patterns vary due to seasonality in environmental conditions and human mobility dynamics. This research also has strong implications for tourism policy and the potential response strategies in case of an emergent epidemic.

Long-term validation of inner-urban mobility metrics derived from Twitter/X

Urban mobility analysis using Twitter as a proxy has gained significant attention in various application fields; however, long-term validation studies are scarce. This paper addresses this gap by assessing the reliability of Twitter data for modeling inner-urban mobility dynamics over a 27-month period in the. metropolitan area of Rio de Janeiro, Brazil. The evaluation involves the validation of Twitter-derived mobility estimates at both temporal and spatial scales, employing over 1.6 × 1011 mobile phone records of around three million users during the non-stationary mobility period from April 2020 to. June 2022, which coincided with the COVID-19 pandemic. The results highlight the need for caution when using Twitter for short-term modeling of urban mobility flows. Short-term inference can be influenced by Twitter policy changes and the availability of publicly accessible tweets. On the other hand, this long-term study demonstrates that employing multiple mobility metrics simultaneously, analyzing dynamic and static mobility changes concurrently, and employing robust preprocessing techniques such as rolling window downsampling can enhance the inference capabilities of Twitter data. These novel insights gained from a long-term perspective are vital, as Twitter - rebranded to X in 2023 - is extensively used by researchers worldwide to infer human movement patterns. Since conclusions drawn from studies using Twitter could be used to inform public policy, emergency response, and urban planning, evaluating the reliability of this data is of utmost importance.

Visual experience and sensitivity to motion sequences, from human movement patterns to animated shapes

Visual experience and sensitivity to motion sequences, from human movement patterns to animated shapes

Buckling-inspired triboelectric sensor for multifunctional sensing of soft robotics and wearable devices

Amidst the rapid development of intelligent robotics and wearable health monitoring devices, there is an urgent demand for flexible sensor with high sensitivity and precise feedback. In this paper, a multifunctional triboelectric sensor inspired by the phenomenon of buckling is introduced, which integrates kirigami structures with triboelectric nanogenerator principles to accurately detect one-dimensional (1D) and two-dimensional (2D) deformations. The sensor employs a contact-separating structure, with its friction layers consisting of 2D laser-cut kirigami electrodes, which deform into three-dimensional (3D) structures through tensile or compressive buckling. The tensile buckling sensor (BS) developed is capable of sensing the bending angle of soft grippers, allowing the detection of the size of grasped objects with a high recognition rate of 93.75 %. Additionally, in its 1D compressive buckling form, the sensor can be used to recognize various human motion patterns effectively, while the 2D compressive version can be used to measure the expansion of curved surfaces with a sensitivity of 1.3307 V/mm. By combining the unique properties of buckling structures, the innovative triboelectric multifunctional sensor proposed offers a new solution for high-precision motion feedback control with potential application in soft robotics and health monitoring.

Dynamic Contact Networks in Confined Spaces: Synthesizing Micro-Level Encounter Patterns through Human Mobility Models from Real-World Data.

This study advances the field of infectious disease forecasting by introducing a novel approach to micro-level contact modeling, leveraging human movement patterns to generate realistic temporal-dynamic networks. Through the incorporation of human mobility models and parameter tuning, this research presents an innovative method for simulating micro-level encounters that closely mirror infection dynamics within confined spaces. Central to our methodology is the application of Bayesian optimization for parameter selection, which refines our models to emulate both the properties of real-world infection curves and the characteristics of network properties. Typically, large-scale epidemiological simulations overlook the specifics of human mobility within confined spaces or rely on overly simplistic models. By focusing on the distinct aspects of infection propagation within specific locations, our approach strengthens the realism of such pandemic simulations. The resulting models shed light on the role of spatial encounters in disease spread and improve the capability to forecast and respond to infectious disease outbreaks. This work not only contributes to the scientific understanding of micro-level transmission patterns but also offers a new perspective on temporal network generation for epidemiological modeling.

An Effective Deep Learning Framework for Fall Detection: Model Development and Study Design.

Fall detection is of great significance in safeguarding human health. By monitoring the motion data, a fall detection system (FDS) can detect a fall accident. Recently, wearable sensors-based FDSs have become the mainstream of research, which can be categorized into threshold-based FDSs using experience, machine learning-based FDSs using manual feature extraction, and deep learning (DL)-based FDSs using automatic feature extraction. However, most FDSs focus on the global information of sensor data, neglecting the fact that different segments of the data contribute variably to fall detection. This shortcoming makes it challenging for FDSs to accurately distinguish between similar human motion patterns of actual falls and fall-like actions, leading to a decrease in detection accuracy. This study aims to develop and validate a DL framework to accurately detect falls using acceleration and gyroscope data from wearable sensors. We aim to explore the essential contributing features extracted from sensor data to distinguish falls from activities of daily life. The significance of this study lies in reforming the FDS by designing a weighted feature representation using DL methods to effectively differentiate between fall events and fall-like activities. Based on the 3-axis acceleration and gyroscope data, we proposed a new DL architecture, the dual-stream convolutional neural network self-attention (DSCS) model. Unlike previous studies, the used architecture can extract global feature information from acceleration and gyroscope data. Additionally, we incorporated a self-attention module to assign different weights to the original feature vector, enabling the model to learn the contribution effect of the sensor data and enhance classification accuracy. The proposed model was trained and tested on 2 public data sets: SisFall and MobiFall. In addition, 10 participants were recruited to carry out practical validation of the DSCS model. A total of 1700 trials were performed to test the generalization ability of the model. The fall detection accuracy of the DSCS model was 99.32% (recall=99.15%; precision=98.58%) and 99.65% (recall=100%; precision=98.39%) on the test sets of SisFall and MobiFall, respectively. In the ablation experiment, we compared the DSCS model with state-of-the-art machine learning and DL models. On the SisFall data set, the DSCS model achieved the second-best accuracy; on the MobiFall data set, the DSCS model achieved the best accuracy, recall, and precision. In practical validation, the accuracy of the DSCS model was 96.41% (recall=95.12%; specificity=97.55%). This study demonstrates that the DSCS model can significantly improve the accuracy of fall detection on 2 publicly available data sets and performs robustly in practical validation.

EM-Gait: Gait recognition using motion excitation and feature embedding self-attention

Gait recognition, which can realize long-distance and contactless identification, is an important biometric technology. Recent gait recognition methods focus on learning the pattern of human movement or appearance during walking, and construct the corresponding spatio-temporal representations. However, different individuals have their own laws of movement patterns, simple spatial–temporal features are difficult to describe changes in motion of human parts, especially when confounding variables such as clothing and carrying are included, thus distinguishability of features is reduced. To this end, we propose the Embedding and Motion (EM) block and Fine Feature Extractor (FFE) to capture the motion mode of walking and enhance the difference of local motion rules. The EM block consists of a Motion Excitation (ME) module to capture the changes of temporal motion and an Embedding Self-attention (ES) module to enhance the expression of motion rules. Specifically, without introducing additional parameters, ME module learns the difference information between frames and intervals to obtain the dynamic change representation of walking for frame sequences with uncertain length. By contrast, ES module divides the feature map hierarchically based on element values, blurring the difference of elements to highlight the motion track. Furthermore, we present the FFE, which independently learns the spatio-temporal representations of human body according to different horizontal parts of individuals. Benefiting from EM block and our proposed motion branch, our method innovatively combines motion change information, significantly improving the performance of the model under cross appearance conditions. On the popular dataset CASIA-B, our proposed EM-Gait is better than the existing single-modal gait recognition methods.

Ultrasensitive electrospinning fibrous strain sensor with synergistic conductive network for human motion monitoring and human-computer interaction

With the rapid development of wearable electronic skin technology, flexible strain sensors have shown great application prospects in the fields of human motion and physiological signal detection, medical diagnostics, and human-computer interaction owing to their outstanding sensing performance. This paper reports a strain sensor with synergistic conductive network, consisting of stable carbon nanotube dispersion (CNT) layer and brittle MXene layer by dip-coating and electrostatic self-assembly method, and breathable three-dimensional (3D) flexible substrate of thermoplastic polyurethane (TPU) fibrous membrane prepared through electrospinning technology. The MXene/CNT@PDA-TPU (MC@p-TPU) flexible strain sensor had excellent air permeability, wide operating range (0–450 %), high sensitivity (Gauge Factor, GFmax = 8089.7), ultra-low detection limit (0.05 %), rapid response and recovery times (40 ms/60 ms), and excellent cycle stability and durability (10,000 cycles). Given its superior strain sensing capabilities, this sensor can be applied in physiological signals detection, human motion pattern recognition, and driving exoskeleton robots. In addition, MC@p-TPU fibrous membrane also exhibited excellent photothermal conversion performance and can be used as a wearable photo-heater, which has far-reaching application potential in the photothermal therapy of human joint diseases.

A gait phase recognition method for obstacle crossing based on multi-sensor fusion

The analysis of human motion patterns and gait phase detection holds significant importance for both human motion analysis and precise control of exoskeleton devices. Presently, research on gait during obstacle crossing primarily focuses on the medical rehabilitation domain, with limited studies concerning gait phase analysis during obstacle crossing. This study conducts a detailed analysis of gait data during obstacle crossing and proposes a CNN-PCA-LSTM algorithm that fuses multi-sensor data to accurately identify gait phases when crossing obstacles in the horizontal direction. A wearable multi-sensor data acquisition system was designed, utilizing flexible capacitive pressure sensors to collect pressure data from the soles of both feet and IMUs to gather foot motion data. Kinematic simulation analyses of human obstacle-crossing motions were performed using Anybody software to determine various gait phases, validating the efficacy of the simulation model. Subsequently, one-dimensional and two-dimensional convolutional neural networks were separately employed to extract features from sole pressure data and foot motion data. Principal Component Analysis (PCA) was utilized to reduce the dimensionality of these feature datasets, which were then inputted into LSTM networks. Finally, the Softmax function was employed for the classification and recognition of gait phases when crossing obstacles in the horizontal direction. The experimental results indicate that employing the CNN-PCA-LSTM algorithm integrating multiple sensor data achieves a recognition accuracy of 97.91 % for identifying gait phases during horizontal obstacle crossing.

Wearable sensors based on artificial intelligence models for human activity recognition.

Human motion detection technology holds significant potential in medicine, health care, and physical exercise. This study introduces a novel approach to human activity recognition (HAR) using convolutional neural networks (CNNs) designed for individual sensor types to enhance the accuracy and address the challenge of diverse data shapes from accelerometers, gyroscopes, and barometers. Specific CNN models are constructed for each sensor type, enabling them to capture the characteristics of their respective sensors. These adapted CNNs are designed to effectively process varying data shapes and sensor-specific characteristics to accurately classify a wide range of human activities. The late-fusion technique is employed to combine predictions from various models to obtain comprehensive estimates of human activity. The proposed CNN-based approach is compared to a standard support vector machine (SVM) classifier using the one-vs-rest methodology. The late-fusion CNN model showed significantly improved performance, with validation and final test accuracies of 99.35 and 94.83% compared to the conventional SVM classifier at 87.07 and 83.10%, respectively. These findings provide strong evidence that combining multiple sensors and a barometer and utilizing an additional filter algorithm greatly improves the accuracy of identifying different human movement patterns.

The Impact of Scale on Extracting Individual Mobility Patterns from Location-Based Social Media.

Understanding human movement patterns is crucial for comprehending how a city functions. It is also important for city planners and policymakers to create more efficient plans and policies for urban areas. Traditionally, human movement patterns were analyzed using origin-destination surveys, travel diaries, and other methods. Now, these patterns can be identified from various geospatial big data sources, such as mobile phone data, floating car data, and location-based social media (LBSM) data. These extensive datasets primarily identify individual or collective human movement patterns. However, the impact of spatial scale on the analysis of human movement patterns from these large geospatial data sources has not been sufficiently studied. Changes in spatial scale can significantly affect the results when calculating human movement patterns from these data. In this study, we utilized Weibo datasets for three different cities in China including Beijing, Guangzhou, and Shanghai. We aimed to identify the effect of different spatial scales on individual human movement patterns as calculated from LBSM data. For our analysis, we employed two indicators as follows: an external activity space indicator, the radius of gyration (ROG), and an internal activity space indicator, entropy. These indicators were chosen based on previous studies demonstrating their efficiency in analyzing sparse datasets like LBSM data. Additionally, we used two different ranges of spatial scales-10-100 m and 100-3000 m-to illustrate changes in individual activity space at both fine and coarse spatial scales. Our results indicate that although the ROG values show an overall increasing trend and the entropy values show an overall decreasing trend with the increase in spatial scale size, different local factors influence the ROG and entropy values at both finer and coarser scales. These findings will help to comprehend the dynamics of human movement across different scales. Such insights are invaluable for enhancing overall urban mobility and optimizing transportation systems.

Hazard exposure heterophily in socio-spatial networks contributes to post-disaster recovery in low-income populations

Despite recognition of the influence of socio-spatial networks on disaster recovery, limited data-driven evidence exists regarding the nature of this relationship. To address this knowledge gap, this study investigated the duration of the human mobility recovery process and its relationship with hazard-exposure heterophily (as a measure for the resourcefulness of social ties) and various other socio-demographic characteristics. We applied the concept of hazard-exposure heterophily to the possibility of resource sharing, through social-spatial networks, between communities with different hazard exposures. We used human movement patterns to measure recovery duration. To understand the disparities during recovery, we examined the correlations between recovery duration and the social demographic characteristics of race and income caused by Hurricane Harvey which made landfall in Harris County, Texas, on August 25, 2017. We discerned high/low hazard-exposure heterophily through the use of a threshold-classification method, and we got four clusters according to the flood-inundation percentage and social connectedness values. Then we applied correlation analysis to analyze the relations between hazard-exposure heterophily/income and recovery weeks among different income groups. The results revealed that hazard-exposure heterophily considerably accelerates human mobility recovery in low-income areas, as the hazard-exposure heterophily allows for the increased exchange of knowledge and resources between members of diverse groups, ultimately accelerating the recovery process. The results of this study could benefit local agents to better allocate recovery resources.

Development of a protocol for using geo-trackers to identify zoonotic enteric pathogen transmission pathways in a pilot study in Kenya

Abstract Humans and animals can be exposed to fecally-transmitted pathogens in both private and public domain environments. Animals may also acquire pathogen infections from these two environments and transmit them to humans through different pathways. Evidence on how often and where interactions occur between the two environments could improve the effectiveness of public health programs for preventing zoonotic disease transmission and response to disease outbreaks. This study aimed to develop a protocol for using geo-trackers to measure the spatial-temporal movement and interaction between animals and children in households and public areas in low- and middle-income categories of urban neighbourhoods in Kenya. It also aimed to identify opportunities and challenges for the scale-up of these methods’ for surveillance of other zoonotic diseases of public health significance. One commercial geo-tracker device with the best technological performance and usability that met pre-defined criteria for the study context was identified. Community engagement meetings were then conducted to gather input on a proposed study protocol. Afterwards, infants and animals were geo-tracked in 10 households in urban informal settlements of Kibera and Jericho, Nairobi over two consecutive weeks with iterative improvements to protocols. The effectiveness of the geo-tracking exercise was evaluated through in-depth interviews with Community Health Volunteers (CHVs) and infant caregivers. Animal and infant behaviour and battery reliability of the geo-trackers were also monitored; and observed opportunities and challenges in implementing the protocol during the exercise were documented. Community members were receptive and accepted the use of geo-trackers on animals and children. In pilot testing, there was no change of behaviour from 10 infants tracked. Discomfort was observed for up to 30 min after the placement for some of the seven animals tracked, but the animals quickly adjusted. The battery for all geo-tracker devices lasted for the 24-h geo-tracking period. Some caregivers and CHVs were concerned whether the geo-trackers could record personal information. It was shown that geo-tracker devices can be successfully deployed to study animal-child interactions and movement in different categories of urban neighbourhoods. Recommendations have been made on the lessons learnt from the study to help scientists who would use geo-trackers for future community-based human and animal research. One Health impact statement Assessing human and animal risks of acquiring enteric infections from environmental contamination, or contributing to pathogen contamination of the environment requires understanding spatial-temporal patterns of human and animal movement between private households and the community. Movement of domestic animals is relevant for understanding household-community spatial patterns of pathogen transmission because animal ownership and contact with humans is common, and free-roaming animals carrying enteric pathogens can spread diverse faecal pathogens between environments. The community-based participatory research approach of this study evaluated implementation feasibility from the lens of technical usability and reliability of geo-tracking devices, as well as social acceptability and ethics related to using such devices for the community-based study of human and animal subjects in urban settings. Evidence and recommendations from this study will be useful to anyone interested in understanding human-animal-environment interactions and their contributions to infectious disease transmission.

The importance of understanding kinesiology for gymnastics: A biomechanist's perspective

Kinesiology has progressed to become one of the most dynamic disciplines for studying and scrutinizing the various aspects of human bodily movement from a scientific perspective. The field has expanded into major areas of sports such as athletics, martial arts, gymnastics, and more. This study aimed to delve deeper into this subject and explore the importance of understanding kinesiology in gymnastics, as it provides a comprehensive context for studying such aspects. The primary goal was to assess the opinions of biomechanists to gain insight into the underlying factors. The study employed a quantitative design, using an original research approach to report findings on the current subject. A total of 36 biomechanists (29 males and 7 females) participated in this study, and data were collected using a questionnaire. The questionnaire included demographic questions and 24 statements. These statements were developed based on extensive evaluations of kinesiology features and their associations or implications for gymnastics. The study outcomes revealed that, over time, gymnasts are becoming increasingly aware of the importance of having a solid understanding of kinesiology for analyzing human body movement patterns. They are optimistically applying this knowledge to optimize their training routines, use their energy and skills more efficiently, and ensure better performance and faster recovery from physical injuries when faced with such challenges.

Comparison of computational pose estimation models for joint angles with 3D motion capture

Tools to calculate human movement patterns can benefit musculoskeletal clinicians and researchers for rehabilitation assessments. The research objective of this study was to compare two human pose estimation models (HRNet, MediaPipe) against the laboratory marker-based reference standard for joint angles and range of motion (ROM) for several movement parameters. Twenty-two healthy volunteers (Female n = 16, Male n = 6), participated to compare outputs for knee and elbow kinematics. Joint angles were calculated by selecting three marker points defining the joint and angle between them in Qualisys Track Manager software. Using predicted key points, pose estimation model calculations for the same musculoskeletal kinematic outputs were computed. Coefficient of Variation (CoV) was used as a variation statistic for joint angle during movements. All comparison results were under 10%, implying that both models compute reliable joint angle data during the five tested activities. When comparing ROM as a discrete parameter, CoV values remain low, though not all below 10%. Intra-class Correlation Coefficients were computed across the ROM data as a measure of statistical similarity. Each exercise displayed good-excellent and significant correlations for both models compared to Qualisys apart from left knee sit-to-stand. Exploration from this data sampling imply that flexion/extension exercises give stronger consistency results than full sit-to-stand movements when compared to 3D motion analysis, and there is little distinction between these two models. Finer tuning of models will give further reliability for in-depth analysis as these results are restricted, but valuable for a rehabilitative setting with limited objective analysis alternative.

Human motion activity recognition and pattern analysis using compressed deep neural networks

ABSTRACT This work presents an on-device machine learning model with the ability to identify different mobility gestures called human activity recognition (HAR), which includes running, walking, squatting, jumping, and others. The data is collected through an Arduino Nano 33 BLE Sense board with a sampling rate of 119 Hz, which is embedded with an Inertial Measurement Unit (IMU) sensor. The same board is used as a microcontroller to identify human gestures by developing an end-to-end edge computing application. A deep neural network model is trained and then compressed for deployment on the board to create a self-contained, embedded device capable of identifying the type of gesture performed. Three deep learning models, namely Multi-Layer Perceptron (MLP), Convolutional Neural Network – Long Short Term Memory (CNN-LSTM) & CNN-Gated Recurrent Unit (CNN-GRU), are evaluated in the identification of the mobility gestures. The observed accuracy of the models is 96%, 97.1% and 97.8%, respectively, MLP, CNN-LSTM & CNN-GRU across the different gesture categories. The study shows the utility of embedded devices with deep neural network-based models, which can provide low cost, minimal power usage, and meet data privacy requirements in HAR.

Walk This Way: How Weight Distorts Gender Identification of Point-Light Walkers

Extensive research in social perception and biological motion has converged on the finding that humans are particularly accurate in identifying gender from the gait of minimal visual conspecific stimuli (e.g., point-light walkers). Despite the preponderance of evidence in favor of this ability, we return to the original paradigm and vary a single parameter—weight. Across nine pre-registered studies, participants (N = 3,196) were assigned to view the gait of point-light walkers based on actual human motion patterns. We find a decline in the accuracy of identifying the gender of female point-light walkers as their weight increases. However, as the weight of female walkers decreases, gender identification accuracy is recovered. These findings carry implications for the gendered nature of weight bias and the role of weight in human perception.