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

BACKGROUND: 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. OBJECTIVE: 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. METHODS: 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. RESULTS: 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%). CONCLUSIONS: 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.

Design and Characterization of a Wearable Inertial Measurement Unit.

The utilization of inertial measurement units as wearable sensors is proliferating across various domains, such as health care, sports, and rehabilitation. This expansion has produced a market of devices tailored to accommodate very specific ranges of operational demands. Simultaneously, this growth is creating opportunities for the development of a new class of devices more oriented towards general-purpose use and capable of capturing both high-frequency signals for short-term, event-driven motion analysis and low-frequency signals for extended monitoring. For such a design, which combines flexibility and low cost, a rigorous evaluation of the device in terms of deviation, noise levels, and precision is essential. This evaluation is crucial for identifying potential improvements and refining the design accordingly, yet it is rarely addressed in the literature. This paper presents the development process of such a device. The results of the design process demonstrate acceptable performance in optimizing energy consumption and storage capacity while highlighting the most critical optimizations needed to advance the device towards the goal of a smart, general-purpose unit for human motion monitoring.

Validation of Step Detection and Distance Calculation Algorithms for Soccer Performance Monitoring.

This study focused on developing and evaluating a gyroscope-based step counter algorithm using inertial measurement unit (IMU) readings for precise athletic performance monitoring in soccer. The research aimed to provide reliable step detection and distance estimation tailored to soccer-specific movements, including various running speeds and directional changes. Real-time algorithms utilizing shank angular data from gyroscopes were created. Experiments were conducted on a specially designed soccer-specific testing circuit performed by 15 athletes, simulating a range of locomotion activities such as walking, jogging, and high-intensity actions. The algorithm outcome was compared with manually tagged data from a high-quality video camera-based system for validation, by assessing the agreement between the paired values using limits of agreement, concordance correlation coefficient, and further metrics. Results returned a step detection accuracy of 95.8% and a distance estimation Root Mean Square Error (RMSE) of 17.6 m over about 202 m of track. A sub-sample (N = 6) also wore two pairs of devices concurrently to evaluate inter-unit reliability. The performance analysis suggested that the algorithm was effective and reliable in tracking diverse soccer-specific movements. The proposed algorithm offered a robust and efficient solution for tracking step count and distance covered in soccer, particularly beneficial in indoor environments where global navigation satellite systems are not feasible. This advancement in sports technology widens the spectrum of tools for coaches and athletes in monitoring soccer performance.

A Segmented Cross-Correlation Algorithm for Dynamic North Finding Using Fiber Optic Gyroscopes.

Fiber optic gyroscope (FOG)-based north finding is extensively applied in navigation, positioning, and various fields. In dynamic north finding, an accelerated turntable speed shortens the time required for north finding, resulting in a rapid north-finding response. However, with an increase in turntable speed, the turntable's jitter contributes to signal contamination in the FOG, leading to a deterioration in north-finding accuracy. This paper introduces a divide-and-conquer algorithm, the segmented cross-correlation algorithm, designed to mitigate the impact of turntable speed jitter. A model for north-finding error is established and analyzed, incorporating FOG's self-noise and the turntable's speed jitter. To validate the feasibility of our method, we implemented the algorithm on a FOG. The simulation and experimental results exhibited a strong concordance, affirming the validity of our proposed north-finding error model. The experimental findings indicate that, at a turntable speed of 180°/s, the north-finding bias error within a 360 s duration is 0.052°, representing a 64% improvement over the traditional algorithm. These results indicate the effectiveness of the proposed algorithm in mitigating the impact of unstable turntable speeds, offering a solution for north finding with both prompt response and enhanced accuracy.

Gesture-Based Interactions: Integrating Accelerometer and Gyroscope Sensors in the Use of Mobile Apps.

This study investigates the feasibility and functionality of accelerometer and gyroscope sensors for gesture-based interactions in mobile app user experience. The core of this innovative approach lies in introducing a dynamic and intuitive user interaction model with the device sensors. The Android app developed for this purpose has been created for its use in controlled experiments. Methodologically, it was created as a stand-alone tool to both capture quantitative (time, automatically captured) and qualitative (behavior, collected with post-task questionnaires) variables. The app's setting features a set of modules with two levels each (randomized presentation applied, minimizing potential learning effects), allowing users to interact with both sensor-based and traditional touch-based scenarios. Preliminary results with 22 participants reveal that tasks involving sensor-based interactions tend to take longer to complete when compared to the traditional ones. Remarkably, many participants rated sensor-based interactions as a better option than touch-based interactions, as seen in the post-task questionnaires. This apparent discrepancy between objective completion times and subjective user perceptions requires a future in-depth exploration of factors influencing user experiences, including potential learning curves, cognitive load, and task complexity. This study contributes to the evolving landscape of mobile app user experience, emphasizing the benefits of considering the integration of device sensors (and gesture-based interactions) in common mobile usage.

Hammerstein-Wiener Motion Artifact Correction for Functional Near-Infrared Spectroscopy: A Novel Inertial Measurement Unit-Based Technique.

Participant movement is a major source of artifacts in functional near-infrared spectroscopy (fNIRS) experiments. Mitigating the impact of motion artifacts (MAs) is crucial to estimate brain activity robustly. Here, we suggest and evaluate a novel application of the nonlinear Hammerstein-Wiener model to estimate and mitigate MAs in fNIRS signals from direct-movement recordings through IMU sensors mounted on the participant's head (head-IMU) and the fNIRS probe (probe-IMU). To this end, we analyzed the hemodynamic responses of single-channel oxyhemoglobin (HbO) and deoxyhemoglobin (HbR) signals from 17 participants who performed a hand tapping task with different levels of concurrent head movement. Additionally, the tapping task was performed without head movements to estimate the ground-truth brain activation. We compared the performance of our novel approach with the probe-IMU and head-IMU to eight established methods (PCA, tPCA, spline, spline Savitzky-Golay, wavelet, CBSI, RLOESS, and WCBSI) on four quality metrics: SNR, △AUC, RMSE, and R. Our proposed nonlinear Hammerstein-Wiener method achieved the best SNR increase (p < 0.001) among all methods. Visual inspection revealed that our approach mitigated MA contaminations that other techniques could not remove effectively. MA correction quality was comparable with head- and probe-IMUs.

Eating Event Recognition Using Accelerometer, Gyroscope, Piezoelectric, and Lung Volume Sensors.

In overcoming the worldwide problem of overweight and obesity, automatic dietary monitoring (ADM) is introduced as support in dieting practises. ADM aims to automatically, continuously, and objectively measure dimensions of food intake in a free-living environment. This could simplify the food registration process, thereby overcoming frequent memory, underestimation, and overestimation problems. In this study, an eating event detection sensor system was developed comprising a smartwatch worn on the wrist containing an accelerometer and gyroscope for eating gesture detection, a piezoelectric sensor worn on the jaw for chewing detection, and a respiratory inductance plethysmographic sensor consisting of two belts worn around the chest and abdomen for food swallowing detection. These sensors were combined to determine to what extent a combination of sensors focusing on different steps of the dietary cycle can improve eating event classification results. Six subjects participated in an experiment in a controlled setting consisting of both eating and non-eating events. Features were computed for each sensing measure to train a support vector machine model. This resulted in F1-scores of 0.82 for eating gestures, 0.94 for chewing food, and 0.58 for swallowing food.

Spacecraft Attitude Measurement and Control Using VSMSCSG and Fractional-Order Zeroing Neural Network Adaptive Steering Law.

In order to improve the accuracy and convergence speed of the steering law under the conditions of high dynamics, high bandwidth, and a small deflection angle, and in an effort to improve attitude measurement and control accuracy of the spacecraft, a spacecraft attitude measurement and control method based on variable speed magnetically suspended control sensitive gyroscopes (VSMSCSGs) and the fractional-order zeroing neural network (FO-ZNN) steering law is proposed. First, a VSMSCSG configuration is designed to realize attitude measurement and control integration in which the VSMSCSGs are employed as both actuators and attitude-rate sensors. Second, a novel adaptive steering law using FO-ZNN is designed. The matrix pseudoinverses are replaced by FO-ZNN outputs, which solves the problem of accuracy degradation in the traditional pseudoinverse steering laws due to the complexity of matrix pseudoinverse operations under high dynamics conditions. In addition, the convergence and robustness of the FO-ZNN are proven. The results show that the proposed FO-ZNN converges faster than the traditional zeroing neural network under external disturbances. Finally, a new weighting function containing rotor deflection angles is added to the steering law to ensure that the saturation of the rotor deflection angles can be avoided. Semi-physical simulation results demonstrate the correctness and superiority of the proposed method.

Going beyond PA: Assessing sensorimotor capacity with wearables in multiple sclerosis-a cross-sectional study.

BACKGROUND: Wearable technologies are currently clinically used to assess energy expenditure in a variety of populations, e.g., persons with multiple sclerosis or frail elderly. To date, going beyond physical activity, deriving sensorimotor capacity instead of energy expenditure, is still lacking proof of feasibility. METHODS: In this study, we read out sensors (accelerometer and gyroscope) of smartwatches in a sample of 90 persons with multiple sclerosis over the course of one day of everyday life in an inpatient setting. We derived a variety of different kinematic parameters, in addition to lab-based tests of sensorimotor performance, to examine their interrelation by principal component, cluster, and regression analyses. RESULTS: These analyses revealed three components of behavior and sensorimotor capacity, namely clinical characteristics with an emphasis on gait, gait-related physical activity, and upper-limb related physical activity. Further, we were able to derive four clusters with different behavioral/capacity patterns in these dimensions. In a last step, regression analyses revealed that three selected smartwatch derived kinematic parameters were able to partially predict sensorimotor capacity, e.g., grip strength and upper-limb tapping. CONCLUSIONS: Our analyses revealed that physical activity can significantly differ between persons with comparable clinical characteristics and that assessments of physical activity solely relying on gait can be misleading. Further, we were able to extract parameters that partially go beyond physical activity, with the potential to be used to monitor the course of disease progression and rehabilitation, or to early identify persons at risk or a sub-clinical threshold of disease severity.

Combined Polarization/Magnetic Modulation of a Transverse NMR Gyroscope.

In this paper, we describe a new approach to the continuous operation of a transverse spin-exchange optically pumped NMR gyroscope that utilizes modulation of both the applied bias field and the optical pumping. We demonstrate the simultaneous, continuous excitation of 131Xe and 129Xe using this hybrid modulation approach and the real-time demodulation of the Xe precession using a custom least-squares fitting algorithm. We present rotation rate measurements with this device, with a common field suppression factor of ∼1400, an angle random walk of 21 µHz/Hz, and a bias instability of ∼480 nHz after ∼1000 s.

Homogeneous Sensor Fusion Optimization for Low-Cost Inertial Sensors.

The article deals with sensor fusion and real-time calibration in a homogeneous inertial sensor array. The proposed method allows for both estimating the sensors' calibration constants (i.e., gain and bias) in real-time and automatically suppressing degraded sensors while keeping the overall precision of the estimation. The weight of the sensor is adaptively adjusted according to the RMSE concerning the weighted average of all sensors. The estimated angular velocity was compared with a reference (ground truth) value obtained using a tactical-grade fiber-optic gyroscope. We have experimented with low-cost MEMS gyroscopes, but the proposed method can be applied to basically any sensor array.

A Prototype for a Passive Resonant Interferometric Fiber Optic Gyroscope with a 3 × 3 Directional Coupler.

Reducing the dimensions of optical gyroscopes is a crucial task and resonant fiber optic gyroscopes are promising candidates for its solution. The paper presents a prototype of a miniature resonant interferometric gyroscope of a strategic accuracy class. Due to the use of passive optical elements in this gyroscope, it has a great potential for miniaturization, alongside a low production cost and ease of implementation, since it does not require many feedback loops. The presented prototype shows results on a zero instability of 20°/h and an angle random walk of 0.16°/√h. A theoretical model explaining the nature of the multipath interference of resonant spectra and establishing the relationship between the resonator parameters and the output parameters of the presented prototype is proposed. The results predicted are in agreement with the experimental data. The prototype gyroscope demonstrates a scale factor instability and a change in the average signal level, which is due to the presence of polarization non-reciprocity, occurring due to the induced birefringence in the single-mode fiber of the contour. This problem requires further investigation to be performed.

An Effective Sensor Architecture for Full-Attitude Determination in the HERMES Nano-Satellites.

The High Energy Rapid Modular Ensemble of Satellites (HERMES) is a constellation of 3U nano-satellites for high energy astrophysics. The HERMES nano-satellites' components have been designed, verified, and tested to detect and localize energetic astrophysical transients, such as short gamma-ray bursts (GRBs), which are the electromagnetic counterparts of gravitational wave events, thanks to novel miniaturized detectors sensitive to X-rays and gamma-rays. The space segment is composed of a constellation of CubeSats in low-Earth orbit (LEO), ensuring an accurate transient localization in a field of view of several steradians exploiting the triangulation technique. To achieve this goal, guaranteeing a solid support to future multi-messenger astrophysics, HERMES shall determine its attitude and orbital states with stringent requirements. The scientific measurements bind the attitude knowledge within 1 deg (1σa) and the orbital position knowledge within 10 m (1σo). These performances shall be reached considering the mass, volume, power, and computation constraints of a 3U nano-satellite platform. Thus, an effective sensor architecture for full-attitude determination was developed for the HERMES nano-satellites. The paper describes the hardware typologies and specifications, the configuration on the spacecraft, and the software elements to process the sensors' data to estimate the full-attitude and orbital states in such a complex nano-satellite mission. The aim of this study was to fully characterize the proposed sensor architecture, highlighting the available attitude and orbit determination performance and discussing the calibration and determination functions to be implemented on-board. The presented results derived from model-in-the-loop (MIL) and hardware-in-the-loop (HIL) verification and testing activities and can serve as useful resources and a benchmark for future nano-satellite missions.

An Interface ASIC Design of MEMS Gyroscope with Analog Closed Loop Driving.

This paper introduces a digital interface application-specific integrated circuit (ASIC) for a micro-electromechanical systems (MEMS) vibratory gyroscope. The driving circuit of the interface ASIC uses an automatic gain circuit (AGC) module instead of a phase-locked loop to realize a self-excited vibration, which gives the gyroscope system good robustness. In order to realize the co-simulation of the mechanically sensitive structure and interface circuit of the gyroscope, the equivalent electrical model analysis and modeling of the mechanically sensitive structure of the gyro are carried out by Verilog-A. According to the design scheme of the MEMS gyroscope interface circuit, a system-level simulation model including mechanically sensitive structure and measurement and control circuit is established by SIMULINK. A digital-to-analog converter (ADC) is designed for the digital processing and temperature compensation of the angular velocity in the MEMS gyroscope digital circuit system. Using the positive and negative diode temperature characteristics, the function of the on-chip temperature sensor is realized, and the temperature compensation and zero bias correction are carried out simultaneously. The MEMS interface ASIC is designed using a standard 0.18 µM CMOS BCD process. The experimental results show that the signal-to-noise ratio (SNR) of sigma-delta (ΣΔ) ADC is 111.56 dB. The nonlinearity of the MEMS gyroscope system is 0.03% over the full-scale range.

Sagnac Effect Compensations and Locked States in a Ring Laser Gyroscope.

Frequency lock-in-induced deadband phenomena are major problems of ring laser gyroscopes (RLGs), which deteriorate linear responses to changes in the applied rotation rate. In this work, the frequency lock-in phenomenon occurring in the RLG was successfully investigated by compensating for the Sagnac effect through frequency analysis using a newly defined error function. Integrative and generalized viewpoints from the analyzed results provide new possibilities for relevant performance improvements of optical gyroscopes, as well as a deeper understanding of locked states in principle aspects.

Digital Control and Demodulation Algorithm for Compact Open-Loop Fiber-Optic Gyroscope.

With the advantages of small size, low cost, and moderate accuracy, an open-loop fiber-optic gyroscope (FOG) has a wide range of applications around control and automation. For the most cost-sensitive applications, a simple and stable digital algorithm with a reduced control-circuit volume and cost is highly desirable to realize high-precision control of a FOG. In this work, a new algorithm for an open-loop FOG is proposed based on the discrete multi-point demodulation in the sinusoidal modulation period. Utilizing this algorithm, stable control and angular velocity calculation of a gyro are realized with effectively suppressed gyro error. The use of this algorithm greatly reduces the requirements for processing power and simplifies the gyro circuit. Based on this algorithm, a digital FOG with a volume of only 25 × 20 × 40 mm3 achieves a bias instability of less than 0.15°/h, an angle random walk (ARW) of less than 0.015°/√h, a start-up time of less than 1 s, and a 3 dB bandwidth beyond 160 Hz. This low-cost, compact, and high-performance gyro is sufficient to satisfy the requirements of applications in the navigation and control fields such as unmanned driving.

Automatic Sequential Stitching of High-Resolution Panorama for Android Devices Using Precapture Feature Detection and the Orientation Sensor.

Image processing on smartphones, which are resource-limited devices, is challenging. Panorama generation on modern mobile phones is a requirement of most mobile phone users. This paper presents an automatic sequential image stitching algorithm with high-resolution panorama generation and addresses the issue of stitching failure on smartphone devices. A robust method is used to automatically control the events involved in panorama generation from image capture to image stitching on Android operating systems. The image frames are taken in a firm spatial interval using the orientation sensor included in smartphone devices. The features-based stitching algorithm is used for panorama generation, with a novel modification to address the issue of stitching failure (inability to find local features causes this issue) when performing sequential stitching over mobile devices. We also address the issue of distortion in sequential stitching. Ultimately, in this study, we built an Android application that can construct a high-resolution panorama sequentially with automatic frame capture based on an orientation sensor and device rotation. We present a novel research methodology (called "Sense-Panorama") for panorama construction along with a development guide for smartphone developers. Based on our experiments, performed by Samsung Galaxy SM-N960N, which carries system on chip (SoC) as Qualcomm Snapdragon 845 and a CPU of 4 × 2.8 GHz Kyro 385, our method can generate a high-resolution panorama. Compared to the existing methods, the results show improvement in visual quality for both subjective and objective evaluation.

A Developed Jerk Sensor for Seismic Vibration Measurements: Modeling, Simulation and Experimental Verification.

Acceleration-based sensors are widely used in indicating the severity of damage caused to structural buildings during dynamic events. The force rate of change is of interest when investigating the effect of seismic waves on structural elements, and hence the calculation of the jerk is necessary. For most sensors, the technique used for measuring the jerk (m/s3) is based on differentiating the time-acceleration signal. However, this technique is prone to errors especially in small amplitude and low frequency signals, and is deemed not suitable when online feedback is required. Here, we show that direct measurement of the jerk can be achieved using a metal cantilever and a gyroscope. In addition, we focus on the development of the jerk sensor for seismic vibrations. The adopted methodology optimized the dimensions of an austenitic stainless steel cantilever and enhanced the performance in terms of sensitivity and the jerk measurable range. We found, after several analytical and FE analyses, that an L-35 cantilever model with dimensions 35 × 20 × 0.5 (mm3) and a natural frequency of 139 (Hz) has a remarkable performance for seismic measurements. Our theoretical and experimental results show that the L-35 jerk sensor has a constant sensitivity value of 0.05 ((deg/s)/(G/s)) with ±2% error in the seismic frequency bandwidth of 0.1~40 (Hz) and for amplitudes in between 0.1 and 2 (G). Furthermore, the theoretical and experimental calibration curves show linear trends with a high correlation factor of 0.99 and 0.98, respectively. These findings demonstrate the enhanced sensitivity of the jerk sensor, which surpasses previously reported sensitivities in the literature.

Comparison of Grip Strength, Forearm Muscle Activity, and Shock Transmission between the Forehand Stroke Technique of Experienced and Recreational Tennis Players Using a Novel Wearable Device.

Upper limb tennis injuries are primarily chronic, resulting from repetitive overuse. We developed a wearable device which simultaneously measures risk factors (grip strength, forearm muscle activity, and vibrational data) associated with elbow tendinopathy development resulting from tennis players' technique. We tested the device on experienced (n = 18) and recreational (n = 22) tennis players hitting forehand cross-court at both flat and topspin spin levels under realistic playing conditions. Using statistical parametric mapping analysis, our results showed that all players showed a similar level of grip strength at impact, regardless of spin level, and the grip strength at impact did not influence the percentage of impact shock transfer to the wrist and elbow. Experienced players hitting with topspin exhibited the highest ball spin rotation, low-to-high swing path brushing action, and shock transfer to the wrist and elbow compared to the results obtained while hitting the ball flat, or when compared to the results obtained from recreational players. Recreational players exhibited significantly higher extensor activity during most of the follow through phase compared to the experienced players for both spin levels, potentially putting them at greater risk for developing lateral elbow tendinopathy. We successfully demonstrated that wearable technologies can be used to measure risk factors associated with elbow injury development in tennis players under realistic playing conditions.

Methods for Comprehensive Calibration of a Low-Frequency Angular Acceleration Rotary Table.

The total harmonic distortion (THD) index and its calculation methods are presented to calibrate the sinusoidal motion of the low-frequency angular acceleration rotary table (LFAART) and make up the incomprehensive evaluation based on the angular acceleration amplitude and frequency error indexes. The THD is calculated from two measurement schemes: a unique scheme combining the optical shaft encoder and the laser triangulation sensor and a regular scheme using the fiber optical gyroscope (FOG). An improved reversing moments recognition method is presented to upgrade the accuracy of solving the angular motion amplitude based on optical shaft encoder output. The field experiment shows that the difference in the THD values achieved using the combining scheme and FOG is within 0.11% when the signal-to-noise ratio of the FOG signal is higher than 7.7 dB, indicating the accuracy of the proposed methods and the feasibility of taking THD as the index.