Search Results (870)

The integration of flexible electronics and machine learning (ML) algorithms has become a revolutionary force driving the field of intelligent sensing, giving rise to a new generation of intelligent devices and systems. This article provides a systematic review of core technologies and practical applications of ML in flexible electronics. It focuses on analyzing the theoretical frameworks of algorithms such as the Long Short-Term Memory Network (LSTM), Convolutional Neural Network (CNN), and Reinforcement Learning (RL) in the intelligent processing of sensor signals (IPSS), multimodal feature extraction (MFE), process defect and anomaly detection (PDAD), and data compression and edge computing (DCEC). This study explores the performance advantages of these technologies in optimizing signal analysis accuracy, compensating for interference in high-noise environments, optimizing manufacturing process parameters, etc., and empirically analyzes their potential applications in wearable health monitoring systems, intelligent control of soft robots, performance optimization of self-powered devices, and intelligent perception of epidermal electronic systems. Full article

This review provides a comprehensive analysis of modern strategies for the synthesis, functionalization, and application of carbon nanotubes (CNTs) and graphene for the development of high-performance polymer composites in the field of strain sensing. The paper systematically organizes key synthesis methods for CNTs and graphene (chemical vapor deposition (CVD), such as arc discharge, laser ablation, microwave synthesis, and flame synthesis, as well as approaches to their chemical and physical modification aimed at enhancing dispersion within polymer matrices and strengthening interfacial adhesion. A detailed examination is presented on the structural features of the nanofillers, such as the CNT aspect ratio, graphene oxide modification, and the formation of hybrid 3D networks and processing techniques, which enable the targeted control of the nanocomposite’s electrical conductivity, mechanical strength, and flexibility. Central focus is placed on the fundamental mechanisms of the piezoresistive response, analyzing the role of percolation thresholds, quantum tunneling effects, and the reconfiguration of conductive networks under mechanical load. The review summarizes the latest advancements in flexible and stretchable sensors capable of detecting both micro- and macro-strains for structural health monitoring, highlighting the achieved improvements in sensitivity, operational range, and durability of the composites. Ultimately, this analysis clarifies the interrelationship between nanofiller structure (CNTs and graphene), processing conditions, and sensor functionality, highlighting key avenues for future innovation in smart materials and wearable devices. Full article

The accurate prediction of human movement direction plays a crucial role in fields such as rehabilitation monitoring, sports science, and intelligent military systems. Based on plantar pressure and inertial sensor data, this study developed a hybrid deep learning model integrating a Convolutional Neural Network (CNN) and a Bidirectional Long Short-Term Memory (BiLSTM) network to enable joint spatiotemporal feature learning. Systematic comparative experiments involving four distinct deep learning models—CNN, BiLSTM, CNN-LSTM, and CNN-BiLSTM—were conducted to evaluate their convergence performance and prediction accuracy comprehensively. Results show that the CNN-BiLSTM model outperforms the other three models, achieving the lowest RMSE (0.26) and MAE (0.14) on the test set, with an R² of 0.86, which indicates superior fitting accuracy and generalization ability. The superior performance of the CNN-BiLSTM model is attributed to its ability to effectively capture local spatial features via CNN and model bidirectional temporal dependencies via BiLSTM, thus demonstrating strong adaptability for complex motion scenarios. This work focuses on the optimization and comparison of deep learning algorithms for spatiotemporal feature extraction, providing a reliable framework for real-time human motion prediction and offering potential applications in intelligent gait analysis, wearable monitoring, and adaptive human–machine interaction. Full article

In the context of the rapid growth of the Internet of Things and mobile health services, sensors and smart wearable devices are continuously collecting and uploading dynamic health data. Together with the long-term accumulated electronic medical records and multi-source heterogeneous clinical data from healthcare institutions, these data form the cornerstone of intelligent healthcare. In the context of medical data sharing, previous studies have mainly focused on privacy protection and secure data transmission, while relatively few have addressed the issue of incentive mechanisms. However, relying solely on technical means is insufficient to solve the problem of individuals’ willingness to share their data. To address this challenge, this paper proposes a three-party Stackelberg-game-based incentive mechanism for medical data sharing. The mechanism captures the hierarchical interactions among the intermediator, electronic device users, and data consumers. In this framework, the intermediator acts as the leader, setting the transaction fee; electronic device users serve as the first-level followers, determining the data price; and data consumers function as the second-level followers, deciding on the purchase volume. A social network externality is incorporated into the model to reflect the diffusion effect of data demand, and the optimal strategies and system equilibrium are derived through backward induction. Theoretical analysis and numerical experiments demonstrate that the proposed mechanism effectively enhances users’ willingness to share data and improves the overall system utility, achieving a balanced benefit among the cloud platform, electronic device users, and data consumers. This study not only enriches the game-theoretic modeling approaches to medical data sharing but also provides practical insights for designing incentive mechanisms in IoT-based healthcare systems. Full article

Human Activity Recognition (HAR) using wearable sensors enables impactful applications in healthcare, fitness, and smart environments, but it also faces challenges related to data privacy, non-independent and identically distributed (non-IID) data, and limited computational resources on edge devices. This study proposes an efficient and personalized federated learning (PFL) framework for HAR that integrates federated training with model compression and per-client fine-tuning to address these challenges and support deployment on resource-constrained devices (RCDs). A convolutional neural network (CNN) is trained across multiple clients using FedAvg, followed by magnitude-based pruning and float16 quantization to reduce model size. While personalization and compression have previously been studied independently, their combined application for HAR remains underexplored in federated settings. Experimental results show that the global FedAvg model experiences performance degradation under non-IID conditions, which is further amplified after pruning, whereas per-client personalization substantially improves performance by adapting the model to individual user patterns. To ensure realistic evaluation, experiments are conducted using both random and temporal data splits, with the latter mitigating temporal leakage in time-series data. Personalization consistently improves performance under both settings, while quantization reduces the model footprint by approximately 50%, enabling deployment on wearable and IoT devices. Statistical analysis using paired significance tests confirms the robustness of the observed performance gains. Overall, this work demonstrates that combining lightweight model compression with personalization providing an effective and practical solution for federated HAR, balancing accuracy, efficiency, and deployment feasibility in real-world scenarios. Full article

Gait recognition using wearable sensor data is crucial for healthcare, rehabilitation, and monitoring neurological and musculoskeletal disorders. This study proposes a deep learning framework for gait classification using inertial measurements from four body-mounted IMU sensors (head, lower back, and both feet). The data were collected from a publicly available, clinically annotated dataset comprising 1356 gait trials from 260 individuals with diverse pathologies. The framework, G-MASA-TCN (Gait Multi-Anchor, Space-Aware Temporal Convolutional Network), integrates multi-scale temporal fusion, graph-informed spatial modeling, and residual dilated convolutions to extract discriminative gait signatures. To ensure both high performance and interpretability, Integrated Gradients is incorporated as an explainable AI (XAI) method, providing sensor-level and temporal attributes that reveal the features driving model decisions. The framework is evaluated via repeated cross-validation experiments, reporting detailed metrics with cross-run statistical analysis (mean ± standard deviation) to assess robustness. Results show that G-MASA-TCN achieves 98% classification accuracy for neurological, orthopedic, and healthy cohorts, demonstrating superior stability and resilience compared to baseline architectures, including Gated Recurrent Unit (GRU), Transformer neural networks, and standard TCNs, and 98.4% accuracy in identifying individual subjects based on gait. Furthermore, the model offers clinically meaningful insights into which sensors and gait phases contribute most to its predictions. This work presents an accurate, interpretable, and reliable tool for gait pathology recognition, with potential for translation to real-world clinical settings. Full article

Wearable antennas for wireless sensor network (WSN) applications require circularly polarized (CP) radiation to maintain stable communication link under human body movement and complex environments. However, many existing wearable CP antennas rely on either linearly polarized (LP) or CP radiator with a single axial ratio (AR) mode combined with external polarization conversion structures, which limit the achievable axial ratio bandwidth (ARBW). In this work, an all-textile wideband CP antenna with a square-ring notched slot radiator, a 50 Ω microstrip line, and a 3 × 3 nonuniform metasurface (MTS) is proposed for 5.85 GHz WSN applications. Unlike conventional CP generation approaches, the square-ring notched slot, analyzed using characteristic mode analysis (CMA), directly excites three distinct AR modes, enabling potential wideband CP radiation. The nonuniform MTS further improves IBW performance by exciting additional surface wave resonances. Moreover, the nonuniform MTS further enhances ARBW by redirecting the incident wave into an orthogonal direction with equivalent amplitude and a 90° phase difference at higher frequency region. The proposed antenna is composed of conductive textile and felt substrates, offering flexibility for wearable applications. The proposed antenna is measured in free space, on human bodies, and fresh pork in an anechoic chamber. The measured results show a broad IBW and ARBW of 84.52% and 43.56%, respectively. The measured gain and radiation efficiency are 4.47 dBic and 68%, respectively. The simulated specific absorption rates (SARs) satisfy both US and EU standards. Full article

Cyber–physical systems (CPS) in healthcare represent a deep integration of computational intelligence, physical medical devices, and human-centric data, enabling continuous, adaptive, and personalized care. These systems combine real-time measurements, artificial intelligence (AI)-based analytics, and networked medical devices to monitor, predict, and optimize patient health outcomes. A key development in the field of CPS is the emergence of patient digital twins (DTs), virtual models of individual patients that simulate biological, behavioral, and social parameters. Using AI, DTs analyze complex medical and social data (genetics, lifestyle, environment, etc.) to support precise diagnosis and treatment planning. The implications of the bibliometric findings suggest that the field emerges from the conceptual phase, justifying the article’s emphasis on both the proposed architectures and their clinical validation. However, most research was conducted in computer science, engineering, and mathematics, rather than medicine and healthcare, suggesting an early stage of technological maturity. Leading countries were India, the United States, and China, but these countries did not have a high number of publications, nor did they record leading researchers or affiliations, suggesting significant research fragmentation. The most frequently observed Sustainable Development Goals indicate an industrial context. Reflecting insights from medical and social research, AI-based DT systems provide a holistic view of the patient, taking into account not only physiological states but also psychological and social well-being. These systems promote personalized therapy by dynamically adapting treatment based on real-time feedback from wearable sensors and electronic medical records. More broadly, CPS and DT systems increase healthcare system efficiency by reducing hospitalizations and supporting remote preventive care. Their implementation poses significant ethical and privacy challenges, particularly regarding data ownership, algorithm transparency, and patient autonomy. Full article

Accurate recognition of lower limb activities is essential for wearable rehabilitation systems and assistive robotics like exoskeletons and prosthetics. This study introduces SDALLE, a custom hardware data acquisition system that integrates surface electromyography sensors (EMGs) and inertial measurement sensors (IMUs) into a wireless, portable platform for locomotor monitoring. Using this system, data were collected from nine healthy subjects performing four fundamental locomotor activities: walking, jogging, stair ascent, and stair descent. The recorded signals underwent an offline structured preprocessing pipeline consisting of time-series augmentation (jittering and scaling) to increase data diversity, followed by wavelet-based denoising to suppress high-frequency noise and enhance signal quality. A temporal one-dimensional convolutional neural network (1D-TCNN) with three convolutional blocks and fully connected layers was trained on the prepared dataset to classify the four activities. Classification using IMU sensors achieved the highest performance, with accuracies ranging from 0.81 to 0.95. The gyroscope X-axis of the left Rectus Femoris achieved the best performance (0.95), while accelerometer signals also performed strongly, reaching 0.93 for the Vastus Medialis in the Y direction. In contrast, electromyography channels showed lower discriminative capability. These results demonstrate that the combination of SDALLE hardware, appropriate data preprocessing, and a temporal CNN provides an effective offline sensing and activity classification pipeline for lower limb activity recognition and offers an open-source dataset that supports further research in human activity recognition, rehabilitation, and assistive robotics. Full article

Parkinson’s Disease (PD) is a neurodegenerative disorder in which gait abnormalities serve as key indicators of motor impairment and disease progression. Although wearable sensor-based gait analysis has advanced, existing methods still face challenges in modeling multi-sensor spatial relationships, extracting adaptive multi-scale temporal features, and effectively integrating time–frequency information. To address these issues, this paper proposes a multi-sensor gait neural network that integrates biomechanical priors with time–frequency collaborative learning for the automatic assessment of PD gait severity. The framework consists of three core modules: (1) BGS-GAT (Biomechanics-Guided Graph Attention Network), which constructs a sensor graph based on plantar anatomy and explicitly models inter-regional force dependencies via graph attention; (2) AMS-Inception1D (Adaptive Multi-Scale Inception-1D), which employs dilated convolutions and channel attention to extract multi-scale temporal features adaptively; and (3) TF-Branch (Time–Frequency Branch), which applies Real-valued Fast Fourier Transform (RFFT) and frequency-domain convolution to capture rhythmic and high-frequency components, enabling complementary time–frequency representation. Experiments on the PhysioNet multi-channel foot pressure dataset demonstrate that the proposed model achieves 0.930 in accuracy and 0.925 in F1-score for four-class severity classification, outperforming state-of-the-art deep learning models. Full article

This study details the design, fabrication, and experimental validation of a wearable, non-invasive microwave sensor for continuous blood glucose monitoring. It incorporates a crescent-loaded elliptical patch antenna with a complementary split-ring resonator (CSRR) tag unit to greatly improve sensing sensitivity. The sensor operates across multiple resonant frequencies, enabling broadband dielectric characterization of glucose-dependent blood permittivity. Incorporation of the CSRR tag unit leads to a marked improvement in electromagnetic coupling and field confinement, resulting in a substantial increase in sensitivity, achieving 1.14 MHz/mg/dL in resonant frequency shift and 0.015 dB/mg/dL in reflection coefficient sensitivity compared to conventional designs. The sensor was fabricated on an FR-4 substrate and experimentally characterized using a vector network analyzer (VNA), showing strong agreement between simulated and measured S11 responses, with minimal frequency deviations and consistent resonance behavior. Experimental results confirmed improved sensitivity in response to glucose concentration variations over the range of 0–500 mg/dL, validating the sensor’s performance under realistic conditions. Furthermore, a physics-informed deep learning (PI-DL) model was developed to predict glucose concentration directly from measured S11 data. The model achieved enhanced prediction accuracy, with a mean absolute error below 1 mg/dL and a strong generalization across unseen samples, demonstrating the power of combining physical modeling with data-driven approaches. These results confirm that the proposed sensor, enhanced with the CSRR tag unit and supported by a PI-DL framework, offers a promising pathway toward next-generation non-invasive, accurate, and wearable glucose monitoring solutions. Full article

Flexible strain sensors capable of conformal integration with living organisms are essential for advanced wearable electronics, human–machine interaction, and plant health. However, many existing sensors require complex fabrication or rely on non-breathable elastomer substrates that interfere with the physiological microenvironment of skin or plant tissues. Here, we present a low-cost, breathable, and highly stretchable strain sensor constructed from biomedical materials, in which a double-layer medical elastic bandage serves as the porous substrate and an intermediate conductive medical elastic tape impregnated with carbon nanotubes (CNTs) ink acts as the sensing layer. Owing to the hierarchical textile porosity and the deformable CNTs percolation network, the sensor achieves a wide strain range of 100%, a gauge factor of up to 2.72, and excellent nonlinear second-order fitting (R² = 0.997). The bandage substrate provides superior air permeability, allowing long-term attachment without obstructing moisture and gas exchange, which is particularly important for maintaining skin comfort and preventing disturbances to plant epidermal physiology. Demonstrations in human joint-motion monitoring and real-time plant growth detection highlight the device’s versatility and biological compatibility. This work offers a simple, low-cost yet effective alternative to sophisticated strain sensors designed for human monitoring and plant growth monitoring, providing a scalable route toward multifunctional wearable sensing platforms. Full article

The progressive nature of dementia necessitates early detection strategies capable of identifying preclinical cognitive decline. Gait disturbances, mediated by higher-order cognitive functions, have emerged as potential digital biomarkers in this context. This bibliometric review systematically maps the scientific output from 2010 to 2025 on the application of wearable sensor technologies and gait analysis in the early diagnosis of dementia. A targeted search of the Scopus database yielded 126 peer-reviewed studies, which were analyzed using VOSviewer for performance metrics, co-authorship networks, bibliographic coupling, co-citation, and keyword co-occurrence. The findings delineate a multidisciplinary research landscape, with major contributions spanning neurology, geriatrics, biomedical engineering, and computational sciences. Four principal thematic clusters were identified: (1) Cognitive and Clinical Aspects of Dementia, (2) Physical Activity and Mobility in Older Adults, (3) Technological and Analytical Approaches to Gait and Frailty and (4) Aging, Cognitive Decline, and Emerging Technologies. Despite the proliferation of research, significant gaps persist in longitudinal validation, methodological standardization, and integration into clinical workflows. This review emphasizes the potential of sensor-derived gait metrics to augment early diagnostic protocols and advocates for interdisciplinary collaboration to advance scalable, non-invasive diagnostic solutions for neurodegenerative diseases. Full article

Manual material handling is one of the leading causes of work-related low-back disorders, and an accurate assessment of the biomechanical risk is essential to support prevention strategies. Despite workers’ interest in wearable sensor networks for quantifying exposure metrics, these systems still present several limitations, including potential interference with natural movements and workplaces, and concerns about durability and cost-effectiveness. For these reasons, alternative motion capture methods are being explored. Among them, completely markerless (ML) technologies are being increasingly applied in ergonomics. This study aimed to compare a wearable sensor network and an ML system in the evaluation of lifting tasks, focusing on the variables and multipliers used to compute the recommended weight limit (RWL) and the lifting index (LI) according to the revised NIOSH lifting equation. We hypothesized that ML systems equipped with multiple cameras may provide reliable and consistent estimations of these kinematic variables, thereby improving risk assessments. We also assumed that these ML approaches could represent valuable input for training AI algorithms capable of automatically classifying the biomechanical risk level. Twenty-eight workers performed standardized lifts under three risk conditions. The results showed significant differences between wearable sensor networks and ML systems for most measures, except at a low risk (LI = 1). Nevertheless, ML consistently showed a closer agreement with reference benchmarks and a lower variability. In terms of the automatic classification performance, ML–based kinematic variables yielded accuracy levels comparable to those obtained with the wearable system. These findings highlight the potential of ML approaches to deliver accurate, repeatable, and cost-effective biomechanical risk assessments, particularly in demanding lifting tasks. Full article

The development of Micro-electro-magnetic Vibration Energy Harvesters (MEMVEHs) plays a crucial role in advancing self-powered nanophotonic, nanoelectronic, and nanosensor systems. As energy autonomy becomes critical for miniaturized devices, MEMVEHs offer a sustainable power source for low-power nanodevices operating in wireless sensor networks, wearable electronics, and biomedical implants. This study provides a comparative assessment of MEMVEH technologies and evaluates their integration potential within next-generation nanoscale systems, enabling enhanced performance, longevity, and energy efficiency of emerging nanotechnologies. Electromagnetic vibration energy harvesters (EMEHs) based on microelectromechanical system (MEMS) technology are promising solutions for powering small-scale, autonomous electronic devices. In this study, two electromagnetic vibration energy harvesters based on microelectromechanical (MEMS) technology are presented. Two models with distinct vibration structures were designed and fabricated. A permanent magnet is connected to a silicon vibration structure (resonator) and a tiny wire-wound coil as part of the energy harvester. The coil has a total volume of roughly 0.8 cm³. Two energy harvesters with various resonators are tested and compared. Model A’s maximum load voltage is 163 mV, whereas Model B’s is 208 mV. A maximum load power of 59.52 μW was produced by Model A at 347 Hz across a 405 Ω load. At 311.4 Hz, Model B produced a maximum load power of 149.13 μW while accelerating by 0.4 g. Model B features a larger working bandwidth and a higher output voltage than Model A. Model B performs better than Model A in comparable experimental settings. Simple study revealed that Model B’s electromagnetic energy harvesting produced superior outcomes. Additionally, it indicates that a nonlinear spring may be able to raise the output voltage and widen the frequency bandwidth. Full article