Search Results (611)

Background/Objectives: Sleep disorders are heterogeneous conditions with diverse neural, muscular, and ocular manifestations, making polysomnography (PSG) the gold standard for accurate diagnosis. Artificial intelligence-based approaches, particularly deep learning (DL) models capable of integrating heterogeneous information, offer a promising solution for reliable decision-making in such clinical scenarios. However, most existing DL studies have focused on a single disorder, relied on limited datasets, or employed epoch-level labeling strategies that overlook the episodic nature of sleep pathophysiology, thereby limiting clinical applicability. To address these gaps, we propose a novel multimodal attention-enhanced fusion network (MAF-SleepNet) for automatic multi-class sleep disorder classification based on the International Classification of Sleep Disorders. Methods: MAF-SleepNet jointly processes electroencephalography (EEG), electrooculography (EOG), and leg electromyography (EMG) signals through modality-specific feature extraction and adaptive attention mechanisms, capturing both intra- and inter-modality dependencies. The model was evaluated on a combined dataset of 141 recordings from three public databases, including five PSG-requiring disorders and a healthy class. Results: Experimental results demonstrated that MAF-SleepNet achieved 86.07 ± 3.66% accuracy and 82.67 ± 4.46% macro-F1 under a strict subject-independent cross-validation, and 98.96 ± 0.66% accuracy and 98.93 ± 0.60% macro-F1 under subject-dependent cross-validation. Conclusions: These results demonstrate that the proposed approach provides a more reliable and clinically meaningful assessment compared to many existing studies that rely on subject-dependent evaluation or epoch-level labeling. The findings highlight the effectiveness of adaptive multimodal fusion for robust and clinically relevant sleep disorder classification. Future work should investigate the integration of respiratory and autonomic modalities and validation on larger multi-center cohorts. Full article

Reliable monitoring of uterine contractility underpins the assessment of labor, the diagnosis of preterm labor, and the timing of obstetric intervention, yet routine methods measure only the mechanical consequences of contraction. External tocodynamometry and cardiotocography (CTG) are operator- and position-dependent and perform poorly with maternal obesity, detecting as few as ~54% of the contractions confirmed by an intrauterine pressure catheter, whereas surface electrohysterography (EHG) detects upward of ~94% by recording the myometrial electrical activity that drives contraction. This review examines EHG and uterine electromyography (EMG) as measurement modalities, covering their physiological origin, acquisition hardware, signal characteristics, feature extraction, and machine-learning analysis, and compares them with CTG against the intrauterine pressure catheter reference standard. Electrical approaches additionally yield predictive parameters, notably spectral peak frequency and propagation velocity, that mechanical methods cannot provide. The principal barrier to translation is methodological heterogeneity rather than physiology: differences in electrodes, filtering, feature definitions, and outcome measures preclude cross-study comparison and meta-analysis. As its central contribution, this review consolidates prior calls for standardization into a minimum reporting set for EHG studies and appraises translational readiness, identifying prospective external validation, shared datasets, explainable models, and outcome-linked trials as priorities. Full article

Reliable hand gesture recognition (HGR) using wearable sensors remains challenging due to variability in arm posture, motion, and individual muscle activation patterns. To facilitate systematic investigation of multi-modal sensing strategies under realistic operating conditions, this paper presents a comprehensive dataset of surface electromyography (sEMG) and pressure-based force myography (pFMG) signals. The dataset includes three complementary subsets acquired under controlled static arm posture, multiple static arm postures, and combined static and dynamic arm postures. Signals were recorded using a custom-designed co-located sEMG-pFMG armband, enabling the simultaneous capture of electrical muscle activation and mechanical muscle deformation. System validation is conducted from three perspectives. First, hardware-level signal quality is assessed through signal-to-noise ratio (SNR) analysis across all gestures and sensing channels, demonstrating stable and reliable signal acquisition. Second, representative raw waveform examples are provided to qualitatively illustrate modality-specific and condition-dependent signal characteristics under static and dynamic arm-posture scenarios. Third, reproducible baseline gesture recognition experiments are performed using conventional machine learning classifiers. By providing multi-modal data acquired under both static and dynamic arm-posture conditions, along with clearly de-fined experimental protocols and baseline benchmarks, this dataset serves as a valuable resource for developing, evaluating, and comparing gesture recognition algorithms and arm-wearable human–machine interface (HMI) systems. Full article

Diabetic peripheral neuropathy (DPN) causes progressive sensorimotor dysfunction of the lower extremities, impairing mobility and increasing fall risk. Phase-specific neuromuscular activation during repeated sit-to-stand (STS) transfers in DPN remains poorly characterized. This study compared lower-limb muscle activation during repeated STS transfers in fifteen ambulatory DPN patients and fifteen age- and gender-matched healthy controls. Bilateral surface electromyography (sEMG) was recorded from the vastus lateralis, vastus medialis, biceps femoris, gluteus maximus, and gluteus medius; normalized to maximum voluntary contractions (MVCs), and analyzed separately for STS and stand-to-sit phases. No significant between-group differences were observed for any muscle during either phase (all p > 0.05). Effect sizes were small-to-medium, with confidence intervals indicating uncertainty around the estimated between-group differences. Notably, gluteus maximus activation during STS demonstrated the largest between-group effect size, favoring the DPN group (d = 0.54), although this difference was not statistically significant. This finding may indicate a possible proximal hip-extensor strategy that requires confirmation in larger studies. Both groups demonstrated significantly greater vastus lateralis and gluteus maximus activation during STS than stand-to-sit, reflecting greater neuromuscular demand during rising. Overall, the non-significant between-group findings, together with the effect-size and confidence-interval analysis, suggest that MVC-normalized EMG amplitude alone may have limited sensitivity for detecting neuropathic neuromuscular alterations during transitional tasks in ambulatory patients with DPN. Full article

Background/Objectives: This study examined whether foot progression angle and stance width modify lower-limb muscle activation during the holding phase of forward lunges in healthy adults. Methods: Thirty-six healthy adults completed five randomized within-participant lunge conditions: standard stance with 0°, 30°, and 60° foot progression angles and narrow and wide stances with a 0° foot progression angle. Surface electromyographic activity was recorded from the rectus femoris, gluteus medius, vastus lateralis, vastus medialis, biceps femoris, and semitendinosus. The central 2 s of each 6 s trial were analyzed and normalized to maximal voluntary contraction. Data were analyzed using repeated-measures analysis of variance with Bonferroni-adjusted pairwise comparisons. Results: Foot progression angle significantly affected rectus femoris, gluteus medius, vastus lateralis, and vastus medialis activation. Rectus femoris and vastus lateralis activation were higher at 60° than at 0° and 30°, vastus medialis activation was higher at 60° than at 30° only, and gluteus medius activation decreased as foot progression angle increased. Stance width significantly affected rectus femoris, gluteus medius, vastus lateralis, and vastus medialis activation. Rectus femoris activation increased with stance width, gluteus medius activation was lower in standard and wide stances than in narrow stance, and vastus lateralis and vastus medialis activation were higher in standard and wide stances than in narrow stance. Biceps femoris and semitendinosus did not differ across conditions. Conclusions: Foot progression angle and stance width modified selected acute lower-limb EMG responses during the holding phase of bodyweight forward lunges. Full article

Background/Objectives: Lower-limb exoskeletons require reliable movement recognition mechanisms to support adaptive locomotor assistance and rehabilitation. Electromyographic (EMG) signals provide valuable information on muscle activation and user intention, enabling safe and responsive human–exoskeleton interaction. This study aims to develop and experimentally validate an EMG-based platform for intelligent lower-limb movement recognition and locomotor assistance applications. Methods: The proposed platform integrates multichannel EMG acquisition, embedded signal processing, and artificial intelligence for movement classification. EMG signals associated with six movement classes (left/right kneeling, stepping, and dash) were acquired from ten healthy male participants aged 19–24 years. Signal preprocessing, normalization, dataset generation, and model training were performed using a dedicated processing framework. Continuous EMG acquisition without threshold-based segmentation was employed to preserve complete neuromuscular information and improve dataset consistency. Movement classification was implemented using a lightweight one-dimensional convolutional neural network (1D-CNN). Model performance was evaluated using Stratified 5-Fold Cross-Validation and Leave-One-Subject-Out (LOSO) protocols. Results: A dataset containing 608 multichannel EMG recordings was generated for training and validation. The proposed 1D-CNN model achieved an accuracy of 92.43 ± 1.69% and a macro F1-score of 0.9093 ± 0.0247 under Stratified 5-Fold Cross-Validation. LOSO evaluation yielded an accuracy of 62.11 ± 23.26%, highlighting the significant impact of inter-subject variability on classification performance. Conclusions: The developed platform provides an effective framework for EMG-based lower-limb movement recognition in intelligent exoskeleton systems. The results demonstrate the feasibility of integrating multichannel EMG sensing and AI-based inference into adaptive locomotor assistance systems while emphasizing the importance of improving subject-independent generalization. The proposed platform also establishes a foundation for future research on multimodal sensing and real-time adaptive exoskeleton control. Full article

Parkinson’s disease (PD) is a neurodegenerative disorder affecting motor and autonomic nervous system functions. In this study, six synchronized modalities—electroencephalography (EEG), electrocardiography (ECG), electromyography (EMG), respiration (Resp), photoplethysmography (PPG), and gait (Gait)—were recorded from 25 PD patients and 25 healthy controls. A Random Forest classifier was used to perform both unimodal and multimodal signal classification. Among unimodal models, ECG achieved the highest accuracy (84%), whereas the performance of multimodal combinations did not increase linearly with the number of modalities; integrating three or more complementary signals was sufficient to substantially improve classification. The full six-modality model achieved an accuracy of 95.00%, precision of 94.17%, recall of 97.14%, F1 score of 95.21%, and an AUC of 0.98. Incremental analysis further indicated that selecting key complementary modalities can maintain high classification performance while reducing equipment requirements, simplifying experimental procedures, and improving participant comfort, providing guidance for the development of efficient, non-invasive PD diagnostic tools. Full article

This study presents the development of a proof of concept, functional hydrogel interface designed for the acquisition of physiological signals, such as electrocardiogram (ECG) and electromyography (EMG). The hydrogel is synthesized using chitosan extracted from the shells of Tenebrio molitor larvae through a sustainable acid–alkaline protocol, blended with glycerol, polyvinyl alcohol (PVA), and ionized with NaCl to enhance conductivity. The resulting hydrogel membranes were cast and cut into circular shapes to provide a uniform contact geometry. The fabrication process yielded flexible membranes exhibiting ionic conductivity and partial surface conformity and handling stability. The extracted chitosan was characterized by Fourier-transform infrared spectroscopy (FTIR), degree of deacetylation (DDA), and molecular weight determination. Mechanical characterization included compression and tensile testing, while electrical characterization was performed through impedance spectroscopy and comparison with a commercial hydrogel interface. Functional evaluation was conducted through ECG and EMG signal acquisition under controlled experimental conditions. Preliminary in situ ECG and EMG recordings demonstrated successful signal acquisition using the proposed hydrogel interface. Future work may further investigate the mechanical and electrical behavior of the hydrogel under broader experimental conditions, as well as the optimization of the hydrogel formulation and extended physiological signal acquisition. Studies may help further characterize its potential as a chitosan-based bio interface material for bioelectrical sensing applications. Full article

Electromyography (EMG)-controlled orthoses and wearable assistive systems require classifiers that combine accurate motion recognition, efficient embedded deployment, and support for local model formation. Long-term EMG deployment is affected by signal drift and inter-session variability, motivating architectures that can support post-deployment model updates. Local processing can reduce dependence on external computation and data transfer. To address these challenges, this work presents a hardware-targeted Hyperdimensional Computing (HDC) classifier for trainable EMG classification on the resource-constrained GateMate A1 FPGA from Cologne Chip. The proposed architecture performs on-device HDC model formation and inference directly on FPGA. Linear Discriminant Analysis (LDA) serves as a conventional offline-trained and FPGA-inferred baseline. The evaluation uses eight anonymized single-session EMG recordings from seven healthy participants and one participant with spinal cord injury with 32 channels and five motion classes and includes accuracy, robustness, and routed FPGA implementation analysis. Across all recordings, the per-recording best-case HDC configurations reach 95.20% classification accuracy, while the LDA baseline achieves 98.54% overall accuracy. Under controlled input perturbation with a standard deviation of $\sigma =0.10$, HDC retains 92.98% mean accuracy compared with 80.98% for LDA. A first board-level power measurement indicates an energy cost of approximately 0.40 mJ per inference sample and 0.87 mJ per training sample. The experiments demonstrate single-session on-device HDC model formation and inference, while longitudinal validation, fatigue robustness, electrode-shift robustness, and inter-session adaptation remain future work. The results indicate that HDC provides an architectural foundation for trainable wearable edge–AI systems with local model updates. Full article

Objective fitting measures offer a means to circumvent the subjectivity of cochlear implant programming, with the stapedius reflex representing one robust predictor of the maximum comfortable loudness level. With the present study, it was investigated whether long-term electromyographic measurements of the stapedius muscle using implanted electrodes are feasible. In nine sheep, myoelectrical activities were recorded intraoperatively and synchronized with middle ear admittance as a reference signal. For acoustic stimulation pure tones with different frequencies were used. The electrodes were placed at the stapedius muscle surface after exposing it via the retrofacial approach. EMG-based detection of the stapedius reflex was achievable over six months when electrode integrity and placement were preserved. The treated muscles were subsequently excised, cut and examined histologically. No signs of atrophy were found in the muscles examined. However, the histological section series showed a clear division of the muscle from proximal to distal, the ratio between tendon and muscle fibers being most pronounced in favor of the muscle fibers in the proximal section. The integration of an electromyography-based measurement method for the objective determination of the stapedius reflex threshold and thus, for the long-term adjustment of cochlear implants, appears possible and could potentially enable autonomous fitting of implants. Full article

Self-powered monitoring systems capable of scavenging ambient mechanical energy are a highly desirable solution to eliminate the reliance on batteries and grid power in remote and distributed atmospheric sensing networks. However, the widespread adoption of such systems is severely hindered by the insufficient output power density of current energy harvesters, which struggle to simultaneously drive environmental sensors, data acquisition units, and wireless transmission modules. In this work, we report a highly integrated hybrid power generation system that couples a triboelectric nanogenerator (TENG) and an electromagnetic generator (EMG) to efficiently harvest low-frequency mechanical energy from the surroundings. Through systematic structural optimization and synergistic matching of the two transduction mechanisms, the device achieves an outstanding volumetric power density of 129.9 W·m⁻³, which represents one of the highest values ever reported for hybrid nanogenerators targeting self-powered environmental applications. The output characteristics of both the TENG and EMG units under varying load impedances are thoroughly characterized, revealing the optimal operating points for maximum power extraction. A tailored power management module, consisting of rectification, energy storage, and regulation circuits, is designed to convert the irregular alternating output into a stable direct-current supply. To demonstrate the practical viability of the system, we construct a complete self-powered atmospheric environment monitoring node, which integrates multiple environmental sensors, a data acquisition module, and a wireless transmission module. Driven exclusively by the hybrid TENG–EMG generator under ambient mechanical excitation, the node successfully performs real-time sensing, signal processing, and remote data communication without any external power input. This work not only provides a record-high power density among hybrid generators for environmental monitoring, but also establishes a feasible pathway toward maintenance-free, widely distributed, and truly autonomous atmospheric sensing networks. The presented strategy of maximizing volumetric power density through hybrid design and impedance engineering can be readily extended to other self-powered systems. Full article

The high rate of reinjury after anterior cruciate ligament reconstruction (ACLR), persistent sex-related disparity, and increased susceptibility during transitional stages in female football players highlight the need for more specific strategies to identify biomechanical parameters associated with valgus collapse. This study aimed to compare three-dimensional knee kinematic and synchronized electromyography (EMG) muscle activity of the medial and lateral thigh muscle complexes during a change of direction (COD) task, between ACLR and healthy controls players. A cross-sectional case–control study was conducted with 26 under-23 semiprofessional female football players (22.89 ± 2.68 years), divided into ACLR (n = 13) and control (n = 13) groups. The maximum and minimum peaks and range of knee angular velocity across three planes, along with the average and peak electromyography (EMG) muscle activity of the Biceps Femoris (BF), Semitendinosus (ST), Vastus Medialis (VM), and Vastus Lateralis (VL), were recorded during the preparation and load phases. Between-group differences were assessed using independent t-tests or Mann–Whitney U tests. Statistical significance after Holm–Bonferroni correction was established at p-Holm < 0.05. ACLR players demonstrated significant increased knee valgus angular velocity, alongside 4% reduced average ST muscle activity and 27% diminished peak BF muscle activity during the load phase, compared to controls. These findings indicate altered knee kinematic and muscle activity patterns during COD in ACLR players, suggesting persistent long-term functional adaptations in female football players after ACLR. Full article

Background: We present a miniature (30 × 34 mm) 64-channel data acquisition headstage optimized for high-density surface electromyography. Methods: The headstage is made up of a multi-channel ASIC analogue front-end utilizing only MOS transistors, fabricated in 350 nm CMOS technology (IC die dimensions 6.9 × 1.8 mm), combined with an off-the-shelf multi-channel current-input ADC (DDC264, Texas Instruments). The ASIC analogue front-end employs MOS-based capacitors for both processing and AC-coupling. Results: The combination of these two sub-circuits enables the simultaneous recording of 64 channels at a typical sampling rate of 4 KHz with a maximum analogue bandwidth of 0.5–1500 Hz and a resolution of 20-bits. Typical input-referred-noise, determined by the analogue front-end, is 3.5 μV_RMS for a surface EMG bandwidth of interest of 20–500 Hz. This two-chip solution results in a power consumption of 5 mW per channel. Analogue performance variability of the custom ASIC was characterized across a dataset of 960-channels (15 dies) from two fabrication runs. Conclusions: This work practically demonstrates the viability of using both a MOS-only analogue front-end and commercially available off-shelf high-performance back-end hardware already developed for medical imaging applications to record high-density surface biosignals. The aforementioned techniques can be employed to reduce the size and cost for systems or wearable devices; facilitating the translation of high-density bio-acquisition setups from the research environment to more affordable commercial products. Full article

Transcutaneous spinal cord stimulation (tSCS) is a promising neuromodulation approach for motor recovery after spinal cord injury (SCI), yet clinical programming remains largely dependent on subjective parameter selection. This study evaluated high-density surface EMG (HD-sEMG)–derived spatial and temporal features as objective biomarkers for tSCS optimization in three adults with chronic cervical SCI. A 64-channel electrode array recorded stimulation-evoked responses across five cervical stimulation levels, four pulse widths, and graded amplitudes. Features describing activation magnitude, spatial distribution, cluster morphology, and temporal dynamics were extracted from epoch-based activation maps. Of the three enrolled participants, two demonstrated measurable stimulation-evoked responses and contributed to the paired-pulse analyses, whereas pulse-width analyses were limited to a single responsive muscle (left flexor carpi) in one participant. Paired-pulse analysis identified root mean square (RMS) as the most discriminative feature, revealing nonlinear, muscle- and level-specific dose–response relationships in which maximal suppression often occurred at intermediate rather than maximal amplitudes. Increasing pulse width expanded the spatial extent of recruitment (active area: p = 0.006; convex hull area: p = 0.004) without altering response timing. Polarity reversal analysis demonstrated stable innervation zone localization across stimulation levels and amplitudes. These findings establish a spatially resolved HD-sEMG framework that may support individualized tSCS parameter selection in SCI. Full article

Background/Objectives: Minimally invasive surgery benefits patients but places physical and cognitive demands on surgeons. While robot-assisted laparoscopic surgery (RALS) reduces musculoskeletal strain, it may increase cognitive load. This study examined whether physical and psychological preparatory protocols (warm-ups) influence surgeon strain during RALS. Methods: Ten consultant surgeons from East Lancashire Hospitals NHS Trust (UK) participated in a preregistered, randomized study. Each performed RALS under three conditions: control, physical warm-up (10 min simulation tasks on the Da Vinci system), and psychological warm-up (10 min PETTLEP-based mental imagery). Electromyography (EMG) and electroencephalography (EEG) were recorded during key surgical phases. EMG data were normalized to maximal voluntary contractions. Results: The physical warm-up significantly increased EMG activity in the right deltoid and right trapezius (p < 0.05) compared to control, with no differences observed in other muscle groups. EEG alpha power data did not significantly differ between conditions. Conclusions: These findings suggest that brief physical warm-up can enhance muscle activation in key regions involved in RALS, potentially improving motor control and reducing fatigue. Incorporating such strategies may support surgeon performance and well-being. Full article