To integrate MR-based servo navigation in MPRAGE and 3D-TSE sequences and demonstrate its potential for prospective head motion correction in structural imaging. Repeated modules of servo navigators were inserted before each preparation pulse of MPRAGE and 3D-TSE sequences (before-prep) for rapid convergence of motion parameter estimation. Additionally, reference navigators were distributed along the imaging echo train to test a within-echo train correction (within-ET) with contrast-matched models, enabling more frequent geometry updates. The methods were evaluated in instructed motion experiments and in high-resolution scans at 7T. Artifacts caused by instructed abrupt motion were clearly reduced in scans with servo navigation for both sequences using the before-prep correction. Residual artifacts of rapid in-train motion were further mitigated by the within-ET correction in MPRAGE. In high-resolution scans with minimal motion, servo navigation preserved fine anatomical details, while it improved image quality under involuntary motion in long acquisitions. Servo navigation provides accurate real-time head motion correction for structural brain imaging.
Robotic endoscope holders improve visual stability, yet enforcing a fixed remote center of motion (RCM) often induces lateral interaction forces due to fulcrum drift. This study proposes an autonomous guidance framework combining deep learning-based visual servoing with force-based pivoting to track surgical instruments while minimizing interaction forces, without a fixed geometric RCM. To test this, a UR3e manipulator with an integrated force/torque sensor controlled a laparoscope in a pelvitrainer. Instruments were localized using a fine-tuned YOLOv11n on 3,695 annotated frames. The detected tools' centroids drove a hysteresis-based visual servoing law with jerk-limited planning, while an admittance-based controller generated angular motion from measured forces. Performance was assessed in continuous tracking and step-response tasks, comparing fixed-RCM with the proposed pivoting approach. The localization model achieved 0.91-0.93 precision, 0.83-0.90 mAP, and 28 FPS on a CPU. Visual servoing re-centered targets in 98.8% of transitions (median recovery: 2.60 s). Force-based pivoting significantly outperformed fixed-RCM control, reducing median interaction forces from 2.35 to 0.19 N. Integrating CNN-driven servoing with force-based pivoting enables autonomous guidance that preserves visual stability while substantially reducing interaction forces. By eliminating fixed RCM assumptions, this framework offers a safer, more adaptive alternative for robotic camera assistance.
In industrial robot visual servoing, the accuracy of end-effector pose directly affects the feedback quality and trajectory tracking performance of the visual servo system. To improve end-effector pose estimation accuracy, this paper establishes a propagation model from marker measurement errors to end-effector pose estimation errors and further derives the covariance expression of pose estimation errors. Based on this model, different marker placement factors affecting translational and rotational errors are analyzed, including marker-set spatial range, spatial distribution balance, and centroid offset distance. In addition, the influence of the number of markers on pose estimation errors is derived by adding a new marker to an existing point set. The accuracy of the analytical model is validated through Monte Carlo simulations and experiments, and guidelines for marker placement and marker number are provided.
To address the challenge of maintenance decision-making for critical components in electro-hydraulic servo material fatigue testing machine, characterized by weak state observability and difficulty in degradation prediction, a multi-component joint maintenance decision-making method based on multi-head deep reinforcement learning is proposed. Considering the heterogeneity of the degradation mechanisms and observation methods for the four components-bearing beam, fixture, main machine sensors, and hydraulic oil tank-a continuous-discrete hybrid state Markov decision process (HS-MDP) is constructed. To account for differences in maintenance strategies across components, a differentiated discrete action space for each component is designed, and engineering feasibility constraints are explicitly integrated into the policy through action masking. A data-quality loss term, determined by the degradation level of the sensors, is introduced into the reward function to align the optimization objective with the metrological properties of the fatigue testing machine. Based on the Branching Dueling DQN framework, a Q-network structure is constructed, incorporating a shared encoder, an inter-component attention mechanism, and multi-head branched outputs. Taking a 100 kN electro-hydraulic servo fatigue testing machine as a case study, comparisons with baseline strategies such as periodic maintenance, threshold-based condition-based maintenance (CBM), independent DQN, and PPO indicate that the proposed method reduces the average annual total cost by 60.3% compared to periodic maintenance and by 42.6% compared to threshold-based CBM. The number of failures decreases from 9.8 times/year to 1.4 times/year, while data efficiency increases from 82.1% to 96.2%. Ablation experiments and robustness tests further verify the critical contributions of three key design elements: action masking, inter-component attention, and data-quality loss.
Aiming at addressing the problem whereby the traditional time-optimal trajectory planning based on the steady-state torque-speed characteristic cannot fully exploit the short-term dynamic output performance of the servo permanent magnet synchronous motor (SPMSM), a time-optimal trajectory planning method for the SPMSM based on the short-term dynamic feasible region constraint is proposed to effectively improve the response speed. Firstly, the dynamic trapezoidal domain operation boundary is obtained by analyzing the motor working point variation curve and considering factors such as the working temperature and trajectory control, which constitutes the torque-speed value and the dynamic constraint mechanism of trajectory planning. Secondly, based on the energy consumption model, the average thermal power is used to represent the torque overload limit condition, and a dynamic constraint method based on the short-term dynamic torque-speed operation boundary is proposed. Then, in order to reduce the computational load in the online millisecond-level response, a time-optimal trajectory optimization algorithm based on sequential least squares is proposed to calibrate the positioning time of the time-optimal trajectory under different working temperatures and angles. Finally, a simulation and experimental comparisons of the time-optimal trajectories under different angles and working temperatures are carried out to verify the effectiveness of the proposed method.
Automation of organoid and cell culture processes is essential for achieving scalable and standardized experimentation in regenerative medicine and stem cell research. However, existing microfluidic platforms often rely on complex setups, limiting their integration within standard incubator environments. To address these challenges, we developed a compact, scalable multi-well platform featuring 3D-printed, servo-actuated disposable microvalves for fully automated media and drug exchange. This design eliminates the need for external pressure sources and control channels, providing a simplified and cost-effective solution for organoid culture. The platform integrates an internet-connected microscopy module with a motorized XYZ stage, allowing continuous, real-time imaging of individual wells directly within the incubator. It supports precise and reliable fluid handling under physiological conditions, improving throughput, reproducibility, and accessibility. We validate the platform through bench-top testing and in both mouse and human organoid models. Morphological analysis, immunohistochemistry (IHC), and qPCR demonstrate comparable viability, growth, and gene expression profiles between automated and manual culture conditions. These results establish a robust and scalable framework for fully automated organoid culture, offering a simplified and accessible alternative to conventional microfluidic systems with broad applications in regenerative medicine, drug discovery, and scalable biological screening.
Metallic elastic elements serve as critical structural components in advanced electro mechanical assemblies, where cyclic loading induces progressive rigidity reduction. Within electro-hydraulic servo valves (EHSV), the feedback mechanism constitutes an essential closed-loop control element whose mechanical property deterioration directly compromises system precision and operational stability. A predictive methodology is introduced for estimating component lifespan under dynamic loading conditions through integrated System Performance Modeling (SPM) and Stiffness Degradation Prediction (SDP) frameworks. Given physical constraints precluding direct rigidity assessment, this approach leverages externally observable parameters for indirect condition monitoring while SPM supplies necessary loading profiles to SDP. The SDP framework integrates continuum damage mechanics principles with machine learning algorithms, with its output continuously refining SPM structural parameters. Through synchronized simulation protocols, this coupled methodology quantifies component rigidity evolution and evaluates system performance metrics to determine remaining operational lifespan. This integrated approach establishes a foundation for predictive maintenance in precision hydraulic control systems.
Adaptive servo-ventilation (ASV) effectively treats central sleep apnea (CSA), with or without coexisting obstructive sleep apnea (CSA-OSA). However, this is a diverse patient group, and the characteristics of patients who are most likely to benefit from ASV therapy, outside of traditional etiology-based subgroups, are unknown. To identify clusters of ASV-treated patients based on anthropometric data and comorbidities, and to evaluate the differential response to ASV in these patient clusters. This analysis used data from the READ-ASV registry, which enrolled adults with CSA prescribed ASV from September 2017-March 2021. Evaluations included the Epworth Sleepiness Scale, Functional Outcomes of Sleep Questionnaire, Pittsburg Sleep Quality Index, EuroQol-5-dimension, and sleep study data. Latent class analysis was used to identify patient clusters based on baseline anthropometric data and comorbidities. Of 812 patients, 421 (52%) were in Cluster (C) 1 (older, high rate of cardiovascular comorbidities [including stroke], high body mass index [BMI]), 239 (29%) were in C2 (elderly/middle-aged, average-low cardiovascular comorbidity rate, high stroke rate, lower BMI, low diabetes rate), and 152 (19%) were in Cluster 3 (younger, low cardiovascular comorbidity rate, higher BMI, high depression rate). Patients from C3 versus C1/C2 had the highest apnea-hypopnea index at baseline (55 vs. 48/39.5 events/h), were the most symptomatic, had the worst quality of life (QoL), and had the greatest symptomatic and QoL improvement on ASV. They remained the most symptomatic cluster after treatment. The identification of subgroups of ASV users with differing responses to therapy and residual symptom burden could facilitate a more personalized approach to ASV prescription and therapy management. https://www. gov; NCT03032029.
Evidence on adherence to adaptive servo-ventilation (ASV), sleep architecture, and patient-reported outcomes in patients with central sleep apnea (CSA) in routine clinical practice is limited. The primary objective was to assess adherence to ASV in a real-world population with CSA; secondary objectives were to evaluate longitudinal changes in sleep fragmentation, polysomnographic sleep architecture, and patient-reported outcomes. The multicenter, prospective, observational AutoSV Registry (NCT03421704, 2018-2023) enrolled adult patients with CSA, who were prescribed ASV (Philips DreamStation BiPAP auto SV). The follow-up period was up to 24 months. Adherence data were collected via telemonitoring software (Philips EncoreAnywhere™). Polysomnography (PSG) was performed at diagnosis, during ineffective continuous positive airway pressure with residual CSA (CPAPrCSA) and on ASV. Sleepiness (Epworth Sleepiness Scale, ESS) and subjective sleep quality (Pittsburgh Sleep Quality Index, PSQI) were measured. Of the 125 enrolled patients, 66% had impaired subjective sleep quality (PSQI>5) and 33% were sleepy (ESS>10). Adherence to ASV therapy was at least 4 hours/night in 67% of patients. Median adherence was 5.3 (3.1; 6.7) hours/day (n = 103, median (IQR) follow-up 16 months). Sleep fragmentation was reduced during ASV, with decreased sleep stage N1 and increased sleep stage N3 and rapid eye movement sleep compared with diagnostic PSG and CPAPrCSA therapy. The global PSQI (mean [95%CI]) score improved by -1.7 (-2.9; -0.6) (p < 0.001; n = 55), and the ESS score (mean [95%CI]) improved by -1.9 (-3.2; -0.6) (p < 0.001; n = 57) from baseline to 24 months. ASV therapy in a sleep-clinic population with CPAPrCSA was associated with high adherence, possibly reflecting clinically meaningful benefits, including reduced sleep fragmentation, improved objective and subjective sleep quality, and reduced sleepiness.
This paper presents a chopper-stabilized capacitively coupled instrumentation amplifier (CS-CCIA) that achieves large electrode DC offset (EDO) cancellation while featuring a short recovery time and GΩ-level input impedance (Zin). Through analyzing the constraints of noise contribution and recovery time on traditional DC-serv oloops (DSL), a feedforward assisted DSL (FFA-DSL) scheme is proposed. By combining a feedforward SAR ADC for coarse EDO cancellation and an analog DSL (ADSL) for residual offset suppression, the FFA-DSL scheme achieves fast recovery from large EDO steps with low noise contribution. A hybrid positive feedback loop (PFL) is proposed to boost the AC Zin and the DC resistance of the CCIA simultaneously. Additionally, a common-mode cancellation loop based on voltage feedback (FB-CMC) is designed to suppress common-mode interference (CMI) without introducing additional capacitive loading at the input. Fabricated in a 0.18µμm CMOS process, the CCIA achieves ±1 Vpp EDO cancellation with a 2.1-s ADSL recovery time after the feedforward path is activated (trec), while delivering a worst-case total recovery time (ttot) of 4.37 s at a 0.5 Hz high-pass corner frequency. It also provides 1-Vpp CMI tolerance, 2 GΩ Zin at 0.1Hz, and 1.8-μVrms input-referred noise within frequency range of 1 Hz to 200 Hz. It consumes 2.16 μW from a 1.2-V supply.
The study reported in this research article focuses on the machining behavior and phase transformation of Ti₅₀Ni₄₉Co₁ shape memory alloy during wire-electrical-discharge-machining (WEDM). Experiments were performed to determine the combined effect of the pulse-on time (Ton) and servo voltage (SV) under high-energy discharge conditions, using a Taguchi L25 orthogonal array. The results show that the pulse-on time significantly increases discharge energy, which reached a peak MRR of 8.43 mm3/min with Ton = 125 µs and SV = 20 V (Run order 21). However, it also increased the surface roughness (Ra) from 2.85 μm to 5.01 μm, producing deeper craters and hardened recast layers as well as re-solidified debris because of rapid melting and solidification. Major morphological changes were observed at high discharge energy, as confirmed by 3D surface profilometry. The servo voltage influenced the stability of the spark: higher voltages reduced discharge intensity, resulting in lower material removal rate but improved surface finish. XRD analysis revealed the formation of secondary phases, containing NiTi2, TiO2, and CuZn, was produced during machining, resulting from oxidation and transfer of the electrode material. DSC examination showed an increase in transformation peaks, a decrease in transformation enthalpy, and an insignificant change in transformation temperatures, indicating that the martensitic transformation in the recast layer was partially suppressed by compositional alterations and oxide formation. Nevertheless, no loss of shape-memory property was observed in the bulk material, although slight variations in transformation behavior were observed. Overall, the results support the importance of properly optimizing pulse-on time and servo voltage to achieve a balance among good machining performance, consistent surface quality, and good functional performance, particularly in precision biomedical, aerospace, and smart actuator applications.
As predictive maintenance transitions from the data-centric paradigm of Industry 4.0 to the sustainable, human-centric framework of Industry 5.0, diagnosing servo motor conditions faces the dual challenges of data scarcity and a profound lack of labeled fault samples. To address this cold-start problem, we present a pseudo-supervised machine learning framework evaluated on a custom five-channel dataset comprising 199 servo motor telemetry samples (current, voltage, temperature, humidity, and vibration). The methodology integrates hard structural partitioning (k-means) and soft posterior confidence estimation (Gaussian Mixture Models) to characterize operating modes without prior annotation. Concurrently, an Isolation Forest model quantifies anomaly intensity and establishes a dynamic quantile-based threshold. A critical innovation of this research is the deterministic risk mapping derived from engineering priors; it defines the "high-risk" (abnormal) state by inversely weighting the physical safety margins of the sensors. This mechanism strictly maps unsupervised clusters to binary pseudo-labels. These labels are subsequently used to supervise downstream discriminators (Random Forest and Support Vector Machine). The final online diagnostic outputs a score-level fusion of the classifier probability and the GMM posterior, gated by the anomaly threshold. Quantitative evaluation demonstrates that the Random Forest model achieved a perfect F1 score of 1.000, while the comparative SVM yielded an F1 score of 0.997, proving the framework to be a robust, interpretable, and highly accurate solution for cold-start industrial health monitoring.
Autonomous mobile robots are used in optimizing warehouse logistics, yet achieving precise positioning during docking maneuvers and autonomous planning remains a technical challenge. This study presents a custom vision-based control system designed for an autonomous omnidirectional wheeled robot. The proposed methodology acquires visual feedback using a stereo camera integrated within the Robot Operating System framework. Two visual feedback control laws are formulated and rigorously evaluated: a Classic Position-Based Visual Servoing algorithm, which minimizes pose error using a quaternion-based approach, and a second solution that utilizes Dual Lie Algebra to compute the 3D visual sensor's velocities, ensuring convergence towards the desired point-feature configuration. Experimental validation reveals that while both methods achieve docking, the dual pose-free approach enables more robust, effortless movement of the robot platform than Classic Position-Based Visual Servoing. Consequently, these findings indicate that integrating depth-based feature recovery with advanced algebraic strategies offers a stable control strategy for automated industrial scenarios.
Endovascular therapy is preferred over open surgery due to its minimally invasive nature, faster recovery, and lower perioperative risk; however, fluoroscopy guided procedures are limited by radiation exposure, high equipment costs, and reliance on highly skilled operators. This study aims to develop and evaluate a lightweight, portable robotic system for autonomous guidewire navigation to improve safety, accessibility, and operator independence. A compact 400 g robotic device was designed with millimeter scale positioning accuracy, servo current-based real-time haptic feedback, precise axial rotation, automated retraction advance control, and compatibility with standard endovascular tools. Miniature linear actuated servomotors replicate skilled manual maneuvers using impedance control. Upon tip contact, the system advances the guidewire by 1 mm, measures current changes, classifies lesion stiffness (soft, medium, stiff), and adapts virtual mass-spring-damper gains to regulate push speed and applied force. Bench experiments were conducted using flexible tubing with inserts simulating 20-80% stenosis and two current thresholds (75 and 94 mA). Retraction frequency increased with stenosis severity, validating the autonomous control strategy. Lesion stiffness classification achieved F-scores of 0.83, 0.77, and 0.95 for soft, medium, and stiff conditions, respectively, demonstrating reliable discrimination and adaptive force modulation. The proposed system enables autonomous and adaptive guidewire advancement with high classification accuracy using low-cost, current-based sensing and impedance control. Its lightweight and portable design reduces dependence on continuous manual operation and specialized imaging infrastructure, supporting safer and faster interventions and potential deployment in prehospital or resource-limited settings. This prototype advances the development of more accessible and operator-independent endovascular therapy.
Cycloidal (RV-type) reducers are widely used in industrial robot joints due to their high torque density and low backlash, yet their multi-mesh transmission path produces structured, operating-point-dependent vibration components at the disc-mesh order and associated harmonics and sidebands. This paper presents a real-time, accelerometer-in-the-loop vibration suppression framework that reduces these components online while maintaining tracking performance within the bounds observed in our experiments and operating within predefined safety limits. A tri-axial accelerometer mounted on the reducer housing provides high-bandwidth vibration measurements from which order-synchronous, band-limited metrics are computed in streaming form. These metrics define both the optimization objective and vibration exposure constraints. The control architecture retains the vendor servo loops and adds a vibration-targeted layer combining a low-dimensional anti-resonance parameterization (adaptive notch shaping and narrowband feedforward cancellation aligned with the estimated mesh-order family) with a safety-certified contextual Bayesian optimization module that adapts the parameters as a function of operating context (speed, load proxy, and temperature proxy). A barrier-function-based safety filter runs at the servo rate to enforce constraint handling during operation; its effect is evaluated empirically through logged interventions and constraint statistics. Experimental evaluation on a cycloidal joint testbed across multiple speeds and load levels shows attenuation of the dominant mesh-order vibration component and its harmonics. Tracking accuracy and safety-related signals remained within preset limits during the tested operating conditions. The proposed approach provides a deployable pathway for online vibration minimization in cycloidal robot joints without requiring high-fidelity internal contact models, and its logged parameter trajectories and order-tracked metrics also offer a foundation for condition-aware adaptation over long-term operation.
Professional UAV thermal imaging systems are widely used for inspection, environmental monitoring, search and rescue, agriculture, and technical diagnostics. However, their cost limits their use in education, preliminary field screening, rapid prototyping, and low-resource applications. This study evaluates a minimum-cost indirect UAV thermal sensing workflow based on a DJI Mini 4K consumer drone, a lightweight Servo King9000 smartphone, and a UTi260M smartphone-connected infrared thermal camera. In the proposed configuration, the smartphone displayed and recorded the thermal stream, while the onboard RGB camera of the UAV recorded the smartphone-displayed infrared video during flight. The aim was not to develop a radiometric UAV thermal imaging platform, but to determine whether such a low-cost configuration can provide qualitative presence/absence indication of clear thermal hotspots and to identify its operational limits. The system was experimentally assessed under no-payload and payload conditions, daylight and nighttime illumination, and several low-altitude operating heights. Additional motor-region thermal observations were performed using a UTi260T handheld thermal camera under loaded and unloaded operating conditions. The complete UAV-payload configuration had a measured mass of approximately 340 g, corresponding to an effective added payload of 91 g and a payload-to-UAV mass ratio of 36.5%. Payload operation reduced near-ground flight endurance from approximately 25 min to 14 min 40 s. The maximum observed motor-region temperature increased from 24.9 °C under unloaded operation to 42.0 °C under loaded operation, while motor thermal asymmetry increased from 4.8 °C to 7.6 °C. Nighttime and low-glare operation improved the readability of the smartphone-displayed thermal stream, with the most practical usability observed at approximately 10-20 m. The results show that the proposed workflow is feasible only for short-range qualitative thermal screening and clear hotspot presence/absence indication. The UAV-recorded video should not be interpreted as direct thermal data, but as an RGB recording of a smartphone display showing thermal information. Therefore, the workflow is not suitable for quantitative temperature measurement, radiometric thermal mapping, or accurate thermal shape delineation. The main operational limits are payload mass, suspended-load oscillation, display readability, reduced endurance, motor-region thermal loading, sensitivity to payload alignment, and the absence of raw radiometric data. Direct UTi260M smartphone-recorded thermal frames were additionally used for pixel-size-assisted qualitative verification of practical reference thermal targets, including a human-sized target and a vehicle-sized target, at selected low-altitude operating heights.
Background/Objectives: Peri-implantitis is a common complication affecting approximately 24% of dental implants and is characterized by progressive bone loss and reduced implant stability. Implantoplasty, an intraoral procedure used to remove biofilm by machining the titanium implant surface, has become increasingly common in clinical practice. However, this procedure may compromise the mechanical integrity of implants, especially when combined with peri-implant bone loss, potentially leading to premature fatigue failure. This study evaluated the effect of different marginal bone resection depths, with and without implantoplasty, on the cyclic mechanical behavior of dental implants. Methods: A total of 200 commercially pure grade 4 titanium implants were embedded in resin simulating human bone at depths of 3, 4, and 5 mm. A subset of implants underwent implantoplasty with a 0.4 mm surface reduction corresponding to the thread width. Finite element analysis was performed to evaluate von Mises stress distribution and predict fatigue behavior. Numerical results were experimentally validated using a servo-hydraulic MTS Bionix system under ISO 14801:2016 conditions. Fatigue limits were determined from the asymptotic region of the load-cycles-to-failure (S-N) curves, and fracture surfaces were examined by scanning electron microscopy. Results: Maximum von Mises stresses were concentrated at the thread-body transition and increased with greater marginal resection depth, with additional stress amplification observed after implantoplasty. Fatigue limits for untreated implants were approximately 351 N, 285 N, and 210 N for 3-, 4-, and 5 mm resections, respectively. Implants subjected to 0.4 mm implantoplasty showed fatigue limits of 311 N, 270 N, and 90 N, respectively. Failure patterns were load-dependent: higher loads produced coronal fractures, whereas lower loads resulted in failure at the implant-abutment connection. Finite element predictions showed strong agreement with the experimental results. Conclusions: Excessive marginal resection significantly decreases the fatigue resistance and long-term mechanical reliability of dental implants, particularly when combined with implantoplasty. The main limitations of this study include is in vitro design, the assumptions inherent to the numerical models, and the variability associated with implantoplasty procedures.
Cementitious sand and gravel (CSG) is commonly used in construction engineering; however, its mix proportion design is complex, and traditional physical experiments face limitations such as long cycles, high costs, and susceptibility to external factors when obtaining high-quality sample data. In this study, a foundational dataset was first acquired through physical experiments: 100 sets of CSG specimens with different mix proportions (cement content 40, 50, 60, 70 kg/m3; water-to-binder ratio 1.0, 1.2, 1.4; sand ratio 0.1, 0.2, 0.3, 0.4; fly ash content 20, 30, 40, 50 kg/m3) were prepared. After 28 days of standard curing, compressive strength and splitting tensile strength tests were conducted using a WAW-1000 electro-hydraulic servo universal testing machine, yielding 100 sets of real mechanical property data. The coefficients of variation for all test groups were below 10%, confirming the reliability and repeatability of the experimental data. On this basis, a data augmentation method based on Conditional Tabular Generative Adversarial Networks (CTGAN) is proposed. Through adversarial training between the generator and the discriminator, the model learns the multi-dimensional distribution characteristics of the original CSG data and generates 100 synthetic samples, which are then merged with the original data to expand the dataset to 200 samples. The quality of the synthetic data is evaluated using Wasserstein distance and correlation matrix heatmaps. Furthermore, a hybrid XGBoost-LSTM prediction model is proposed-XGBoost is used for feature construction to capture nonlinear interactions among mix proportion variables, and the constructed features are then fed into an LSTM network for sequential learning and regression prediction. The results show that the CTGAN-generated data are highly consistent with the original data in terms of kernel density distributions and variable correlations, with Wasserstein distance significantly superior to four comparative methods: Bootstrap, SMOTE, GaussianCopula, and TVAE. After augmentation, the XGBoost-LSTM model achieves a coefficient of determination (R2) of 0.9897 for compressive strength prediction (vs. 0.9793 before augmentation) and 0.9801 for splitting tensile strength (vs. 0.9882 before augmentation, a slight decrease). The mean absolute percentage errors (MAPE) are 4.49% and 4.11%, and the root mean square errors (RMSE) are 0.201 and 0.049, respectively; both error metrics are reduced compared with those before augmentation. Compared with baseline models including XGBoost, LSTM, Random Forest (RF), and Support Vector Regression (SVR), the XGBoost-LSTM model exhibits the best performance across all evaluation metrics, and Wilcoxon signed-rank tests confirm that the performance differences are statistically significant (p < 0.05). The proposed method of CTGAN-based data augmentation combined with the XGBoost-LSTM hybrid model provides an effective solution to the problem of insufficient CSG sample data and offers a reference for data enhancement and performance prediction of other small-sample materials.
In visual servoing micromanipulation, the hysteresis nonlinearity of piezoelectric stages and image transmission delay significantly degrade positioning accuracy. To address these issues, this paper proposes a dual-layer control strategy based on an improved Extended Kalman Filter (B-W-EKF). First, image block matching combined with a Gaussian kernel interpolation algorithm is employed to obtain high-precision displacement measurements from microscopic image sequences, from which the voltage-displacement hysteresis loop is constructed. Then, the EKF is integrated with the Bouc-Wen (B-W) model, incorporating hysteresis nonlinearity into the state observation equations. Based on this model, a dual-layer control architecture that combines upper-layer Model Predictive Control (MPC) with lower-layer Sliding Mode Control (SMC) is designed: the upper-layer MPC performs global optimization, while the lower-layer SMC regulates position and velocity, thereby improving tracking accuracy. Experimental results show that the RMSE values for SMC, MPC, SMPC, iMPC, and the proposed dual-layer MPC-SMC are 0.1292 µm, 0.1366 µm, 0.0635 µm, 0.0827 µm, and 0.0372 µm, respectively, under triangular wave reference input, demonstrating the effectiveness of the control strategy in enhancing tracking precision.
To evaluate the face and content validity of a novel, 3D-printed nasotracheoscopy simulator with a motorized, moving larynx using expert feedback from board-certified otolaryngologists. Device development and validation study. An academic institution. An anatomically accurate airway model was created from a de-identified normal patient CT scan using open-source segmentation software. Negative impressions of the turbinates and nasopharynx were 3D printed in PLA and cast in silicone for realistic soft tissue. A custom-coded Arduino-controlled servo motor was integrated to simulate laryngeal motion. Ten board-certified otolaryngologists performed flexible nasotracheoscopy on the model and completed a survey assessing face validity, content validity, and general impressions using a 5-point Likert scale. Descriptive statistics summarized responses. The simulator received high scores for face validity (mean 4.6, SD 0.67), content validity (mean 4.8, SD 0.40), and general impressions (mean 4.83, SD 0.42). All participants agreed that the model accurately represented anatomical landmarks and held educational value. Feedback emphasized its utility for telescopic navigation and anatomical recognition training. This 3D-printed simulator demonstrated strong validity and realism. Its anatomical accuracy, dynamic features, and reproducibility support its use as a training adjunct in procedural training. Future studies will assess its effect on learner performance.