The rapid and sensitive detection of trace ammonia (NH3) at room temperature is crucial for environmental monitoring, occupational safety, and noninvasive disease diagnosis. However, pristine two-dimensional tungsten disulfide (WS2) suffers from oxidative instability in air and sluggish room-temperature gas adsorption/desorption kinetics, limiting its sensing performance. Herein, a supercritical CO2 (SC CO2)-assisted heterointerface engineering strategy was developed to construct interface-purified WS2/reduced graphene oxide (rGO) heterostructures via a reductant-assisted one-step process. The results show that SC CO2 treatment at 180 degrees C, combined with the reductant, simultaneously promotes WS2 exfoliation, in situ reduction of GO, and interfacial purification, leading to an "ultrathin nanosheet-clean interface-strong electronic coupling" structure. TEM results confirm that WS2 nanosheets are uniformly anchored on rGO, while X-ray photoelectron spectroscopy (XPS) reveals markedly reduced W6+ and sulfate species. Benefiting from the purified heterointerface and enhanced interfacial charge transfer, the SC-WS2/rGO sensor exhibits fast response/recovery times of 11.4/11.7 s toward 6 ppm NH3 at room temperature, stable and reversible responses over the range of 6-900 ppm, a low theoretical detection limit of 13.11 ppb, negligible degradation after 30 days, and a selectivity up to 25 times higher than that toward interfering gases. Mechanistic analysis indicates that the clean heterointerface lowers the carrier transport barrier and exposes active sites, while electron delocalization provides efficient charge transport pathways. An effective interface engineering strategy for high-performance room-temperature NH3 sensors based on two-dimensional transition metal sulfides is demonstrated in this work.
In this study, we present the development of a smart, dual-axis continuous passive motion (CPM) rehabilitation device for patients undergoing total hip arthroplasty. Most commercial CPM systems provide only uniaxial flexion-extension motion, limiting their application in comprehensive hip rehabilitation. The proposed system introduces an additional abduction mechanism to enable multidirectional movement during postoperative recovery. The mechanical structure adopts a modular design that allows an independent assembly of the flexion and abduction modules. A programmable logic controller (PLC)-based electrical control unit integrated with a human-machine interface (HMI) was developed for coordinated dual-axis operation. Four endpoint position sensors were installed at the 0 and 90 degrees limits of the flexion module and the 0 and 30 degrees limits of the abduction module. Intermediate positions are controlled by the PLC according to the stepper-motor pulse count and linear-actuator stroke, while the endpoint position sensors provide origin confirmation, travel-limit detection, and overtravel protection. Under abnormal overtravel conditions, the PLC stops the corresponding motion and displays a warning on the HMI. Functional testing confirmed smooth and stable dual-axis operation. These results demonstrate the feasibility of the proposed system and indicate greater motion versatility and integrated control functionality than conventional single-axis CPM configurations, supporting its potential for future clinical application.
Electrocardiogram-related physiological activity and blood oxygen concentration (SpO2) are critical indicators for monitoring cardiovascular status. In this study, we integrate pulse rate and SpO2 signals with a multi-frequency bioelectrical impedance sensing distribution analysis (MF-BISA) framework to determine the optimal timing for hemodialysis termination. These sensing modalities provide complementary physiological information for evaluating hemodynamic response and fluid status. In water-retentive patients, body impedance changes dynamically with fluid redistribution, tissue hydration, and electrolyte balance during hemodialysis. Therefore, continuous bioelectrical impedance sensing enables the non-invasive monitoring of physiological fluid variations. The proposed MF-BISA framework applies multi-frequency electrical currents and analyzes voltage responses to characterize tissue-dependent impedance behavior during dialysis. A numerical signal analysis framework processes multi-site impedance data from sensor patches at the cardiac output (chest) and peripheral (calf) regions. Impedance distribution curves over a 4 h session are constructed for both locations. Physiological equilibrium is reached when body fluid volume, tissue hydration, and electrolyte concentration converge, reflected by the intersection of the two curves. Experimental results show a crossover point at approximately 1.94 h, indicating the optimal dialysis termination time. In this work, we developed a sensor-driven multi-modal monitoring framework integrating bioelectrical impedance sensing with pulse oximetry to provide a quantitative, real-time solution for fluid status assessment and dialysis optimization.
Industrial manufacturing systems increasingly rely on large-scale sensor networks to monitor equipment states and production processes. For example, in continuous casting, casting-speed signals are continuously acquired by encoders and transmitted to a programmable logic controller (PLC) or a distributed control system (DCS) through industrial sensing links. These sensor-generated time-series signals provide essential information for intelligent perception and fault diagnosis, yet often lack high-quality annotations. To address this issue, in this paper, we proposed the Industrial Deep Time Series Detection-Q-Network (IDTS-DQN), a semi-supervised deep reinforcement learning algorithm, for anomaly detection in industrial sensing systems. The method integrates deep learning and reinforcement learning through a dual reward mechanism and an anomaly-based experience replay strategy, enabling the agent to learn temporal patterns with limited labeled samples and dynamically improve detection performance. Extensive experiments on real-world industrial datasets demonstrate that IDTS-DQN outperforms baseline methods in recall, precision, and F1-score, providing an adaptive sensing-data analysis approach for industrial monitoring and demonstrating the potential of deep reinforcement learning in sensor-based anomaly perception. However, the framework has been evaluated only offline on historical sensing data, and its applicability to broader signal types and fault modes remains to be examined. Future work will focus on multimodal sensor fusion, edge-based online deployment, and domain-knowledge integration for industrial sensor-signal analysis.
Simple three-layer broadband optical absorbers have attracted considerable attention because of their potential applications in solar-energy harvesting, thermal emitters, and highperformance optical sensing systems. In particular, broadband absorbers can significantly enhance light-matter interaction, making them promising candidates for improving the performance of optical sensors and ultraviolet photodetectors. In this study, a simple three-layer optical absorber composed of a W ground plane, an Al2O3 dielectric spacer, and a top W square resonator containing a centrally hollow cylindrical cavity was designed and analyzed using the finite element method implemented in COMSOL Multiphysics. The effects of structural parameters, including the thicknesses of the W and Al2O3 layers, the dimensions of the top resonator, and the radius of the hollow cylindrical cavity, were systematically investigated to optimize the broadband absorption performance. In addition, optical impedance, electric field distribution, magnetic field distribution, and angular absorption characteristics under both transverse electric and transverse magnetic polarizations were analyzed to clarify the underlying absorption mechanisms. The optimized absorber exhibits absorptivity exceeding 90% over the wavelength range of 200-1760 nm, together with excellent angular stability under oblique incidence. Owing to its simple structure, broadband absorption capability, and robust optical performance, the proposed design is a promising candidate not only for solar-energy harvesting but also for broadband optical sensors, ultraviolet photodetectors, and other integrated optoelectronic sensing devices.
