This paper presents a novel wireless silent speech interface (SSI) integrating multi-channel textile-based EMG electrodes into headphone earmuff for real-time, hands-free communication. Unlike conventional patch-based EMG systems, which require large-area electrodes on the face or neck, our approach ensures comfort, discretion, and wearability while maintaining robust silent speech decoding. The system utilizes four graphene/PEDOT:PSS-coated textile electrodes to capture speech-related neuromuscular activity, with signals processed via a compact ESP32-S3-based wireless readout module. To address the challenge of variable skin-electrode coupling, we propose a 1D SE-ResNet architecture incorporating squeeze-and-excitation (SE) blocks to dynamically adjust per-channel attention weights, enhancing robustness against motion-induced impedance variations. The proposed system achieves 96% accuracy on 10 commonly used voice-free control words, outperforming conventional single-channel and non-adaptive baselines. Experimental validation, including XAI-based attention analysis and t-SNE feature visualization, confirms the adaptive channel selection capability and effective feature extraction of the model. This work advances wearable EMG-based SSIs, demonstrating a scalable, low-power, and user-friendly platform for silent communication, assistive technologies, and human-computer interaction.

Many people worldwide suffer from motor impairments, with aging and stroke being two of the main causes, resulting in limited mobility for carrying out activities of daily living, reduced walking speed, increased metabolic consumption, and higher risk of falling. Exoskeletons are widely seen as a promising assistive technology for addressing these major challenges and mitigating the impacts of motor impairments. This review provides a comprehensive assessment of wearable active exoskeleton technologies, including their sensing, actuating, and control strategies, for elderly support in healthy aging applications. Specifically, we focus on actuators, sensors and control strategies for active exoskeleton technologies applied to the ankle joint, as these play a critical role in human mobility assistance. By defining a complexity index that accounts for both mechanical and computational complexity of different exoskeleton solutions, we provide a comparative analysis of the landscape and conclude with a new perspective on future trends in soft, wearable, and modular ankle exoskeleton technologies.

Air pollution has been associated with several health problems. Detecting and measuring the concentration of harmful pollutants present in complex air mixtures has been a long-standing challenge, due to the intrinsic difficulty of distinguishing among these substances from interferent species and environmental conditions, both indoor and outdoor. Despite all efforts devoted by the scientific and industrial communities to tackling this challenge, the availability of suitable device technologies able to selectively discriminate these pollutants present in the air at minute, yet dangerous, concentrations and provide a quantitative measure of their concentrations is still an unmet need. Thermal conductivity detectors (TCDs) show promising characteristics that make them ideal gas sensing tools capable of recognising different gas analytes based on their physical fingerprint characteristics at the molecular level, such as their density, thermal conductivity, dynamic viscosity, and others. In this paper, the operation of TCD gas sensors is presented and explored using a finite element simulation of Joule heating in a sensing electrode placed in a gas volume. The results obtained show that the temperature, and hence, the resistance of the individual suspended microbridge sensor device, depends on the surrounding gas and its thermal conductivity, while the sensitivity and power consumption depend on the properties of the constitutive metal. Moreover, the electrode resistance is proven to be linearly dependent on the applied voltage.

Walking is a key movement of interest in biomechanics, yet gold-standard data collection methods are time- and cost-expensive. This paper presents a real-time, multimodal, high sample rate lower-limb motion capture framework, based on wireless wearable sensors and machine learning algorithms. Random Forests are used to estimate joint angles from IMU data, and ground reaction force (GRF) is predicted from an instrumented insole, while joint moments are predicted from angles and GRF using deep learning based on the ResNet-16 architecture. All three models achieve good accuracy compared to the literature, and the predictions are logged at 1 kHz with a minimal delay of 23 ms for 20s worth of input data.