Sonomyography (SMG) is a promising alternative to surface electromyography (sEMG) for wearable human-machine interfaces. The first wearable PZT-based ultrasound armbands achieve up to 96% hand gesture recognition accuracy. To enhance compactness and acoustic compatibility, flexible piezoelectric micromachined ultrasound transducers (pMUTs) are being explored as replacements for rigid PZT elements. However, the acceptable requirements on operational parameters for optimal recognition performance remain unclear. This study simulates pMUT signals by filtering off-the-shelf PZT probe data and injecting noise to examine how frequency, bandwidth, and noise floor affect gesture recognition.

The RGB (de)multiplexer is a key component in laser display systems and demands advanced design solutions that achieve high efficiency within a small form factor. Here, a CMOS-compatible silicon nitride (SiN) visible light photonic integrated circuits (PICs) platform is employed to successfully demonstrate an ultracompact RGB (de)multiplexer, featuring a footprint as small as 86 × 2.3 µm 2 . Leveraging a novel, to the best of our knowledge, dual-segmented multimode waveguide design, the device achieves a low insertion loss of ~1–2 dB at target wavelengths and sets the stage for monolithic integration of laser sources on the same PIC chip. This breakthrough further paves the way for the development of integrated, miniaturized CMOS-compatible RGB light engines offering superior brightness, high resolution, and a wide field-of-view. We anticipate that our design can contribute significantly to the development of immersive AR/VR/MR display technology, while simultaneously extending its impact into sensing and visible light communication applications.

Next generation quantum technologies will need to rely on efficient transduction between electrical, optical, and mechanical quantum degrees of freedom to generate large-scale entanglement over large distances. The performance of such transducers is fundamentally limited by the cryogenic properties of the underlying materials. Here, we demonstrate that engineering strain in ferroelectric thin-film strontium titanate (SrTiO_3) not only results in an exceptionally large Pockels coefficient, but also in a robust linear piezoelectric response at cryogenic temperatures, surpassing previous thin-film benchmarks. We measure piezoelectric tensor elements of d_15 = 151.8 ± 1.5 pm/V and d_33 = 54.8 ± 4 pm/V, and an effective photoelastic coefficient of p_eff = 0.56 at 5 K. Utilizing these enhanced properties, we demonstrate the first SrTiO_3-on-oxide acousto-optic modulator with a voltage-length product (V_πL) of 0.874 ± 0.084 V.cm, outperforming state-of-the-art unreleased modulators that typically feature a V_πL of a few V.cm. Our results establish thin-film SrTiO_3 as a promising material system for integrated quantum photonics operating at cryogenic temperatures.

Optical phase-change materials (OPCMs) offer a powerful platform for electrically driven reconfigurable photonics, yet the development of low-loss optical phase change materials, in particular in the visible range, remains a significant challenge. Currently investigated materials such as Sb2Se3 and Sb2S3 remain quite absorptive in that wavelength range. In this work, we demonstrate the use of molybdenum oxide (MoOx, x approximate to 3) as a low-loss OPCM. Through ellipsometry, we observed large changes in the refractive index between its amorphous and crystalline phases. This transition is achieved by annealing the material above its crystallization temperature, with crystallization confirmed via X-ray diffraction. In both phases, the imaginary part of the refractive index for green light was below 1x10 (4), demonstrating the material's potential for reconfigurable photonic devices. With an on-chip heater, Joule heating pulses were used to induce a reversible amorphous-to-crystalline phase transition, confirmed with selected area electron diffraction (SAED). During optical characterization, a change in reflectivity was observed. However, signs of layer delamination were also identified, which are expected to contribute to the reflectivity variation. Nevertheless, the observed reversible crystallization of MoO. using electrical pulses marks an important step toward its use in reconfigurable flat optics. Thermal simulations guided the microheater design to enable localized switching, while optical simulations provide a baseline for expected modulation. This work represents the first experimental demonstration of electrically induced phase switching of MoO. thin films integrated into an optical device, highlighting both the promise and practical challenges of this emerging material system.

To achieve efficient ultrasonic power transfer using piezoelectric micromachined ultrasonic transducers (pMUTs), both interdependent electrical matching and distance dependent acoustic coupling must be considered together. Exploring this design space with full-wave multiphysics simulations alone can become computationally demanding, especially when multiple parameters are involved. In this work, we develop a frequency domain system-level model for a single Tx–Rx pMUT link that integrates electrical terminations, electromechanical conversion, acoustic radiation and coupling into a mixed-domain two-port framework. The propagation path is described by a closed-form propagating wave transfer impedance, which allows rapid sweeps over frequency and normalized spacing d/λ. To identify practical matching conditions, we further introduce a power landscape procedure that evaluates the power delivered across the complex load plane on both the Rx and Tx ports. The corresponding target impedances can then be selected and used for Impedance Matching Network (IMN) synthesis. The overall approach is compared with COMSOL simulations for different device radii and wavelength normalized separations. In these cases, the model reproduces the main trends in peak transferred power while reducing the evaluation time by orders of magnitude.

Scandium-doped aluminum nitride (ScAlN)-based piezoelectric micromachined ultrasonic transducer (PMUT) arrays have attracted increasing attention in acoustofluidics for micro total analysis systems (μTAS), particularly for applications involving acoustic radiation force for bioparticle manipulation and cell manipulation. However, their use for fluid handling via acoustic streaming remains underexplored. This study, for the first time, examines the potential of a rectangular membrane ScAlN-based PMUT array to generate directional acoustic streaming for micro-pumping applications. The PMUT array is embedded within a PDMS microfluidic channel and is driven by a set of AC signals with a 120° phase difference between adjacent PMUT cells to induce directional streaming flow. The device features a compact active area of 1.2 × 1.6 mm and demonstrates a volumetric flow rate of 0.12 μL/min, in good agreement with predictions from numerical multiphysics simulations. Further numerical optimization suggests that the flow rates of 1.0 μL/min are achievable by optimizing the array kerf (lateral spacing between adjacent PMUT cells) and applied phase difference to adjacent PMUTs. A comparative analysis with state-of-the-art chip-integrable micropumps highlights the advantages of the proposed device, including its miniaturized footprint, CMOS compatibility, and ease of on-chip integration. These attributes position the proposed micropump as a promising solution for μTAS applications, especially where compact size and precise, low-flow-rate fluid control are critical.

Ultrasound (US) has emerged as a promising modality for Human-Machine Interfaces (HMIs), with recent research efforts exploring its potential for Hand Pose Estimation (HPE). A reliable solution to this problem could introduce interfaces with simultaneous support for up to 23 degrees of freedom encompassing all hand and wrist kinematics, thereby allowing far richer and more intuitive interaction strategies. Despite these promising results, a systematic comparison of models, input modalities and training strategies is missing from the literature. Moreover, there is only one publicly available dataset, namely the Ultrasound Adaptive Prosthetic Control (Ultra-Pro) dataset, enabling reproducible benchmarking and iterative model development. In this paper, we compare the performance of six different deep learning models, selected based on diverse criteria, on this benchmark. We demonstrate that, by using a step learning rate scheduler and the envelope of the RF signals as input modality, our 4-layer deep UDACNN surpasses XceptionTime's performance by 2.28 percentage points while featuring 87.52% fewer parameters. This result (77.72%) constitutes an absolute improvement of 0.88% from previously reported baselines. According to our findings, the appropriate combination of model, preprocessing and training algorithm is crucial for optimizing HMI performance.

We introduce a suspended graphene membrane-based optomechanical sensor with integrated photonic readout. Numerical simulations, a precisely controlled CMOS-compatible fabrication process, and initial optical measurements promise high displacement sensitivity and potential scalability for parallel in vivo probing in lab-on-chip platforms for biomedical sensing and pharmacology applications.

A novel, CMOS compatible ScAlN-based micropump driven by a discrete rectangular membrane piezoelectric micromachined ultrasonic transducer (PMUT) array is developed. PDMS containing a microfluidic channel is bonded to the top of the actuator. When subjected to a phase-delayed AC electrical excitation, the PMUT array generates a directional acoustic wave within the microfluidic channel, producing a directional streaming flow pattern. The compact device, with an active area of 1.2 x 1.6 mm(2), achieved a mean flow velocity of 0.16 mm/min, which is consistent with predictions from numerical multiphysics simulations. Further optimization via numerical simulations indicates augmentation of mean flow velocity up to 2.1 mm/min is possible. The micropump has the potential for integration into mu TAS applications, where liquid control and delivery in a small form factor are critical.

We present the fabrication and characterization of capacitive micromachined ultrasound transducers (CMUTs) on glass substrates. The devices exhibit electromechanical coupling efficiency, fractional bandwidth, and transmit sensitivity comparable to state-of-the-art silicon-based CMUTs for an in-water frequency range of 4 - 12 MHz. Performance improves with higher DC bias, though dielectric charging may affect long-term stability. The used process flow is compatible with existing flat-panel display production lines, paving the way for large-area and flexible transducer arrays tailored for continuous ultrasound monitoring in and outside hospitals.

We experimentally demonstrate the use of a micro-transfer-printed on-chip distributed Bragg reflector laser to interrogate an opto-mechanical ultrasound sensor. By eliminating the need for an external tunable laser source, this work presents a scalable pathway toward an all-on-chip photonic platform for high-performance acoustic sensing and imaging.

The pull-in and pull-out voltages are important characteristics of Capacitive Micromachined Ultrasound Transducers (CMUTs), marking the transition between conventional and collapse operation regimes. These voltages are commonly determined using capacitance–voltage (C-V) sweeps. By modeling the operating conditions of an LCR meter in COMSOL Multiphysics®, we demonstrate that the measured capacitance comprises both static and dynamic capacitances, with the dynamic capacitance causing the appearance of a peak in the effective C-V curve. Furthermore, Laser Doppler Vibrometer (LDV) measurements and electromechanical simulations indicate the occurrence of collapse–snapback phenomena during the C-V sweeps. This study, through advanced simulations and experimental analyses, demonstrates that the transient membrane behavior significantly affects the apparent capacitance–voltage characteristics of electrostatically actuated Micro-Electromechanical Systems (MEMS).

Integration of polarization sensing into conventional imaging devices can uncover crucial information, unlocking applications across a wide range of scientific domains from biomedical imaging to remote sensing and astronomy. Rapid progress in nanofabrication technology has driven the trend of pixel downscaling in imaging devices, leading to improved device functionality and performance. For polarization imaging to engage with high-resolution image sensor a further scaling of per-pixel polarizing objects is needed, while maintaining the optical throughput. In this paper, we propose a design for an ultracompact polarization splitting device based on a vertical dielectric waveguide architecture, compatible with sub-micron pixel sizes (560 nm), which is close to the state-of-the-art CMOS imager sensor resolution. The device implements an adiabatic mode converter to effectively separate orthogonal linear polarization states of incident light across the entire visible spectrum. Moreover, we demonstrate the design versatility by expanding its polarization splitting functionality across the entire linear polarization space (0°–180°), enabling robust polarization detection for advanced polarization imaging applications.

A 5-Gbit/s joint OWC and FMCW LiDAR system is experimentally demonstrated over a 1.5 m free-space optical link. No optical amplification is needed in the optical transmission system by using a calibration-free and low-loss (similar to 12 dB) 512-element OPA. (c) 2025

A breakthrough in endoscopy imaging resolution, coupled with a drastic reduction in its invasiveness, is on the horizon with emerging techniques based on multimode fibers. These imaging techniques rely on illumination wavefront control to deliver diffraction-limited images through a thin multimode fiber (diameter similar to 100 mu m). By controlling the wavefront at the fiber's input, a grid of focused spots can be raster scanned at the fiber's end to form an image. However, there's room for performance improvement in spatial light modulators commonly used for focusing through multimode fibers. The ones based on liquid crystals suffer from low sub-kHz modulation speed, while the other, digital micromirror devices, require an intricate optical setup to achieve preferred phase modulation. We propose a novel approach for focusing through a fiber by employing a photonic integrated circuit to modulate the wavefront, which brings improvements in the modulation rate and the optical setup compactness. Using an optical phased array with 128 antennas, operating at a wavelength of lambda = 852 nm, we demonstrate focusing through a graded-index multimode fiber. Spot size as low as 2.3 mu m, approaching the diffraction limit, and an average focus power ratio equal to 93% of the theoretical maximum are achieved. Additionally, we demonstrate 2D focus steering at the fiber's distal end for raster-scan imaging. By focusing in different planes, we show that OPAs can enable volumetric imaging through a multimode fiber. Finally, our device was fabricated on a silicon nitride platform, offering the prospect of large-scale fabrication.

In recent years, researchers constantly attempt to derive Ultrasound-based (US-based) hand gesture recognition (HGR) solutions suitable for edge applications. This process involves improving several design aspects of US-based HGR systems such as the transducers, the wearable US acquisition systems and the algorithms employed in terms of energy consumption, computational complexity and robustness. The subject of this paper is the latter. In this paper, we present a spiking framework for US-based HGR. The proposed approach leverages a single-layer Spiking Neural Network (SNN) equipped with Spike-Timing-Dependent Plasticity (STDP) as a feature descriptor for Rate-based (RB) coded A-line US signals coupled with a lightweight linear support vector machine (SVM) classifier. According to our findings, our proposed approach achieves performance comparable to that of the state-of-the-art in the ultrasound-based adaptive prosthetic control (Ultra-Pro) dataset. Furthermore, we demonstrate that our feature descriptor exhibits inter-session generalization capabilities, i.e. re-training is not required between within-day sessions and thus reduces the burden of periodic extensive data collection from the user.

High-resolution optical imaging in thick tissue samples remains elusive, mainly because of the scattering exhibited by the tissue. With increasing depth, the number of nonscattered photons exponentially decreases – limiting the use of conventional imaging techniques at depth. Wavefront shaping is a novel technique that aims to enable imaging at depth by refocusing the scattered light. However, significant wavefront-control hardware improvements are necessary to unlock the applications in in vivo microscopy. Optical phased arrays (OPAs), realized in integrated photonics, can provide improvements in the pixel pitch, operation speed, and system compactness compared to conventionally employed spatial light modulators. We compare different OPA designs for focusing in tissue-like forward-scattering samples. OPA design trade-offs, such as the array pitch, number of antennas, and antenna emission profile, are experimentally studied, and their influence on the device performance is highlighted. We do this for increasing thickness of the forward-scattering sample and observe two distinct regimes. The devices, operating at the wavelength of λ = 852 nm, were fabricated on a SiN photonics platform suitable for both near-infrared (NIR) and visible (VIS) light.