A comprehensive wind tunnel investigation was conducted to analyze noise generation, propagation, and perception mechanisms in a boundary layer ingesting (BLI) ducted fan through integrated aeroacoustic and psychoacoustic assessments. The study examines interactions between an incoming adverse pressure gradient turbulent boundary layer flow, developed over a curved wall, and the ducted fan. The fundamental investigation confirms that the fan thrust regime influences aerodynamic, aeroacoustic, and psychoacoustic characteristics, exhibiting various haystacking phenomena. High-thrust operation induces a pronounced upstream suction effect, accelerating the boundary layer flow, amplifying bulk momentum, and intensifying turbulence ingestion, leading to fan aeroacoustics and associated fan haystacking in noise spectrum. In contrast, low-thrust operation minimally alters the boundary layer flow, with reduced suction and noise dominated by duct aeroacoustics and the associated duct haystacking due to interactions between ingested turbulence and the duct's acoustic field. The psychoacoustic assessments indicate that both fan and duct haystacking contribute to higher perceived noise in the high- and low-thrust regime, respectively.
Recent advances in cough sound analysis using deep learning techniques enable smartphone-based respiratory disease screening suitable for self-management care in a home setting, yet their utility is limited by device heterogeneity, population diversity, and challenges in multimodal integration. We propose a device-invariant, multimodal deep learning framework that jointly models cough acoustics, demographic data, and symptom descriptions for multi-label classification of adult respiratory diseases. To address the issues of device effect, an adversarial branch is embedded in the audio encoder to enforce device-invariant feature learning, while an invariant risk minimization-augmented loss enhances robustness to non-structural shifts. To evaluate the effectiveness of our proposed method, a real-world, multi-center dataset containing over 10,000 cases spanning seven major respiratory conditions was curated. On the tasks of individual respiratory disease identification for chronic obstructive pulmonary disease (COPD), lower respiratory tract infection (LRTI) and pulmonary shadows (PS), our method achieves superior performance with the area under the receiver operating characteristic curve (AUROC) of 0.9698, 0.8483 and 0.8720, respectively. It also shows promising results in identifying the presence of comorbidities for 7 respiratory diseases with an overall AUROC of 0.8907. More importantly, extensive experimental results demonstrate our method mitigates the issues of device effect and facilitates the cross-device generalization for cough-based respiratory disease diagnoses. This work demonstrates a scalable and transferable AI-based approach for cough-driven respiratory screening, emphasizing the importance of multimodal fusion and robust representation learning in advancing clinical applicability.
Patient safety and high treatment quality are essential in modern healthcare, but analyzing safety incidents for insights is labor-intensive and inconsistent process. To address this, we developed the Artificial Intelligence-based Incident Analysis and Learning System (AI-ILS), trained on 1548 expertly curated incidents categorized by the Human Factors Analysis and Classification System (HFACS). AI-ILS identifies latent safety threats and classifies incident causes with high accuracy, achieving an average AUROC of 0.92, MCC of 0.72, and overall accuracy of 79%. In testing on 350 real-world clinical incidents, AI-ILS showed 88% concordance with expert reviewers and operated 29 times faster than manual analysis. We deployed and validated AI-ILS using real-world radiation oncology data, where it improved retrospective incident analysis at our institution by generating aggregated HFACS-based results and addressing challenges related to inconsistent review processes and lack of standardized taxonomies.
Anthropogenic underwater noise poses a significant threat to marine ecosystems, disrupting key biological functions. Common mitigation strategies include enclosing noise sources within acoustic barriers. Current designs include locally resonant absorbers, which offer narrow-band performance, and reflective systems with limited effectiveness at low frequencies. In this work, we propose an approach to design thin anisotropic metamaterial-based acoustic barriers for broadband underwater noise attenuation at deep sub-wavelength scales using topology optimization to maximize the coupling between normal stresses and shear strains. Unlike conventional methods, the proposed optimization is formulated in the static regime, relying solely on the homogenized elastic properties of the structured material and not on the characteristics of the surrounding fluid. The resulting metabarriers achieve a high sound transmission loss (STL, 100 dB peak) above 2 kHz, while maintaining a thickness-to-wavelength ratio as low as 1/70 below 1 kHz and STL of approximately 20-30 dB. The influence of hydrostatic pressure on performance is also evaluated, and structural modifications for practical deployment are proposed. The results demonstrate the potential of anisotropy-driven metamaterials as compact and efficient solutions for the control of underwater noise, offering a promising avenue for future acoustic insulation technologies.
Transcranial focused ultrasound (tFUS) is a non-invasive neuromodulatory tool that holds promise for various neuropsychiatric disorders. While it offers several distinct advantages, it also faces notable technical challenges. The irregular shape and inhomogeneous acoustic properties of the human skull impede efficient acoustic energy transmission through the skull. So far, clinical semi-spherical (hemispherical) arrays still suffer from strong wave reflection and refraction at the skull interface, especially with steering. We propose a flexible ultrasound array that conforms to individual skull shapes and can be optimized to target vertex-accessible subcortical regions. The impact of flexible array configuration was investigated by comparing the flexible array with a semi-spherical array commonly used in the clinical setting. Numerical results show that the random-patterned flexible array reduces the z-axis -6 dB full width at half maximum (FWHM) by 29.4% and enhances the focal peak pressure by 44.4% when compared to the semi-spherical array without steering. In addition, it achieves a wide steering range over a 30 × 20 mm2 region while maintaining the focusing performance. We expect that our proposed tFUS stimulation with a flexible array may provide a theoretical framework for improving the therapeutic efficiency for various neuropsychiatric conditions.
This paper presents Soundscape Experience Activities and Mapping (SEAM), a new method for exploring how older adults perceive and relate to their indoor acoustic environments. With global ageing, populations are increasingly choosing to age in place, creating opportunities to enhance older life through the intentional design of supportive home soundscapes. Through a mixed-method approach combining Ecological Momentary Assessment with Cultural Probe methods, we engaged eight older adults (age 56-76) in Belgium to document their domestic soundscape experiences. Reflexive thematic analysis constructed four patterns of meaning: personal agency in shaping acoustic environments, temporal routines structured by sound, sound-memory associations fostering place attachment, and social presence through acoustic monitoring. Within this study context, sounds functioned as spatiotemporal anchors, structuring daily routines while fostering place attachment through memory. This exploratory design research offers situated insights for soundscape interventions that support independence, while highlighting methodological considerations for situated soundscape research.
Neuroinflammation contributes to the progression of many neurological diseases. Here, we explore whether ultrasound can reduce microglia-mediated inflammation in vitro and in vivo. We tested a broad range of ultrasound parameters in a BV2 microglial cell line, treated with lipopolysaccharide (LPS) to induce an inflammatory response. We found that specific combinations of centre frequency, acoustic pressure and treatment duration can significantly lower the levels of pro-inflammatory cytokines, including tumor necrosis factor (TNF)-α, interleukin (IL)-1β and IL-6. These effects lasted up to 72 h and were associated with the downregulation of the nuclear factor κB (NF-κB), suggesting a mechanistic link between ultrasound and inflammation. Further investigation in vivo, in LPS-treated mice, revealed a reduction in TNF-α expression in the hippocampus following ultrasound. Overall, our findings showcase the potential of ultrasound as a non-invasive therapeutic strategy to reduce neuroinflammation and restore brain homeostasis.
[18F] fluorodeoxyglucose positron emission tomography - computed tomography (FDG PET-CT) is increasingly used for staging of breast cancer in the primary and recurrent setting, as well as in evaluating treatment response and in follow-up. Quantitative parameters derived from the primary tumor, even in non-metastatic patients (i.e., without distant metastases but possibly with nodal involvement), have shown prognostic value. Beyond visual interpretation, quantitative evaluations may improve diagnostic accuracy and reproducibility. However, current studies often rely on predefined parameters such as maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV), and total lesion glycolysis (TLG), which may overlook the high-dimensional patterns inherent in FDG PET-CT. To address this, we conducted a deep-learning-based analysis of FDG PET-CT from a large retrospective cohort of non-metastatic breast cancer patients, evaluating prognostic value from multiple perspectives. To improve patient prognosis and risk stratification, we developed a multi-omics prognostic stratification (MOPS) model that integrates clinical data, FDG PET-CT, and corresponding medical reports using CMA and transformer-based architectures to predict overall survival (OS) and disease-free survival (DFS). To support clinical applicability, we incorporated interpretability into the model, providing causal explanations, visualization-based insights, and semantic interpretations to help clinicians understand and apply the predictions transparently. The MOPS model markedly improves survival prediction, outperforming single-omics models, TN staging, and molecular subtyping, with C-index values of 0.75 (95% CI: 0.69-0.81) for OS and 0.71 (95% CI: 0.65-0.77) for DFS.
The impact of minor manufacturing deviations in facesheet orifice geometries on the acoustic impedance of liners is studied. Using the lattice-Boltzmann method, simulations of a normal impedance tube (NIT) with plane acoustic waves at sound pressure levels of 130 and 145 dB and frequencies of 800, 1400, and 2000 Hz were performed. Experimental validation was conducted at the Federal University of Santa Catarina using a baseline geometry obtained via 3D scanning and characterized by rounded orifice edges. This geometry was modified to investigate the influence of various edge configurations: sharp edges, double chamfers, and single top chamfers. Results show that sharp-edged orifices increase acoustic resistance and absorption, while geometries with rounded or chamfered edges reduce resistance by up to 28% and lower the absorption coefficient. This is similar to what was found experimentally by performing NIT measurements over different parts of the liner sample. Velocity field analysis reveals that flow separation at the orifice edge is the primary mechanism driving impedance variation, independent of frequency or sound pressure level. These findings underscore the significant influence of small geometric imperfections, often introduced during manufacturing, on liner performance, highlighting the need to consider such variations in industrial design and quality assurance processes.
Potential opportunities for unmanned aircraft systems (UAS) to offer societal benefits are accompanied by noise impact risks. Accordingly, it is important to develop greater understanding of perception and response to UAS sound. A laboratory listening experiment was undertaken to address this aim by investigating psychoacoustics of UAS sound exposure. The experiment incorporated contextual auditory and soundscape factors by embedding spatially-rendered UAS sounds within urban acoustic environments. The UAS covered varying aircraft designs, operating modes and numbers of flights. The experiment was focussed on determining noticeability and noise annoyance. The results indicate that annoyance responses were influenced by UAS type, operational mode, sound characteristics, quantities of flights, and the ambient acoustic environments in which UAS events occurred. Annoyance also appeared to have associations with personal attitude towards advanced air mobility technology, and with classification of residence area. Noticeability appeared to be influenced by UAS type, operating mode, loudness and ambient environment.
Efficient sound absorption at low and mid frequencies is essential for mitigating anthropogenic noise but remains a scientific and engineering challenge. While porous materials are effective at medium and high frequencies, they typically require considerable thickness to absorb low-frequency sound. Inspired by the Sierpiński carpet, we propose a hierarchical pattern of through-thickness holes in porous materials to enhance low- and mid-frequency absorption. Numerical simulations and impedance tube measurements demonstrate significant performance gains (up to a 46% of improvement in absorption at 500 Hz using 10% less material for a thickness of 80 mm). This enhancement is attributed to the presence of small-scale geometric features that spatially localize the acoustic pressure field, increasing energy dissipation within the porous medium. This hierarchical approach holds promise for developing lightweight, high-performance panels for sound insulation applications, particularly where improved low- and mid-frequency absorption is required without increasing material bulk.
Photo-activated localization microscopy (PALM) has been a game-changer, breaking the diffraction limit in spatial resolution. This study presents the Deactivation Super Resolution (DSR) method, which utilises the deactivation of genetically encodable contrast agents, enabling us to super-resolve and pinpoint individual cells with ultrasound as they navigate through structures which cannot be resolved by conventional B-Mode imaging. DSR takes advantage of Gas Vesicles (GVs), which are air-filled sub-micron particles that have been expressed in genetically engineered bacterial and mammalian cells to produce acoustic contrast. Our experimental results show that DSR can distinguish sub-wavelength microstructures that standard B-mode ultrasound images fail to resolve by super-localising individual mammalian cells. This study provides a proof of concept for the potential of DSR to serve as a super-resolution ultrasound technique for individual cell localisation, opening new horizons in the field.
Acoustic holography can reconstruct desired pressure fields and overcome phase aberrations caused by refractions in heterogeneous media, such as the skull. However, the accuracy of holographic targeting within the brain has not yet been thoroughly evaluated. We sought to characterize the holographic focusing limits for focused and unfocused single-element transducers. Holographic lenses enlarged the focal size by more than 4-fold. Bifocal lenses achieved foci separation of 7-68 mm and focal depths of 7-83 mm. Lenses were tested in silico and in free-field experiments, with RMS errors of 0.03-0.33. Focused transducers were preferable at low F numbers and were better-suited for murine brain targeting. However, planar transducers can focus over larger areas so have higher clinical relevance in the human brain. Finally, simulations with a human skull showed an RMS error < 0.01. This work provides valuable insight into the accuracy of acoustic holography, demonstrating that transducer design is essential for clinical brain applications.
Understanding the heterogeneity of tumor vascular function and oxygenation is key in individualizing treatments, especially with therapies that are ineffective in hypoxic microenvironments. Our previous work has demonstrated that ultrasound-guided photoacoustic imaging (US-PAI)-based blood oxygen saturation (StO2) measurements can be used as a surrogate marker for predicting the regionalized efficacy of photodynamic therapy (PDT). However, monitoring of StO2 during therapy could provide additional insights, specifically informing "on the spot" dosing decisions. In this work, we demonstrate the heterogeneous oxygen consumption during PDT by integrating light delivery fibers with the US-PAI transducer and tested the setup on murine tumor models with vascular-targeting benzoporphyrin derivative (BPD) PDT. Besides mapping dose-dependent oxygen utilization in real time, we also show that areas of reoxygenation post-PDT retain vascular function, confirmed with immunohistochemistry. Our results demonstrate the high potential of US-PAI in heterogenous tumoral oxygenation mapping for online dosimetry of cancer therapies such as PDT.
Sonobiopsy uses focused ultrasound to enrich circulating biomarkers from ultrasound-targeted disease regions, enabling sensitive and spatially targeted molecular diagnosis through liquid biopsy. Since the concept of integrating ultrasound with liquid biopsy was introduced in 2009, the field has advanced to incorporate various ultrasound modalities for applications across a range of diseases. To drive clinical translation of sonobiopsy, it is essential to optimize focused ultrasound techniques, enhance biomarker analysis, and elucidate underlying mechanisms.
Although volumetric ultrasound is limited by cost and availability of 2D arrays, 3D volumes can be reconstructed from 2D slices if transducer position is known, which is not usually the case. Even with position data, existing algorithms for reconstruction are impractical due to their discrete nature that struggles with scale. We propose a 1D array on a programmable motor for scanning and implicit neural representations for continuous reconstruction. Our network's ability to sample at arbitrary positions was compared to classic algorithms, achieving x7.9 performance while maintaining accuracy. Based on these, a reconstruction pipeline was tested on simulated data with 93% accuracy using only 36 B-mode images. This was evaluated in-vivo to measure tumor volumes in mice, with 6.3% mean error. Our findings suggest implicit neural representations can reduce data needed to recreate volumes from 2D slices and replace interpolation methods to enable interactive analysis.
Ultrasound (US) imaging is a fundamental tool in healthcare for the diagnosis of diverse conditions. Wearable, flexible ultrasound patches could expand the scope of US imaging to continuous, at-home monitoring without professional intervention, but require scaling to large numbers of transducer elements. This poses challenges in interconnect density, power consumption, and data bandwidth. To improve interconnect density, we present the first integration of flexible ultrasound transducers with flexible a-IGZO thin-film transistor (TFT) multiplexing electronics. In the Si CMOS readout chip, a new circuit technique cuts front-end power, while a log-delta ADC compresses data efficiently. Our system achieves an 8× reduction in required front-end circuitry and a 42% decrease in front-end power. The data needed to describe the ultrasound image are reduced five-fold, decreasing data transmission power by the same factor. These advances bring the vision of wearable high-density, large-area ultrasound imaging patches for monitoring one step closer.
Passive sound mitigation techniques have garnered attention whether for absorption, isolation, reverberation or new wave phenomena observation. In parallel, a wide range of research has been devoted to active control strategies, which complement passive techniques, particularly for low-frequency. We review the main control techniques related with airborne acoustic wave in the audible regime, emphasizing electrodynamic loudspeakers and piezo-diaphragms, and their applications. We conclude by discussing perspectives in this evolving field.
We present a tunable phononic crystal which can be switched from a mechanically insulating to a mechanically conductive (transmissive) state. Specifically, in our simulations for a phononic lattice under biaxial tension (σ xx = σ yy = 0.01 N m-1), we find a bandgap for out-of-plane phonons in the range of 48.8-56.4 MHz, which we can close by increasing the degree of tension uniaxiality (σ xx/σ yy) to 1.7. To manipulate the tension distribution, we design a realistic device of finite size, where σ xx/σ yy is tuned by applying a gate voltage to a phononic crystal made from suspended graphene. We show that the bandgap closing can be probed via acoustic transmission measurements and that the phononic bandgap persists even after the inclusion of surface contaminants and random tension variations present in realistic devices. The proposed system acts as a transistor for MHz-phonons with an on/off ratio of 105 (100 dB suppression) and is thus a valuable extension for phonon logic applications. In addition, the transition from conductive to isolating can be seen as a mechanical analogue to a metal-insulator transition and allows tunable coupling between mechanical entities (e.g. mechanical qubits).
Magnetic resonance imaging and X-ray computed tomography provide the two principal methods available for imaging the brain at high spatial resolution, but these methods are not easily portable and cannot be applied safely to all patients. Ultrasound imaging is portable and universally safe, but existing modalities cannot image usefully inside the adult human skull. We use in silico simulations to demonstrate that full-waveform inversion, a computational technique originally developed in geophysics, is able to generate accurate three-dimensional images of the brain with sub-millimetre resolution. This approach overcomes the familiar problems of conventional ultrasound neuroimaging by using the following: transcranial ultrasound that is not obscured by strong reflections from the skull, low frequencies that are readily transmitted with good signal-to-noise ratio, an accurate wave equation that properly accounts for the physics of wave propagation, and adaptive waveform inversion that is able to create an accurate model of the skull that then compensates properly for wavefront distortion. Laboratory ultrasound data, using ex vivo human skulls and in vivo transcranial signals, demonstrate that our computational experiments mimic the penetration and signal-to-noise ratios expected in clinical applications. This form of non-invasive neuroimaging has the potential for the rapid diagnosis of stroke and head trauma, and for the provision of routine monitoring of a wide range of neurological conditions.