Ultrasound is an emerging technology for cleaning produce and surface sterilization in the food industry; however, inappropriate intensity of ultrasound treatment may induce stress responses in food spoilage bacteria and enhance their stress resistance. This study systematically evaluated roles of mechanosensitive (MS) channel proteins: MscL and MscS on ultrasonic stress responses in Pseudomonas fluorescens as a typical bacterium leading to food spoilage. The MS channel proteins deletion mutants of P. fluorescens ATCC 13525 were initially constructed, including ΔmscL, ΔmscS, and ΔmscL ΔmscS. MscL was more sensitive to mechanical stimulation of ultrasound which rapidly released intracellular osmotic pressure to alleviate membrane tension damage; however, MscS only provided complementary protection under the ultrasound treatment as proved by Ca2+ fluorescence labeling and antibacterial activity assays. Acoustic pressure distribution and thresholds of MscL (289.45 kPa) and MscS (334.21 kPa) initiating acoustic stress response were determined by COMSOL Multiphysics. Phenotypic assays showed that ultrasound induced membrane potential disruption, ATP depletion, leakage of intracellular components, and oxidative stress response in an intensity-dependent manner with the strongest disturbance observed in the ΔmscL ΔmscS mutant. Ten stress-related genes expression at different ultrasonic intensities further indicated that MscL primarily governed the initial response to ultrasound stimulation, whereas MscS acted as a regulatory factor during sustained and amplified signaling. Both MscL and MscS maintained membrane integrity and cellular homeostasis which jointly regulated the initial transcriptional response including oxidative stress, biofilm formation, and DNA repair. Results of this study would provide theoretical evidences for ultrasound-activation of MS proteins to achieve efficient antibacterial activity.
For half a century, cervical auscultation has been proposed as a classic method for screening aspiration in patients with dysphagia. However, the effects of standardized liquid viscosity and food texture on swallowing sounds are not fully understood due to the lack of uniform standardized bolus preparation. Currently, there is insufficient guidance in the literature for the appropriate liquids and foods for swallowing training, and there is also a lack of studies using acoustic signals to monitor swallowing progress continuously.Objective Based on the international dysphagia diet standardisation initiative (IDDSI) grading, we detect the effects of different levels of food on the swallowing characteristics of healthy older people, which will lay a preliminary research foundation for further early screening and rehabilitation dietary treatment of patients with dysphagia. MethodsThirty healthy older people who met the criteria were selected as the research subjects. Eight kinds of food with different levels were prepared based on the IDDSI grading standard. A throat microphone was used to collect the swallowing sounds of eating different levels of food. RavenPro1.6.0 software was used to perform sound segmentation, annotation and swallowing acoustic index measurement.ResultsThere were significant differences in the duration and peak of the power spectral density of the sound signals of different liquids and foods. As the food level increased, the duration of swallowing also increased. As the food level increased, the peak frequency density of the sound decreased, and there were statistical differences between men and women. Although IDDSI Level 4 is classified as "liquid", its acoustic characteristics approximated those of solid foods in our cohort, suggesting physiological similarity rather than advocating reclassification. It suggests that swallowing acoustic monitoring can be used for clinical screening of dysphagia, and provide a basis for long-term swallowing behavior detection and clinical management.
An unexpected loud sound can elicit an acoustic startle reflex and potentially cause a temporary threshold shift (TTS) for hearing and noise-induced tinnitus for a short period of time. This occurs, for example, with the high-amplitude noise burst that accompanies a bright light and impulsive force when police or military use a "flashbang" grenade to temporarily disable or confuse an adversary. This pilot study evaluated the effects of an unexpected loud acoustic impulse sound on human subject performance in terms of (1) word recognition in noise, and (2) sound localization, both examined in a hemi-anechoic free field with a horizontal plane 32 loudspeaker array. Eleven normal hearing, young, adult, male subjects were tested to determine whether acoustic startle, simulated TTS, simulated tinnitus, and combinations thereof, negatively impact performance. Results indicate that performance in a localization task is substantially affected by all three (and combinations) of these manipulations, whereas word in noise performance is principally degraded by TTS simulation. While these results are preliminary, due to the small sample size, the findings have potential implications for development of non-lethal flashbang grenades, i.e., without a concussive incendiary component.
This study used the sound-induced flash illusion (SIFI) paradigm to investigate how sensory congruency, audiovisual rhythm configurations, and momentary neural states jointly shape perceptual decisions by integrating trial-level EEG and behavioral data within a hierarchical Bayesian regression framework. Thirty healthy adults performed six beep-flash conditions (B0F1, B0F2, B1F1, B1F2, B2F1, and B2F2), encompassing two visual-only baseline conditions, two audiovisually congruent conditions, and two incongruent audiovisual illusion conditions, while response accuracy and reaction time (RT) were recorded together with prestimulus occipital α-band power, auditory and visual N1 amplitudes, and the centro-parietal positivity (CPP) indexing evidence accumulation. We first characterized condition effects and EEG-behavior relationships using repeated-measures ANOVAs and correlations, and then jointly modeled accuracy (Bernoulli likelihood with logit link; effects reported as odds ratios) and RT (Student-t likelihood on logRT; effects interpreted as percentage change in median RT) using trial-level hierarchical Bayesian regression, with model generalizability assessed via Pareto-smoothed importance sampling leave-one-out (PSIS-LOO) cross-validation. Behaviorally, unimodal and congruent conditions yielded faster and more accurate responses, whereas incongruent conditions-particularly the "2 beep-1 flash" configuration-produced elevated illusion rates and markedly prolonged RTs, highlighting the role of causal structure and signal reliability in cross-modal weighting. Neurally, higher trial-level prestimulus α power was reliably associated with reduced odds of a correct response, consistent with lower cortical excitability and a more conservative decision criterion; larger auditory N1 amplitudes predicted faster responses; and CPP slope closely tracked RT in descriptive analyses, consistent with its interpretation as an accumulation-to-bound signal. Model comparisons indicated that experimental conditions accounted for most of the behavioral variance, whereas EEG covariates added limited and non-robust incremental out-of-sample predictive value under the present linear additive specification. Nevertheless, prestimulus α showed a stable negative association with accuracy, and auditory N1 showed a weaker association with response speed, indicating that these neural measures remain theoretically informative even though they did not materially improve overall model generalizability.
暂无摘要（点击查看详情）
Various studies have been conducted in the field of sound event classification, with a specific emphasis on urban environments. However, there has been limited investigation in the realm of forest-oriented sound event classification. We introduce FSM5, a task-specific forest acoustic dataset curated from multiple publicly available sources to support forest sound event classification research, and propose a two-stage framework, ForNet, for effectively classifying critical sound events in forest. Initially, a Convolutional Neural Network (CNN) is used to extract discriminative audio embeddings. In the subsequent phase, ensemble classifiers such as XGBoost and Random Forest are employed for classification. The performance of MFCC, Log-Mel, and Mel spectrogram features is systematically evaluated, and the results indicate that MFCC and Log-Mel features significantly enhance classification performance. The findings indicate that combining handcrafted acoustic representations with CNN-derived embeddings yields improved performance compared to end-to-end CNN classification. The efficacy of deep CNN's feature representation and the discriminative capability of shallow classifiers are evaluated. To showcase the reliability of ForNet, we have also conducted tests on the benchmark: Urbansound8k dataset. ForNet achieves an accuracy of 94% under 10-fold cross-validation on UrbanSound8K, outperforming several state-of-the-art methods, and attains an accuracy of 91.4% using 10-fold cross-validation on the FSM5 dataset.
This in vitro study evaluated the diagnostic accuracy of visual and radiographic assessments for secondary caries (ICDAS-merged) on posterior permanent teeth. Teeth were obtained from a university tooth bank and selected based on the presence of occlusal or proximal restorations. Two trained examiners assessed 296 restored surfaces using ICDAS-merged secondary caries criteria, grouping scores into sound (0), initial (1-2), moderate (3-4), and extensive (5-6). Scores were validated against a histological gold standard using three diagnostic thresholds: (1) sound vs. initial+moderate+extensive; (2) sound+initial vs. moderate+extensive; and (3) sound+initial+moderate vs. extensive. Analyses included Spearman correlation, weighted kappa, and calculation of sensitivity, specificity, and accuracy. After sample evaluation included 129 premolar surfaces and 167 molar surfaces, restored either with amalgam (n = 194) or composite resin (n = 102). Both methods showed substantial inter- and intra-examiner reproducibility (kappa: 0.70-0.80). Correlation with histology was substantial for both visual (=0.73) and radiographic (=0.70) assessments. At the threshold of sound vs. initial+moderate+extensive, specificity was high (visual: 0.85; radiographic: 0.93), while sensitivity was higher for visual (0.80) than for radiographic criteria (0.68). Overall accuracy was high for both methods (visual: 0.83; radiographic: 0.81). Across restorative materials, visual assessment showed consistent accuracy for amalgam and composite restorations. In contrast, radiographic assessment showed lower sensitivity for amalgam (0.61) and composite (0.64) restorations at the same threshold. ICDAS-merged visual criteria demonstrated strong accuracy and reproducibility for detecting secondary caries in restored permanent teeth, regardless of restorative material. Radiographic criteria exhibited high specificity but lower sensitivity, particularly for early lesions, highlighting the value of visual inspection as a reliable method for detecting secondary caries in clinical settings.
Although humans and animals excel at interpreting target sounds in competing noise, sound recognition abilities can vary widely due to statistical characteristics of the interfering background. Yet, how the brain leverages statistical sound information to segregate sounds in noise remains unclear. Here, using many natural auditory textures as background stimuli, we test how the encoding of speech is altered by natural noises in the inferior colliculus (IC) of unanesthetized rabbits. We identify foreground- and background-driven neural response components to sound mixtures and find that the background statistics alter the neural representation of speech in a frequency- and modulation-specific manner. Neural encoding signal-to-noise ratios (SNRs) vary extensively across background categories, frequencies, and modulations, and these influence the encoding of speech independently of the acoustic SNR. This variability is driven by both the background spectrum and modulation statistics, which distort the speech representation and show distinct interference effects in the modulation ranges for rhythm and pitch. Thus, spectrum and modulation statistics critical to speech recognition in noise are reflected in IC neural population activity. These neural response statistics and resulting distortions likely determine the coding fidelity for speech in downstream cortical regions and provide a neural basis for differences in speech intelligibility in real-world noises.
Internal derangement (ID) of the temporomandibular joint (TMJ) is a common condition that is characterized by an abnormal relationship between the articulating disc and the mandibular condyle, often resulting in pain during jaw movements, joint sounds, and restricted mouth opening. Etiology of ID of TMJ is often multifactorial, involving functional, mechanical and biological factors that disturb the normal relationship between the disc and the condyle. Management strategies range widely from conservative therapy, minimally invasive procedures, to invasive surgical procedures. Arthrocentesis is a procedure that is minimally invasive, involving lavage of the superior joint space of the TMJ using needles to remove inflammatory mediators and release adhesions of the disc to the mandibular fossa. Also, there is an increasing interest in minimally invasive regenerative approaches that can heal the injured disc and attachments of the articular disc. Injectable platelet-rich fibrin (i-PRF), which is an autologous blood product, is rich in growth factors and has shown very promising results in reducing inflammation and promoting tissue repair. i-PRF can easily be produced by centrifuging autologous blood at 700 rpm for a very short time of 3 minutes. This case series evaluates the clinical outcomes of arthrocentesis followed by intra-articular i-PRF injections into the upper joint space in three patients diagnosed with TMJ internal derangement. All three patients presented with pain, limited mouth opening, and joint sounds. MRI investigation was done, and found one patient with an anteriorly displacing disc with reduction during opening and closing the mouth, and two patients with an anteriorly displacing disc without the disc reducing during opening and closing the mouth. Clinical parameters like intensity of pain measured by the visual analogue scale (VAS), joint sounds, maximum mouth opening (MMO), functional limitation and patient satisfaction were assessed preoperatively and during follow-up at one week, one month and three months postoperatively. The results demonstrated a significant reduction in pain scores, with mean VAS decreasing from 5.67 preoperatively to 0 at three months post-treatment. Mean maximum mouth opening improved from 32 mm to 39.3 mm, and joint sounds were eliminated in all cases at three months post-intervention. No complications or adverse effects were observed during and after the procedure. All three patients were satisfied with the treatment outcome. These findings suggest that arthrocentesis, which is followed by intra-articular injection of i-PRF, is a safer and effective treatment option for TMJ internal derangement, providing significant symptomatic relief and functional improvement. Further studies with larger sample sizes, post-intervention MRI scan evaluation after three months and longer follow-up periods were recommended to validate these outcomes.
Interaural time and level differences are crucial in sound localization, yet their contributions to sound source segregation and spatial selection remain underspecified. Here, participants completed a spatial auditory selective attention task while we measured hemodynamic activity in the prefrontal cortex and superior temporal gyrus using functional near-infrared spectroscopy. Participants listened to a target sound stream and a simultaneous spatially separated speech stream or white noise masker. Sound streams were spatialized with either 50 μs interaural time differences (ITDs), 500 μs ITDs, naturally occurring interaural level differences (ILDs) from a non-individualized head-related transfer function (HRTF), or broadband 10 dB ILDs. Behavioral results revealed a stronger effect of spatial cues when the masker was speech. Error patterns differed in the two difficult conditions, small ITDs and natural ILDs: Small ITDs produced lower hit rates, while naturally occurring ILDs produced higher false alarm rates. Small ITDs led to greater activity in prefrontal cortex and activity in superior temporal gyrus that was lateralized, greater in the hemisphere contralateral to attentional focus, consistent with previous reports. These results suggest that natural ILDs alone support source segregation even if they are insufficient to cause large shifts in perceived lateralization, explaining high false alarm rates (confusions between target and distractor words). In contrast, small ITDs alone may be insufficient to segregate competing sources, leading to low hit and false alarm rates. Together, these results reveal differences in how ITDs and ILDs contribute to auditory scene analysis and spatial attention.
Natural listening environments are filled with competing sounds. One mechanism that helps overcome this challenge is spatial release from masking (SRM), whereby spatial separation between a target signal and interfering sounds improves perception of the target. SRM has been observed in a wide range of species including songbirds, crocodilians, ferrets, and human adults, suggesting an evolutionarily conserved strategy for listening in noise. However, an important question remains: Is an infant's developing brain able to tap into the same mechanism to cope with the noisy world? To address this question, we recorded electroencephalography (EEG) from 7- and 11-month-old infants (N = 53; 25 female) and adults (N = 20; 12 female) as they listened to natural speech in quiet, collocated noise, and spatially segregated noise. Our results revealed that both infants and adults showed robust cortical tracking of speech across quiet and noisy listening conditions. For adults, spatial separation between the target speaker and distracting talkers led to enhanced cortical tracking of the target speech, consistent with SRM. Infants also showed SRM, with stronger tracking in segregated than collocated noise, although the effect was confined to a fronto-central region rather than broadly distributed across the scalp as in adults. These findings provide the first neurophysiological evidence that, although immature, the developing brain can benefit from spatial cues in the first year of life. The results add new insight into how the infant brain solves a fundamental perceptual problem - identifying a relevant voice in noise - using an evolutionarily grounded mechanism.Significance statement Understanding how the human brain separates relevant speech from competing sounds is a central problem in auditory neuroscience, yet little is known about how this ability emerges early in life. By investigating cortical representation of natural speech in the developing brain, this work provides new insight into the early foundations of human auditory processing. Using electroencephalography in human infants, we show that the neural mechanisms supporting spatial release from masking emerges in infancy, prior to mature language comprehension and attentional control. These results help bridge established accounts of speech-in-noise perception in adults with their developmental origins, providing an objective neural basis for studying how infants navigate complex auditory environments.
The necessity of sentinel lymph node biopsy (SLNB) in early-stage, node-negative (cN0) breast cancer is increasingly questioned in the era of surgical de-escalation and biologically driven treatment strategies. While SLNB has long replaced axillary lymph node dissection to reduce morbidity, it nevertheless carries a measurable risk of complications, including lymphedema, pain, and reduced arm mobility. Recent high-quality evidence suggests that SLNB may be safely omitted in selected patients without compromising oncologic outcomes. The SOUND randomized trial demonstrated that omission of SLNB in women with cT1N0, ultrasound-negative axilla treated with breast-conserving surgery (BCS) and whole-breast irradiation yields non-inferior 5-year distant disease-free survival, with axillary recurrence up to 1%. These findings were reinforced by the large INSEMA randomized trial, which included more than 5,000 cT1-2N0 patients and confirmed that SLNB omission does not adversely affect invasive disease-free survival or overall survival (non inferiority study). INSEMA, as SOUND, predominantly included women older than 50 years and postmenopausal, with approximately 90% of patients presenting with cT1N0 tumor. Importantly, INSEMA showed consistent and clinically significant improvements in patient-reported outcomes, including lower rates of chronic arm morbidity and lymphedema. Then the BOOG 2013-08 trial highlighted quality-of-life benefits at three years for patients who avoided SLNB. However, caution is warranted for underrepresented subgroups such as HER2-positive, triple-negative, high-grade, or lobular cancers. Plus, the rate of macrometastatic sentinel lymph nodes was under 10% for cT1N0 patients but was around 20% for cT2N0 patients. Collectively, these data support a paradigm shift toward selective axillary staging, emphasizing tumor biology, imaging accuracy, and systemic therapy over routine surgical assessment. Evidence indicates that SLNB omission is oncologically safe and functionally advantageous in Luminal cT1N0 patients.
Caries risk begins early in life, yet few studies use intraoral cameras to assess risk in toddlers or examine multiple contributing factors. This study aimed to address this gap and conduct intraoral camera-based assessment of caries and evaluate multiple toddler risk factors, including microbial profiles, oral hygiene practices, and potential maternal transmission. Cross-sectional data from 21 maternal-child dyads included intraoral camera assessments coded with an adapted International Caries Detection and Assessment System, saliva-based microbial tests for Streptococcus mutans (SM) and Lactobacillus (LB), and validated surveys of hygiene practices. Among toddlers (N = 21; mean age 3.9 years), intraoral camera use was well tolerated and produced images adequate for assessment. Reliability was good for sound teeth [Intraclass correlation coefficient (ICC) = 0.80], moderate for initial caries (ICC = 0.72), and poor for moderate-to-severe caries (ICC = 0.13). Most teeth were sound (89.0%), with 9.8% mild and 1.1% moderate decay. SM was detected in 9/20 (45.0%) and LB in 4/21 (19.0%) of toddlers. While 81% brushed twice daily, few flossed (14.3%) or used mouthwash (19%); 42.1% consumed soda at least weekly and 19% ate fast food weekly. More frequent brushing was associated with lower SM (Cramer's V = 0.54; p = .02) and LB (V = 0.48; p = .043). Mothers (N = 21; mean age 32.1) had 38.0% of teeth coded as sound, 45.0% with initial caries, and 14.0% with moderate-to-severe decay; 9/21 (42.9%) tested positive for LB and the same proportion tested positive for SM. Most brushed twice daily (81%), with 47.6% flossing and 38.1% using mouthwash. Maternal and toddler brushing frequency (V = 0.68; p = .02), fast-food intake (V = 0.56; p = .01), and microbial positivity (SM: V = 0.93; p = .001; LB: V = 0.54; p = .02) were strongly related. Intraoral cameras with adapted ICDAS scoring were feasible for assessing caries risk in toddlers. Significant concordance between maternal and child microbial profiles suggests a potential maternal influence on early caries risk. Larger longitudinal studies are needed to clarify transmission pathways and inform prevention strategies.
Speaking is the primary way that humans communicate. This communication is enabled by a production system that can plan and execute unique combinations of speech sounds. Although a distributed network of brain regions has been implicated in speaking, it is unclear how planning and execution of speech are coordinated to produce meaningful sounds. Leveraging the high spatio-temporal resolution of intracranial recordings at different spatial scales, we show distinct neural mechanisms that facilitate speech planning and execution. During planning, different levels of speech units are coded discretely at distinct prefrontal sites. These planned units are then dynamically integrated at various cortical levels to guide subsequent execution. During speech execution, speech motor regions generate continuous sequences that reflect both discrete speech sound units and their transitional properties between units. This rapid neural transition from discrete speech units to motor sequences links speech planning with execution and enables our effortless ability to speak.
Speaking-induced suppression (SIS) refers to the phenomenon that the sounds produced through self-generated speech elicit less EEG activity in the auditory cortex than externally-produced sounds. An analogous phenomenon demonstrates that the activity elicited by an audible syllable is suppressed when it co-occurs with a content-matching syllable produced in inner speech. We have argued that this 'inner SIS' effect is likely underpinned by corollary discharges, similar to those assumed to underlie SIS. However, an alternative possibility is that 'inner SIS' reflects a priming effect, wherein the increased subjective expectation of hearing a certain sound causes a reduction in evoked activity. The current study tested this alternative hypothesis using an expectancy-only version of the task in which participants (N=70, 45 female and 25 male) did not produce inner speech. On each trial, a syllable appeared on-screen (/ba/, /bi/, or /fo/), and was followed by an audible syllable (/ba/ or /bi/). The on-screen syllable either matched (Primed condition) or mismatched (Misprimed condition) the content of the audible syllable. In the Control condition, the on-screen syllable never occurred as the audible syllable. Contrary to previous studies in which participants produced inner speech, there were no significant between-condition differences in N1-amplitude, P2-amplitude, pre-stimulus activity or gamma-band power elicited by the audible syllable. Bayesian analysis indicated a meaningful difference between these findings and prior results from the inner speech task. The present data indicate that a simple cue-based priming manipulation, in the absence of inner speech, is insufficient to reproduce previously reported inner-SIS effects.Significance statement Inner speaking-induced suppression ('inner SIS') refers to the phenomenon that the electrophysiological activity elicited by an external vocalization is reduced when participants concurrently produce the same vocalization in their inner speech. We, and others, have argued that inner SIS is likely caused by corollary-discharge-related mechanisms. However, an alternative possibility is that inner SIS reflects a priming effect, wherein the increased subjective expectation of hearing a certain vocalization causes a reduction in evoked activity. Using a modified version of our standard experimental protocol, we found no evidence for this alternative hypothesis. This result suggests that a priming manipulation, in the absence of inner speech, is insufficient to reproduce previously reported inner-SIS effects.
Listeners discretize the speech signal by assigning sounds to phonetic categories, though there is variability in how individuals accomplish categorization. Having more consistent categorization of sounds may be advantageous for understanding speech-in-noise (SIN). Though, it is unclear how different levels of neural processing in the auditory system reflect these perceptual differences. We recorded brainstem frequency-following responses (FFRs) and cortical event-related potentials (ERPs) while listeners actively labeled vowels along an acoustic-phonetic continuum using a visual analog scale. We computed intertrial consistency of neural responses to index the stability of listeners' neural speech representations across stimulus presentations. We also assessed how faithfully midbrain and cortical responses represented stimulus acoustics using representational dissimilarity matrices (RDMs) computed across all token pairs. Neural RDMs were then compared with acoustic and phonetic category RDMs to assess whether FFRs and ERPs carried gradient vs. categorical information of the speech signal. We found greater behavioral consistency during phoneme labeling was correlated with improved SIN scores. Neurally, we found greater cortical or subcortical consistency predicted greater behavioral consistency. RDMs revealed subcortical responses retained more acoustic details, while cortical responses more closely reflected abstract phoneme categories. Our findings reveal important benefits of perceptual consistency to other domains of speech perception. We find perceptual consistency is driven by more consistent encoding of speech at either a cortical or subcortical level. More consistent sensory processing could provide a more stable readout of the speech signal to higher cortical brain areas which could confer advantages to later perceptual processes downstream.
Operating room noise is a recognized occupational exposure with potential implications for worker health and performance. Most workplace noise risk assessments rely on time-averaged levels, which may obscure short-duration, high-intensity exposure peaks occurring during specific workflow phases. Task-aligned exposure characterization may therefore provide additional value for occupational risk assessment. A field measurement study was conducted in 26 operating rooms under routine clinical conditions. A total of 611 real-time sound level measurements were obtained across 13 predefined surgical workflow stages using a calibrated sound level meter. Stage-specific noise levels were recorded to characterize exposure variability and peak events. Surgical nurses completed the Perceived Stress Scale and the Attention Control Scale immediately before and after procedures. Descriptive statistics, paired comparisons, and correlation analyses were performed. The mean noise level across workflow stages was 66.57 dB, while short-term peak levels reached up to 87 dB. Peak exposures clustered in workflow stages involving equipment movement and patient transfer. Noise variability differed substantially between workflow stages. No statistically significant pre- to post-procedure differences were observed in perceived stress or attention control scores. Associations between stage-averaged noise levels and psychological measures were weak and not statistically significant. Reliance on average noise levels alone may underestimate task-dependent peak exposures in operating rooms. Workflow-aligned measurement approaches improve exposure characterization and may strengthen occupational noise risk assessment. Integrating stage-specific monitoring into routine workplace exposure evaluation may support more precise hazard identification and targeted risk control strategies for operating room personnel.
Acoustic holography has emerged as a disruptive technique for freewheelingly reconstructing the spatial sound field. However, existing mechanisms operate exclusively in homogeneous media, leaving their applications in complex cross-media scenarios beyond reach. Here, a passive meta-projector built by acoustic metamaterials is proposed and experimentally demonstrated for high-fidelity cross-media sound holography. The judiciously designed meta-projector consists of both impedance-matching and mismatching components with a subwavelength thickness and high spatial resolution. Using the full-mode expansion method, we analytically derive the acoustic transfer function of the meta-projector and reveal the existence of the quasi-decoupled point within the considered parameter space. At this point, the passive meta-projector is capable of breaking the intrinsic amplitude-phase locking relationship in conventional anti-reflection materials, thereby achieving synergistic amplitude-phase control of cross-media acoustic waves in a decoupled and ergodic way, which is crucial for high-precision holographic reconstruction. As a proof of concept, a distinct example is demonstrated both numerically and experimentally, wherein an incident underwater simple wave is precisely projected into the desired airborne complex pattern. Moreover, the potential of this meta-projector for contactless cross-media particle assembly is also illustrated. Our proposed meta-projector with multi-dimensional control establishes a new paradigm for extending the acoustic hologram to complicated cross-media systems, holding pivotal significance for diverse applications in complex media that require sophisticated manipulation, including transcranial ultrasound therapy, water-air joint communication, and so on.
Accurate identification of indoor occupant presence is crucial for intelligent building energy management. Traditional monitoring methods are often invasive and lack standardized quantitative indicators. Therefore, this study proposes a non-invasive occupant presence state prediction method based on machine learning. Six machine learning algorithms-Logistic Regression, Decision Tree, k-Nearest Neighbor (KNN), Random Forest, CatBoost, and XGBoost-were evaluated across five building scenarios (hospitals, classrooms, dormitories, offices, and dwelling houses) using core environmental features including air temperature, relative humidity, sound pressure level, illuminance, CO2 concentration, formaldehyde (HCHO), PM2.5, PM1.0, and PM10. A temporally ordered walk-forward evaluation framework was adopted to prevent data leakage, with VIF screening and RFECV for feature selection. The best-performing models achieved mean accuracies ranging from 0.61 (dwelling) to 0.88 (office), with the optimal algorithm varying by scenario. SHAP-based interpretability analysis identified CO2 concentration, illuminance as the most consistently influential predictors, with PM2.5, temperature, and sound contributing in scenario-specific patterns. A leave-one-feature-out sensitivity analysis showed that removing CO2 caused the largest performance drop in the hospital scenario. This study provides a comparative analysis of explainable machine-learning predictors across five single-room building scenarios under a temporally aligned protocol, providing comparative observations that may inform subsequent multi-room and closed-loop studies of smart building management.
Language acquisition is a complex process already influenced by prenatal neural development and auditory experiences. From the onset of the third trimester, fetuses perceive sounds already influencing the fetal brain. The study investigates how the length of intrauterine language exposure, indexed by gestational age (GA), and overall maturation, indexed by birth weight (BW), affect full-term newborns' brain responses to linguistic stimuli. Data from 14 near-infrared spectroscopy studies testing responses to different auditory sound patterns in 192 0- to 7-day-old newborns were pooled together and analyzed to assess the impact of GA and BW on changes in oxygenated hemoglobin (HbO) and deoxygenated hemoglobin (HbR). Results showed that, for HbO, activations in both considered conditions were larger in the temporal than in the frontal areas, irrespective of BW or GA. A similar spatial pattern was observed for HbR, with stronger responses in the temporal compared with frontal regions across conditions. In contrast, when considering effect sizes and reflecting discrimination abilities, these were more strongly associated with BW in the bilateral frontal regions, whereas in the bilateral temporal regions, they were more strongly associated with GA. The findings suggest a differential impact of BW and GA on neural measures of linguistic sensitivity in newborns, reflecting their roles in biological maturation and auditory experience, respectively. Overall, the study suggests that both the length of prenatal experience and maturation play significant roles in shaping newborns' hemodynamic responses.