Multidisciplinary Digital Publishing Institute (Switzerland)
Multidisciplinary Digital Publishing InstituteNot a member yet
1861300 research outputs found
Sort by
FakeVoiceFinder: An Open-Source Framework for Synthetic and Deepfake Audio Detection
No full textAI-based audio generation has advanced rapidly, enabling deepfake audio to reach levels of naturalness that closely resemble real recordings and complicate the distinction between authentic and synthetic signals. While numerous CNN- and Transformer-based detection approaches have been proposed, most adopt a model-centric perspective in which the spectral representation remains fixed. Parallel data-centric efforts have explored alternative representations such as scalograms and CQT, yet the field still lacks a unified framework that jointly evaluates the influence of model architecture, its hyperparameters (e.g., learning rate, number of epochs), and the spectral representation along with its own parameters (e.g., representation type, window size). Moreover, there is no standardized approach for benchmarking custom architectures against established baselines under consistent experimental conditions. FakeVoiceFinder addresses this gap by providing a systematic framework that enables direct comparison of model-centric, data-centric, and hybrid evaluation strategies. It supports controlled experimentation, flexible configuration of models and representations, and comprehensive performance reporting tailored to the detection task. This framework enhances reproducibility and helps clarify how architectural and representational choices interact in synthetic audio detection
The Gut Microbiome of the Goitered Gazelle Enables Plasticity by Responding to Environmental Factors in the Qaidam Basin
No full textThe Qaidam Basin on the Qinghai–Tibet Plateau is an extreme arid environment, posing severe survival challenges. The goitered gazelle (Gazella subgutturosa) is a keystone species in this fragile ecosystem, yet the ecological role of its gut microbiota and its associations with environmental drivers remain poorly understood. We collected fecal samples from gazelles across seven regions of the basin. Metagenomic sequencing was employed to characterize the gut microbiome. Statistical analyses (Mantel tests, multiple regression on matrices, co-occurrence networks) were used to link microbial composition and function with key environmental factors. The gut microbiota was dominated by fiber-degrading phyla (Firmicutes, Bacteroidota) and enriched in metabolic pathways, aligning with a high-fiber diet. Regarding environmental drivers of gut microbial composition variations, isothermality and soil organic carbon were significant predictors, likely via vegetation and environmental inoculation. Regarding environmental drivers of gut microbial function variations, winter solar radiation was uniquely associated with metabolic function without altering microbial composition, suggesting a functional plasticity—the capacity to shift metabolic profiles independently of taxonomic turnover—in response to environmental variation. The gut microbiota of the goitered gazelle exhibits a stable core composition alongside environmentally responsive functional modules. This suggests the microbiome may serve as a significant mediator of host resilience, highlighting adaptation as a dynamic interplay between host, microbiome, and environment. These insights are crucial for microbiome-assisted conservation
Red-Pitaya-Based Frequency Stabilization of 1560-nm Fiber Laser to 780-nm Rubidium Atomic Transition via Single-Pass Frequency Doubling
No full textThe single-step Rydberg excitation of cesium atoms requires a 319 nm ultraviolet laser with a narrow laser linewidth, high frequency stability, and high output power. To meet these requirements, in this work, we construct a high-power, single-frequency UV laser system at this wavelength. In this system, the frequency stabilization of the 1560.492 nm seed laser is critical to the performance of the ultraviolet laser. We employ nonlinear frequency conversion technology, the 1560.492 nm laser is frequency-doubled to 780.246 nm via a single pass through a PPLN crystal, and function integration is realized based on the modular parameter adjustment interface provided by the PyRPL software. Subsequently, the 1560.492 nm laser is stabilized to the D2 hyperfine transition line of Rb-87 atoms using polarization spectroscopy (PS) and radio-frequency-modulated saturation absorption spectroscopy (RF-SAS). A comparative study of these two techniques shows that RF-SAS achieves superior stabilization performance, with the residual frequency fluctuation of the frequency-doubled laser being 1.07 MHz over 30 min. According to frequency doubling theory, the actual residual frequency fluctuation of the 1560.492 nm fundamental-frequency laser can be calculated as 0.535 MHz. Compared with our earlier scheme that utilized an ultra-low-expansion (ULE) optical cavity as a frequency reference, the present scheme eliminates the long-term drift induced by environmental factors. In contrast to frequency stabilization relying on discrete instruments, this integrated scheme significantly reduces the cost, simplifies the system architecture, saves space, and greatly enhances the flexibility and controllability of the system. It therefore provides a reliable and cost-effective solution to ensure the portability and practicability of high-performance UV laser sources. This high-precision frequency stabilization scheme directly guarantees the performance of the 319 nm UV laser, suppressing its linewidth below 10 kHz. Thus, it fully meets the stringent laser linewidth and frequency stability requirements for the single-step Rydberg excitation of cesium atoms and provides a reliable light source foundation for subsequent precision spectroscopic measurements
Study on Preparation and Antibacterial Property of DOMA-SBMA Copolymer Coatings on Stainless Steel Surfaces
No full textA combination of surface wettability and antibacterial performance is highly imperative for construction of antibacterial coatings. In this study, motivated by the antibacterial properties of zwitterionic polymer, mussel-inspired adhesion, and the “grafting to”, a novel DOMA-SBMA copolymer with adhesion and wettability is developed for constructing a bacteriostatic surface. Specifically, the antibacterial coating is prepared via free radical polymerization and grafting to methods on the surface of stainless steel, and characterized by SCA, FTIR, XPS, SEM, and AFM to confirm the modification process. Antibacterial activity evaluation using Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli) shows that the coating presents satisfactory antibacterial performance. The results showed that DOMA-SBMA coating is enough for antibacterial application, with high antibacterial efficiency against E. coli (92.2%) and S. aureus (95.0%). In summary, the bioinspired coating developed here may improve the stability of zwitterionic coatings and provides a simple preparation strategy for constructing antibacterial coatings
CAR Intrinsic Design Pre-Shapes Transcriptional and Metabolic Networks in CAR T Cells
No full textBackground/Objectives: Chimeric antigen receptor (CAR) T cells are a powerful cancer therapy, but their function depends heavily on internal signaling domains and metabolic adaptability. Most studies evaluate CAR behavior upon antigen exposure, yet intrinsic signaling properties may pre-program CAR T cell states even in the absence of stimulation. This study investigates how CAR design and metabolic support shape baseline transcriptional programs, focusing on tonic signaling and NF-κB-related pathways. Methods: We engineered CAR T cells targeting HER2 or GPC3 antigens, incorporating either 4-1BB or CD28 co-stimulatory domains, respectively. A subset of cells was further modified with adenosine deaminase 1 (ADA1) and CD26 to degrade extracellular adenosine and supply inosine, a metabolic strategy termed metabolic refueling (MR). Bulk RNA-seq was performed on resting T cells without antigen stimulation. We analyzed differential gene expression, gene set enrichment (GO, KEGG, Hallmarks), and transcription factor activity (DoRothEA) to assess the impact of CAR design and MR on T cell programming. Results: All CAR T cells exhibited activation of NF-κB–centered inflammatory programs at baseline, indicating tonic signaling. GPC3 CAR T cells showed stronger baseline activation than HER2 CAR T cells. Metabolic refueling amplified these programs without altering their directionality, enhancing inflammatory, survival, and effector modules. Transcription factor activity scores mirrored these trends, highlighting RELA, FOS, and STATs as key regulatory nodes. Conclusions: CAR-intrinsic features, notably co-stimulatory domain choice, define the tonic NF-κB activation tone in resting CAR T cells. Metabolic refueling boosts these baseline states without overstimulation, suggesting it may be especially valuable for weaker CAR constructs. These findings provide a framework for tuning CAR T cell function through combinatorial design strategies targeting signaling and metabolism
Evaluation of Different Controllers for Sensing-Based Movement Intention Estimation and Safe Tracking in a Simulated LSTM Network-Based Elbow Exoskeleton Robot
No full textControl of elbow exoskeletons using muscular signals, although promising for the rehabilitation of millions of patients, has not yet been widely commercialized due to challenges in real-time intention estimation and management of dynamic uncertainties. From a practical perspective, millions of patients with stroke, spinal cord injury, or neuromuscular disorders annually require active rehabilitation, and elbow exoskeletons with precise and safe motion intention tracking capabilities can restore functional independence, reduce muscle atrophy, and lower treatment costs. In this research, an intelligent control framework was developed for an elbow joint exoskeleton, designed with the aim of precise and safe real-time tracking of the user’s motion intention. The proposed framework consists of two main stages: (a) real-time estimation of desired joint angle (as a proxy for movement intention) from High-Density Surface Electromyography (HD-sEMG) signals using an LSTM network and (b) implementation and comparison of three PID, impedance, and sliding mode controllers. A public EMG dataset including signals from 12 healthy individuals in four isometric tasks (flexion, extension, pronation, supination) and three effort levels (10, 30, 50 percent MVC) is utilized. After comprehensive preprocessing (Butterworth filter, 50 Hz notch, removal of faulty channels) and extraction of 13 time-domain features with 99 percent overlapping windows, the LSTM network with optimal architecture (128 units, Dropout, batch normalization) is trained. The model attained an RMSE of 0.630 Nm, R2 of 0.965, and a Pearson correlation of 0.985 for the full dataset, indicating a 47% improvement in R2 relative to traditional statistical approaches, where EMG is converted to desired angle via joint stiffness. An assessment of 12 motion–effort combinations reveals that the sliding mode controller consistently surpassed the alternatives, achieving the minimal tracking errors (average RMSE = 0.21 Nm, R2 ≈ 0.96) and showing superior resilience across all tasks and effort levels. The impedance controller demonstrates superior performance in flexion/extension (average RMSE ≈ 0.22 Nm, R2 > 0.94) but experiences moderate deterioration in pronation/supination under increased loads, while the classical PID controller shows significant errors (RMSE reaching 17.24 Nm, negative R2 in multiple scenarios) and so it is inappropriate for direct myoelectric control. The proposed LSTM–sliding mode hybrid architecture shows exceptional accuracy, robustness, and transparency in real-time intention monitoring, demonstrating promising performance in offline simulation, with potential for real-time clinical applications pending hardware validation for advanced upper-limb exoskeletons in neurorehabilitation and assistive applications
Modeling and Optimization of Extruded Corn Product Fortification
No full textThe present study aimed to model and optimize the fortification of corn-based extruded flips with quinoa flour to improve their nutritional, functional, and sensory quality while maintaining desirable technological properties. Corn flour was partially replaced with quinoa flour at levels of 0, 10, 20, and 30%, and the mixtures were processed using a twin-screw extruder at three screw speeds (350, 500, and 650 rpm). The influence of formulation and mechanical energy input on product quality was evaluated through comprehensive characterization, including bulk density, expansion index, texture, color, chemical composition, mineral profile, amino acid and fatty acid composition, and descriptive sensory attributes. Response surface methodology (RSM) was applied to model the effects of quinoa addition and screw speed on 56 quality responses. The Z-score approach was employed to identify optimal processing conditions. The results showed that from a technological and nutritional perspective, formulations containing 20–30% quinoa processed at medium to high screw speeds (500–650 rpm) provided the most balanced products. Z-score optimization identified that the sample with 20% quinoa extruded at 650 rpm showed a balanced combination of enhanced nutritional characteristics and preserved physical and sensory quality
Influence of Blending Model n-Butanol Alcoholysis Derived Advanced Biofuel Blends with Diesel on the Regulated Emissions from a Diesel Hybrid Vehicle
No full textDecarbonisation of the transport sector, whilst reducing pollutant emissions, will likely involve the utilisation of multiple strategies, including hybridisation and the use of alternative fuels such as advanced biofuels as mandated by the EU. Alcoholysis of lignocellulosic feedstocks, using n-butanol as the solvent, can produce such potential advanced biofuel blends. Butyl blends, consisting of n-butyl levulinate (nBL), di-n-butyl ether, and n-butanol, were selected for this study. Three butyl blends with diesel, two at 10 vol% biofuel and one at 25 vol% biofuel, were tested in a Euro 6b-compliant diesel hybrid vehicle to determine the influence of the blends on regulated emissions and fuel economy. Real Driving Emissions (RDE) were measured for three cold start tests with each fuel using a Portable Emissions Measurement System (PEMS) for carbon monoxide (CO), particle number (PN), and nitrogen oxides (NOX = NO + NO2). When using the butyl blends, there was no noticeable change in vehicle drivability and only a small fuel economy penalty of up to 5% with the biofuel blends relative to diesel. CO, NOX, and PN emissions were below or within one standard deviation of the Euro 6 not-to-exceed limits for all fuels tested. The CO and PN emissions reduced relative to diesel by up to 72% and 57%, respectively. NOX emissions increased relative to diesel by up to 25% and increased with both biofuel fraction and the amount of nBL in that fraction. The CO emitted during the cold start period was reduced by up to 52% for the 10 vol% blends but increased by 25% when using the 25 vol% blend. NOX and PN cold start emissions reduced relative to diesel for all three biofuel blends by up to 29% and 88%, respectively. It is envisaged that the butyl blends could reduce net carbon emissions without compromising or even improving air pollutant emissions, although optimisation of the after-treatment systems may be necessary to ensure emissions limits are met
Risks Associated with Dietary Exposure to Contaminants from Foods Obtained from Marine and Fresh Water, Including Aquaculture
No full textAquatic environments have been a critical source of nutrition for millennia, with wild fisheries supplying protein and nutrients to populations worldwide. A notable shift has occurred in recent decades with the expansion of aquaculture, now representing a fast-growing sector in food production. Aquaculture plays a key role in mitigating the depletion of wild fish stocks and addressing issues related to overfishing. Despite its potential benefits, the sustainability of both wild and farmed aquatic food systems is challenged by anthropogenic pollution. Contaminants from agricultural runoff, industrial discharges, and domestic effluents enter freshwater systems and eventually reach marine environments, where they may be transported globally through ocean currents. Maintaining water quality is paramount to food safety, environmental integrity, and long-term food security. In addition to conventional seafood products such as fish and shellfish, foods such as those derived from microalgae are gaining attention in Western markets for their high nutritional value and potential functional properties. These organisms have been consumed in Asia for generations and are now being explored as sustainable foods and ingredients as an alternative source of protein. Contaminants in aquatic food products include residues of agrochemicals, persistent organic pollutants (POPs) such as dioxins, polychlorinated biphenyls (PCBs), and per- and polyfluoroalkyl substances (PFASs), as well as brominated flame retardants and heavy metals. Public and scientific attention has intensified around plastic pollution, particularly microplastics and nanoplastics, which are increasingly detected in aquatic organisms and are the subject of ongoing toxicological and ecological risk assessments. While the presence of these hazards necessitates robust risk assessment and regulatory oversight, it is important to balance these concerns against the health benefits of aquatic foods, which are rich in omega-3 fatty acids, high-quality proteins, vitamins, and trace elements. Furthermore, beyond direct human health implications, the environmental impact of pollutant sources must be addressed through integrated management approaches to ensure the long-term sustainability of aquatic ecosystems and the food systems they support. This review covers regulatory frameworks, risk assessments, and management issues relating to aquatic environments, including the impact of climate change. It aims to serve as a comprehensive resource for researchers, policymakers, food businesses who harvest food from aquatic systems and other stakeholders
A SWOT Analysis of Modular Construction
No full textModular construction is generally defined as a typical offsite construction approach that can improve environmental sustainability throughout the building project lifecycle. Based on this situation, identifying the strengths, weaknesses, opportunities, and threats (SWOT) while promoting this sustainable construction method effectively during the urbanisation process is essential. Generally, modular construction is a sustainable building approach that can improve project sustainability, considering the environmental, social, economic, and technological aspects. A comprehensive understanding of the basic situation of prefabricated construction is worthwhile to ensure the widespread adoption of this offsite building method. By employing the SWOT analytical framework, this study adopts a literature review approach to conduct the investigation. In terms of the project results, the core strengths of using modular construction include improving environmental sustainability, enhancing management effectiveness, and improving construction safety and quality. The major weaknesses, on the other hand, are a lack of expertise and research, excessively high initial costs, and difficulties in stakeholder coordination. On the other hand, the major opportunities include promoting the SDGs and other policies, the Industrial Revolution 4.0, and urbanisation and building demands. The main threats, however, include substitute construction technologies, imperfect building codes and standards, and a lack of social and market acceptance. Further research can increase the sample size and collect more accurate firsthand data to validate the results of the current investigation, which can increase the effectiveness of promoting modular construction in the targeted regions