239700 research outputs found
Sort by
The Manchester driving behavior questionnaire (DBQ) integrating health and technology factors: The DBQ 2025 update with translations in 11 languages
No full textThe Manchester Driving Behavior Questionnaire is a widely-used instrument to assess driving behavior but has become outdated, omitting items addressing modern communication technologies and health-related issues that impact today's drivers. This study updated the original instrument with relevant health and technology items. The instrument updating process involved 5 steps. First, a literature review identified new and relevant items from existing instruments. Second, an international team with expertise in driving behavior, including the original instrument developer, reviewed and suggested revisions, assessed face validity, and recommend changes iteratively. Third, content validity was evaluated via computation of a content validity index (CVI) and content validity ratios (CVR). Fourth, the updated English version was translated into 11 languages by a global team. Finally, reliability of the Persian version was assessed via Cronbach's alpha and intra-class correlation coefficient. The literature review led to new items addressing topics such as smartphone use and health conditions that may impact safe driving. Experts refined these items iteratively, creating an updated MDBQ with the 27 original items and 12 newly-developed ones. Content validity was assessed, yielding average CVIs of 0.95 and CVRs of 0.87. The questionnaire was then translated into 11 languages (Arabic, Azerbaijani Turkish, Chinese, French, German, Italian, Persian, Spanish, Swedish, Tagalog, and Turkish) according to standardized methods. The updated instrument retained all components of the original MDBQ, preserving comparability with existing data while also assessing contemporary topics. It provides a comprehensive instrument to assess driving behaviors and is recommended for use in research, policy, public health, and intervention development
The evolutionary history of ultra-compact accreting binaries
No full textContext. AM Canum Venaticorum (AM CVn) stars are ultra-compact binary systems composed of a white dwarf primary accreting from a hydrogen-deficient donor. They play a crucial role in astrophysics as potential progenitors of Type Ia supernovae and as laboratories for gravitational wave studies. However, their formation and evolutionary history remain incomplete. Three formation channels have been discussed in the literature: the white dwarf, He-star, and cataclysmic variable channels.
Aims. The chemical composition of the accretor atmosphere reflects the material transferred from the donor. In this work we took the first accurate measurements of the fundamental parameters of the accreting white dwarf in ZTF J225237.05−051917.4, including the abundances of key elements such as carbon, nitrogen, and silicon, by analysing ultraviolet spectra obtained with the Hubble Space Telescope (HST). These measurements provide new insight into the evolutionary history of the system and, together with existing optical observations, establish it as a benchmark to develop our pipeline, paving the way for its application to a larger sample of AM CVn systems.
Methods. We determined the binary parameters through photometric analysis and constrained the atmospheric parameters of the white dwarf accretor, including its effective temperature, surface gravity, and chemical abundances, by fitting the HST ultraviolet spectrum with synthetic spectral models. We then inferred the system’s formation channel by comparing the results with theoretical evolutionary models.
Results. According to our measurements, the accretor’s effective temperature (Teff) is 23 300 600 K and the surface gravity (log g) is 8.4 0.3, which imply an accretor mass (MWD) of 0.86 0.16 M. We find a high nitrogen-to-carbon abundance ratio by mass of > 153.
Conclusions. The accretor is significantly hotter than previous estimates based on simplified blackbody fits to the spectral energy distribution, underscoring the importance of detailed spectral modelling for accurately determining system parameters. Our results show that ultraviolet spectroscopy is well suited to constraining the formation channels of AM CVn systems. Of the three proposed formation channels, the He-star channel can be excluded given the high nitrogen-to-carbon ratio. Our results are consistent with both the white dwarf and cataclysmic variable channels
Wideband MIMO radar beampattern shaping in spectrally dense environments
No full textWideband MIMO radar beampattern shaping with Constant Modulus Constraints (CMCs) in spectrally dense environments is critical for future 6G networked sensing technology. Existing methods minimize the weighted function of wideband MIMO radar beampattern matching Mean Square Error (MSE) and the Energy Spectral Density (ESD) of Spatial Spectral Nulling (SSN) region; however, achieving precise ESD control remains a challenge. To address this, we minimize the beampattern matching MSE under CMCs and precise SSN Constraints (SSNCs). The non-convex nature of the CMCs and multiple SSNCs lead to a non-convex Quadratic-Constrained Quadratic Programming (QCQP) problem. To solve the problem, we propose a novel Manifold-Based Exact Penalty (MBEP) method. First, we construct the Complex Circular Manifold (CCM) to satisfy the CMCs and reformulate the SSNCs as an exact penalty function, thereby transforming the problem into an unconstrained optimization problem on the CCM. Subsequently, a Simplified Quasi-Newton (SQN) method is developed to optimize the problem on the CCM. Finally, the penalty factor is adaptively updated to improve the optimization process. Compared with existing methods: 1) the proposed method achieves precise control of the ESD level in the SSN region; and 2) the ESD in the SSN region is reduced by 8.8 dB, while the beampattern matching MSE is decreased by 0.02 dB
Life cycle assessment of polyol synthesis from waste cooking oil versus virgin rapeseed oil: Toward sustainable polyurethane foam production
No full textPolyurethanes (PU) are lightweight polymeric materials with a broad range of properties and applications, and its market has shown consistent growth. Currently, PU are primarily synthesized from petroleum-derived feedstocks through a polyaddition reaction between isocyanates (R-NCO) and polyols (R-OH). For a sustainable development, the adoption of renewable raw materials and innovative manufacturing processes has become an urgent priority, promoting the replacement of the main fossil-based precursors with bio-based compounds. This study investigated the life cycle environmental impacts generated by the production of 1 ton of polyol from waste cooking oil (WCO) at midpoint and endpoint level. A gate-to-gate approach was adopted, accounting for material and energy use in the following steps: WCO collection, transportation, pretreatment and polyol production via epoxidation and ring-opening reactions. The cultivation and use of virgin rapeseed oil (VRO) as feedstock for bio-polyol production was considered as reference. According to ReCiPe 2016 Midpoint (H), WCO-derived bio-polyol production outperformed the VRO route in 17 out of 18 environmental indicators, reducing CO2-equivalent emissions by 15%. Endpoint and sensitivity analyses confirmed the results obtained, highlighting that the combined recovery of chemicals and catalyst used for epoxidation and ring-opening reactions could further increase the environmental benefits in each impact category (e.g., up to 42% reduction in CO2-equivalent emissions). The innovative use of WCO in bio-polyol production has proven to be a valid alternative to VRO, improving process sustainability and increasing land availability for agricultural purposes. © 2025 The Author(s)
From simple to complex: multiscale modeling of metal-mediated protein interactions to uncover mechanistic insights into the action of inorganic drugs
No full textThis Perspective article explores the crucial role of computational and experimental models in protein metalation studies which is at the heart of advancing inorganic medicinal drugs. The intricate world of protein-inorganic drug interactions poses a significant challenge to computational and experimental characterisation. Indeed, the vast conformational landscapes and dynamic behaviours of proteins demand sophisticated modelling strategies to accurately predict drug behaviour and mechanism of action. Using selected examples, this article demonstrates how research progresses from simple models to increasingly complex systems, thereby facilitating the acquisition of comprehensive mechanistic insights. An effective and pragmatic strategy for navigating this complexity is to deliberately use simple models as steppingstones towards understanding more intricate protein metalation phenomena. This hierarchical modelling paradigm enables researchers to systematically develop an understanding of processes ranging from fundamental atomic interactions to the entire protein-drug dynamic, balancing computational and experimental feasibility with deep mechanistic insight
Experimental validation of deep Koopman MPC for real-time pasteurization unit control
No full textThis paper presents a novel deep Koopman Model Predictive Control framework for energy-intensive processes, addressing the critical challenge of real-time nonlinear control in pasteurization systems. The proposed approach employs neural networks to learn lifting functions that transform nonlinear system dynamics into a linear representation in the lifted space, enabling computationally efficient MPC implementation. A key innovation is the introduction of a lifted state correction mechanism that compensates for the mismatch between unmeasurable lifted states and real measurements, significantly improving control accuracy and practical implementability. Experimental validation on a laboratory-scale pasteurization unit demonstrates that the proposed deep Koopman MPC framework achieves 30 % improvement in control performance compared to conventional subspace identification methods, while maintaining real-time execution within 10ms on standard hardware. The enhanced control accuracy translates to reduced product waste and improved energy efficiency in pasteurization operations without additional computational overhead. This work represents one of the first experimental implementations of Koopman MPC in chemical process control and energy-intensive applications, establishing its practical feasibility for industrial deployment
Fuzzy Decision Trees for Explainable Brain Tumor Classification: A Comparative Study with Deep Neural Networks and Classical Binary Decision Trees
No full textBrain Tumor Classification (BTC) using Magnetic Resonance Imaging (MRI) has achieved remarkable progress through Deep Learning (DL) models, particularly Convolutional Neural Networks (CNNs). However, the opaque nature of these models raises concerns regarding explainability, which is critical in clinical decision support. To address this, most research has focused on post-hoc Explainable AI (XAI) methods that provide after-the-fact interpretations of CNN predictions. In contrast, this work investigates an inherently explainable alternative based on Fuzzy Decision Trees (FDTs), which combine the interpretability of rule-based reasoning with the expressiveness of fuzzy logic. Moreover, we enhance model transparency by integrating radiomic features that capture clinically meaningful tumor characteristics such as shape, texture, and intensity. To the best of our knowledge, this is among the first studies to apply FDTs to brain tumor classification from MRI, explicitly coupling radiomics with multi-way FDT architectures. We perform a comprehensive evaluation comparing FDTs against four state-of-the-art CNNs, namely ConvNeXt, ResNet18, ResNet50, and EfficientNetB0, as well as classical binary Decision Trees (DTs). We provide an explicit analysis of the trade-off between accuracy, complexity, and interpretability of the models. Results show that FDTs achieve competitive performance (overall F1-score ≈ 0.84) compared to the best CNN baseline (ResNet50, F1-score ≈ 0.86), while offering substantially higher explainability and interpretability. Overall, this study demonstrates that FDTs can bridge the gap between accuracy and explainability, offering a viable explainable-by-design alternative to deep learning in medical imaging. Future work will focus on validating this generalizability across different imaging domains and dataset variations
Generalization of Repetitiveness Measures for Two-Dimensional Strings
No full textThe problem of detecting and measuring the repetitiveness of one-dimensional strings has been extensively studied in data compression and text indexing. Our understanding of these issues has been significantly improved by the introduction of the notion of string attractor (Kempa and Prezza 2018) and by the results showing the relationship between attractors and other measures of compressibility. When the input data are structured in a non-linear way, as in two-dimensional strings, inherent redundancy often offers an even richer source for compression. However, systematic studies on repetitiveness measures for two-dimensional strings are still scarce. In this paper, we extend to two or more dimensions the main measures of complexity introduced for one-dimensional strings. We distinguish between the measures δ and γ, defined in terms of the substrings of the input, and the measures g, grl, and b, which are based on copy-paste mechanisms. We study the properties and mutual relationships between these two classes and we show that the two classes become incomparable for d-dimensional inputs as soon as d⩾2. Moreover, we show that our grammar-based representation of a d-dimensional string of size N enables direct access to any symbol in O(logN) time. We also compare our measures for two-dimensional strings with the 2D Block Tree data structure (Brisaboa et al., Comput. J. 67(1), 391–406, 2024) and provide some insights for the design of future effective two-dimensional compressors. A preliminary version of this paper appeared in the proceedings of the conference SPIRE 2024
Generative AI as a New Assistive Technology for Web Interaction
No full textFor users who are unfamiliar with technology or rely on assistive tools such as screen readers, interacting with a web page can be challenging. Ensuring a seamless experience requires a well-designed user interface (UI) that prioritizes accessibility and usability. However, achieving this target demands specialized expertise from developers and can involve significant effort. In this context, Generative Artificial Intelligence (GAI) has become a valuable aid for improving access to information and facilitating interaction with web interfaces. To effectively enhance user interaction—such as accessing services or specific functionalities—AI-driven tools must first be capable of understanding the structure and content of a web page. This study investigates if GAIs can be exploited to assist the user when navigating through a website, describing the site contents, explaining the interface structure and interactive elements, and suggesting actions or procedures to follow to perform a certain task or accomplish a specific goal. This kind of assistive technology can benefit not only visually impaired people but also persons with cognitive impairment and, more generally, people that are not “skilled” with modern web applications, like seniors. Specifically, thirteen popular websites were analyzed by asking Copilot one hundred questions. Results suggest that GAIs have the potential to assist people in web tasks. However, limitations have still been detected, with 20% of completely erroneous answers received from the navigation and interaction questions and 15% for those related to structure, mainly detected in pages having scarce accessibility and sites having a complex HTML structure, respectively