Search

This study presents a novel methodology for stratifying post-myocardial infarction patients at risk of ventricular arrhythmias using patient-specific 3D cardiac models derived from late gadolinium enhancement cardiovascular magnetic resonance (LGE-CMR) images. The method integrates imaging and computational simulation with a simplified cellular automaton model, Arrhythmic3D, enabling rapid and accurate VA risk assessment in clinical timeframes. Applied to 51 patients, the model generated thousands of personalized simulations to evaluate arrhythmia inducibility and predict VA risk. Key findings include the identification of slow conduction channels (SCCs) within scar tissue as critical to reentrant arrhythmias and the localization of high-risk zones for potential intervention. The Arrhythmic Risk Score (ARRISK), developed from simulation results, demonstrated strong concordance with clinical outcomes and outperformed traditional imaging-based risk stratification. The methodology is fully automated, requiring minimal user intervention, and offers a promising tool for improving precision medicine in cardiac care by enhancing patient-specific arrhythmia risk assessment and guiding treatment strategies.

The impact of cognitive biases on decision-making in the digital world remains under-explored despite its well-documented effects in physical contexts. This study addresses this gap by investigating the attractiveness halo effect using AI-based beauty filters. We conduct a large-scale online user study involving 2,748 participants who rated facial images from a diverse set of 462 distinct individuals in two conditions: original and attractive after applying a beauty filter. Our study reveals that the same individuals receive statistically significantly higher ratings of attractiveness and other traits, such as intelligence and trustworthiness, in the attractive condition. We also study the impact of age, gender, and ethnicity and identify a weakening of the halo effect in the beautified condition, resolving conflicting findings from the literature and suggesting that filters could mitigate this cognitive bias. Finally, our findings raise ethical concerns regarding the use of beauty filters., Comment: 53 pages, 15 figures, 13 tables

We study the problem of when a periodic square matrix of order $n$ over an arbitrary field $\mathbb{F}$ is decomposable into the sum of a square-zero matrix and a torsion matrix, and show that this decomposition can always be obtained for matrices of rank at least $\frac n2$ when $\mathbb{F}$ is either a field of prime characteristic, or the field of rational numbers, or an algebraically closed field of zero characteristic. We also provide a counterexample to such a decomposition when $\mathbb{F}$ equals the field of the real numbers. Moreover, we prove that each periodic square matrix over any field is a sum of an idempotent matrix and a torsion matrix.

Gallium phosphide (GaP) has recently received considerable attention as a suitable material for building photonic integrated circuits due to its remarkable optical and piezoelectric properties. Usually, GaP is grown epitaxially on III-V substrates to keep its crystallinity and later transferred to silicon wafers for further processing. Here, an alternative promising route for the fabrication of optomechanical (OM) cavities on GaP epitaxially grown on nominally (001)-oriented Si is introduced by using a two-step process consisting of a low-temperature etching of GaP followed by selective etching of the underneath silicon. The low-temperature (-30 $^o$C) during the dry-etching of GaP hinders the lateral etching rate, preserving the pattern with a deviation between the design and the pattern in the GaP layer lower than 5 %, avoiding the complex process of transferring and bonding a GaP wafer to a silicon-on-insulator wafer. To demonstrate the quality and feasibility of the proposed fabrication route, suspended OM cavities are fabricated and experimentally characterized. The cavities show optical quality factors between 10$^3$ and 10$^4$, and localized mechanical resonances at frequencies around 3.1 GHz. Both optical and mechanical resonances are close to those previously reported on crystalline GaP structures. These results suggest a simple and low-cost way to build GaP-based photonic devices directly integrated on industry-standard Si(001) photonic wafers., Comment: 11 pages, 5 figures

Let $\mathbb{F}$ be a field. We show that given any $n$th degree monic polynomial $q(x)\in \mathbb{F}[x]$ and any matrix $A\in\mathbb{M}_n(\mathbb{F})$ whose trace coincides with the trace of $q(x)$ and consisting in its main diagonal of $k$ 0-blocks of order one, with $k

We examine the contribution of tensor modes, in addition to the dominant scalar ones, on the temperature anisotropies of the cosmic microwave background (CMB). To this end, we analyze in detail the temperature two-point angular correlation function $C(\theta)$ from the Planck 2018 dataset, focusing on large angles ($\theta \gtrsim 120^{\circ}$) corresponding to small $\ell$ multipoles. A hierarchical set of infrared cutoffs are naturally introduced to the scalar and tensor power spectra of the CMB by invoking an extra Kaluza-Klein dimension compactifying at about the GUT scale between the Planck epoch and the start of inflation. We associate this set of lower scalar and tensor cutoffs with the parity of the multipole expansion of the $C(\theta)$ function. By fitting the Planck 2018 data we compute the multipole coefficients thereby reproducing the well-known odd-parity preference in angular correlations seen by all three satellite missions COBE, WMAP and Planck. Our fits improve significantly once tensor modes are included in the analysis, hence providing a hint of the imprints of primordial gravitational waves on the temperature correlations observed in the CMB today. To conclude we suggest a relationship between, on the one hand, the lack of (positive) large-angle correlations and the odd-parity dominance in the CMB and, on the other hand, the effect of primordial gravitational waves on the CMB temperature anisotropies., Comment: 12 pages, 2 figures

Federated Learning (FL) has been proposed as a privacy-preserving solution for machine learning. However, recent works have reported that FL can leak private client data through membership inference attacks. In this paper, we show that the effectiveness of these attacks on the clients negatively correlates with the size of the client's datasets and model complexity. Based on this finding, we study the capabilities of model-agnostic Federated Learning to preserve privacy, as it enables the use of models of varying complexity in the clients. To systematically study this topic, we first propose a taxonomy of model-agnostic FL methods according to the strategies adopted by the clients to select the sub-models from the server's model. This taxonomy provides a framework for existing model-agnostic FL approaches and leads to the proposal of new FL methods to fill the gaps in the taxonomy. Next, we analyze the privacy-performance trade-off of all the model-agnostic FL architectures as per the proposed taxonomy when subjected to 3 different membership inference attacks on the CIFAR-10 and CIFAR-100 vision datasets. In our experiments, we find that randomness in the strategy used to select the server's sub-model to train the clients' models can control the clients' privacy while keeping competitive performance on the server's side.

The availability of digital twins for the cardiovascular system will enable insightful computational tools both for research and clinical practice. This, however, demands robust and well defined models and methods for the different steps involved in the process. We present a vessel coordinate system (VCS) that enables the unanbiguous definition of locations in a vessel section, by adapting the idea of cylindrical coordinates to the vessel geometry. Using the VCS model, point correspondence can be defined among different samples of a cohort, allowing data transfer, quantitative comparison, shape coregistration or population analysis. Furthermore, the VCS model allows for the generation of specific meshes (e.g. cylindrical grids, ogrids) necessary for an accurate reconstruction of the geometries used in fluid simulations. We provide the technical details for coordinates computation and discuss the assumptions taken to guarantee that they are well defined. The VCS model is tested in a series of applications. We present a robust, low dimensional, patient specific vascular model and use it to study phenotype variability analysis of the thoracic aorta within a cohort of patients. Point correspondence is exploited to build an haemodynamics atlas of the aorta for the same cohort. The atlas originates from fluid simulations (Navier-Stokes with Finite Volume Method) conducted using OpenFOAMv10. We finally present a relevant discussion on the VCS model, which covers its impact in important areas such as shape modeling and computer fluids dynamics (CFD).

High-order Graph Neural Networks (HO-GNNs) have been developed to infer consistent latent spaces in the heterophilic regime, where the label distribution is not correlated with the graph structure. However, most of the existing HO-GNNs are hop-based, i.e., they rely on the powers of the transition matrix. As a result, these architectures are not fully reactive to the classification loss and the achieved structural filters have static supports. In other words, neither the filters' supports nor their coefficients can be learned with these networks. They are confined, instead, to learn combinations of filters. To address the above concerns, we propose Diffusion-jump GNNs a method relying on asymptotic diffusion distances that operates on jumps. A diffusion-pump generates pairwise distances whose projections determine both the support and coefficients of each structural filter. These filters are called jumps because they explore a wide range of scales in order to find bonds between scattered nodes with the same label. Actually, the full process is controlled by the classification loss. Both the jumps and the diffusion distances react to classification errors (i.e. they are learnable). Homophiliation, i.e., the process of learning piecewise smooth latent spaces in the heterophilic regime, is formulated as a Dirichlet problem: the known labels determine the border nodes and the diffusion-pump ensures a minimal deviation of the semi-supervised grouping from a canonical unsupervised grouping. This triggers the update of both the diffusion distances and, consequently, the jumps in order to minimize the classification error. The Dirichlet formulation has several advantages. It leads to the definition of structural heterophily, a novel measure beyond edge heterophily. It also allows us to investigate links with (learnable) diffusion distances, absorbing random walks and stochastic diffusion.

The catalyst-free synthesis of graphene on dielectrics prevents the damage induced by the transfer process. Although challenging, to master this synthesis would boost the integration of graphene on consumer electronics since defects hinder its optoelectronic properties. In this work, the influence of the different surface terminations of c-plane sapphire substrates on the synthesis of graphene via plasma-enhanced chemical vapour deposition (PECVD) is studied. The different terminations of the sapphire surface are controlled by a plasma etching process. A design of experiments (DoE) procedure was carried out to evaluate the major effects governing the etching process of four different parameters: i.e. discharge power, time, pressure and gas employed. In the characterization of the substrate, two sapphire surface terminations were identified and characterized by means of contact angle measurements, being a hydrophilic (hydrophobic) surface the fingerprint of an Al- (OH-) terminated surface, respectively. The defects within the synthesized graphene were analysed by Raman spectroscopy. Notably, we found that the ID/IG ratio decreases for graphene grown on OH-terminated surfaces. Furthermore, two different regimes related to the nature of graphene defects were identified and depending on the sapphire terminated surface are bound either to vacancy or boundary like defects. Finally, studying the density of defects and the crystallite area, as well as their relationship with the sapphire surface termination paves the way for increasing the crystallinity of the synthesized graphene., Comment: 27 pages, 6 figures

For any $n\ge 2$ and fixed $k\ge 1$, we give necessary and sufficient conditions for an arbitrary nonzero square matrix in the matrix ring $\mathbb{M}_n(\mathbb{F})$ to be written as a sum of an invertible matrix $U$ and a nilpotent matrix $N$ with $N^k=0$ over an arbitrary field $\mathbb{F}$., Comment: arXiv admin note: text overlap with arXiv:2301.06106

For $n\ge 2$ and fixed $k\ge 1$, we study when a square matrix $A$ over an arbitrary field $\mathbb{F}$ can be decomposed as $T+N$ where $T$ is a torsion matrix and $N$ is a nilpotent matrix with $N^k=0$. For fields of prime characteristic, we show that this decomposition holds as soon as the characteristic polynomial of $A\in \mathbb{M}_{n}(\mathbb{F})$ is algebraic over its base field and the rank of $A$ is at least $\frac nk$, and we present several examples that show that the decomposition does not hold in general. Furthermore, we completely solve this decomposition problem for $k=2$ and nilpotent matrices over arbitrary fields (even over division rings). This somewhat continues our recent publications in Lin. \& Multilin. Algebra (2023) and Internat. J. Algebra \& Computat. (2022) as well as it strengthens results due to Calugareanu-Lam in J. Algebra \& Appl. (2016).

Human perception, memory and decision-making are impacted by tens of cognitive biases and heuristics that influence our actions and decisions. Despite the pervasiveness of such biases, they are generally not leveraged by today's Artificial Intelligence (AI) systems that model human behavior and interact with humans. In this theoretical paper, we claim that the future of human-machine collaboration will entail the development of AI systems that model, understand and possibly replicate human cognitive biases. We propose the need for a research agenda on the interplay between human cognitive biases and Artificial Intelligence. We categorize existing cognitive biases from the perspective of AI systems, identify three broad areas of interest and outline research directions for the design of AI systems that have a better understanding of our own biases., Comment: 14 pages, 1 figure; Accepted for presentation at the AAAI Fall Symposium 2022 on Thinking Fast and Slow and Other Cognitive Theories in AI. Corrected typos; v3: Updated figure 1, added table 4

Federated learning (FL) has been proposed as a privacy-preserving approach in distributed machine learning. A federated learning architecture consists of a central server and a number of clients that have access to private, potentially sensitive data. Clients are able to keep their data in their local machines and only share their locally trained model's parameters with a central server that manages the collaborative learning process. FL has delivered promising results in real-life scenarios, such as healthcare, energy, and finance. However, when the number of participating clients is large, the overhead of managing the clients slows down the learning. Thus, client selection has been introduced as a strategy to limit the number of communicating parties at every step of the process. Since the early na\"{i}ve random selection of clients, several client selection methods have been proposed in the literature. Unfortunately, given that this is an emergent field, there is a lack of a taxonomy of client selection methods, making it hard to compare approaches. In this paper, we propose a taxonomy of client selection in Federated Learning that enables us to shed light on current progress in the field and identify potential areas of future research in this promising area of machine learning., Comment: 17 pages, 3 figures, 1 appendix, accepted to TMLR

We interpret the lack of large-angle temperature correlations and the apparent even-odd parity imbalance, observed in the cosmic microwave background by COBE, WMAP and Planck satellite missions, as a possible stringy signal ultimately stemming from a composite inflaton field (e.g. a fermionic condensate). Based on causality arguments and a Fourier analysis of the angular two-point correlation function, two infrared cutoffs $k_{\rm min}^{\rm even,odd}$ are introduced in the CMB power spectrum associated, respectively, with periodic and antiperiodic boundary conditions of the fermionic constituents (echoing the Neveu-Schwarz-Ramond model in superstring theory), without resorting to any particular model., Comment: 11 pages, 7 figures

This paper proposes an interactive system for mobile devices controlled by hand gestures aimed at helping people with visual impairments. This system allows the user to interact with the device by making simple static and dynamic hand gestures. Each gesture triggers a different action in the system, such as object recognition, scene description or image scaling (e.g., pointing a finger at an object will show a description of it). The system is based on a multi-head neural network architecture, which initially detects and classifies the gestures, and subsequently, depending on the gesture detected, performs a second stage that carries out the corresponding action. This multi-head architecture optimizes the resources required to perform different tasks simultaneously, and takes advantage of the information obtained from an initial backbone to perform different processes in a second stage. To train and evaluate the system, a dataset with about 40k images was manually compiled and labeled including different types of hand gestures, backgrounds (indoors and outdoors), lighting conditions, etc. This dataset contains synthetic gestures (whose objective is to pre-train the system in order to improve the results) and real images captured using different mobile phones. The results obtained and the comparison made with the state of the art show competitive results as regards the different actions performed by the system, such as the accuracy of classification and localization of gestures, or the generation of descriptions for objects and scenes.

A mathematical approach to solve the porthole die design problem is achieved by statistical analysis of a large amount of geometric data of successful porthole die designs. Linear and logarithmic regression are used to analyse geometrical data of 596 different ports from 88 first trial dies. Non-significant variables or high correlated variables are discarded according to knowledge of the extrusion process and statistical criteria. This paper focuses on a validation model for a typical case of porthole dies: four cavities and four ports per cavity dies. This mathematical formulation is a way of summarizing in a single expression the experience accumulated in a large number of designs over time. A broad way of research is open to generalise this model or extend it to other types of porthole dies.

The analysis of azimuthal correlations in multiparticle production can be useful to uncover the existence of new physics beyond the Standard Model, e.g., Hidden Valley, in $e^+e^-$ annihilation at high energies. In this paper, based on previous theoretical studies and using PYTHIA8 event generator, it is found that both azimuthal and rapidity long-range correlations are enhanced due to the presence of a new stage of matter on top of the QCD partonic cascade. Ridge structures, similar to those observed at the LHC, show up providing a possible signature of new physics at future $e^+e^-$ colliders., Comment: 8 pages, 7 figures

In order to find a suitable expression of an arbitrary square matrix over an arbitrary finite commutative ring, we prove that every such a matrix is always representable as a sum of a potent matrix and a nilpotent matrix of order at most two when the Jacobson radical of the ring has zero-square. This somewhat extends results of ours in Lin. & Multilin. Algebra (2021) established for matrices considered on arbitrary fields. Our main theorem also improves on recent results due to Abyzov et al. in Mat. Zametki (2017), \v{S}ter in Lin. Algebra & Appl. (2018) and Shitov in Indag. Math. (2019)., Comment: 10 pages

In this paper, analogies between multiparticle production in high-energy collisions and the time evolution of the early universe are discussed. A common explanation is put forward under the assumption of an unconventional early state: a rapidly expanding universe before recombination (last scattering surface), followed by the CMB, later evolving up to present days, versus the formation of hidden/dark states in hadronic collisions followed by a conventional QCD parton shower yielding final-state particles. In particular, long-range angular correlations are considered pointing out deep connections between the two physical cases potentially useful for the discovery of new physics., Comment: 9 pages, 5 figures