Search Results (50)

Background: Tuberculosis (TB) is a significant global health issue, particularly in resource-limited regions where radiological expertise is constrained. This project aims to develop a scalable deep learning system that safeguards privacy and achieves high accuracy in the early identification of tuberculosis using chest X-ray images. The objective is to implement federated learning with an adaptive-weighted ensemble optimised by a Genetic Algorithm (GA) to address the challenges of centralised training and single-model approaches. Method: We developed an ensemble learning method that combines multiple locally trained models to improve diagnostic consistency and reduce individual-model bias. An optimisation system that autonomously selected the optimal ensemble weights determined each model’s contribution to the final decision. A controlled augmentation process was employed to enhance the model’s robustness and reduce the likelihood of overfitting by introducing realistic alterations to appearance, geometry, and acquisition conditions. Federated learning facilitated collaboration among universities for training while ensuring data privacy was maintained during the establishment of the optimal ensemble at each location. In this system, just model parameters were transmitted, excluding patient photographs. This enabled the secure amalgamation of global data without revealing sensitive clinical information. Standard diagnostic metrics, including accuracy, sensitivity, precision, F1 score, AUC, and confusion matrices, were employed to evaluate the model’s performance. Results: The proposed federated, GA-optimized ensemble demonstrated superior performance compared with individual models and fixed-weight ensembles. The system achieved 98% accuracy, 97% F1 score, and 0.999 AUC, indicating highly reliable discrimination between TB-positive and typical cases. Federated learning preserved model robustness across heterogeneous data sources, while ensuring complete patient privacy. Conclusions: The proposed federated, GA-optimized ensemble achieves highly accurate and robust early tuberculosis detection while preserving patient privacy across distributed clinical sites. This scalable framework demonstrates strong potential for reliable AI-assisted TB screening in resource-limited healthcare settings. Full article

The application of unmanned vehicles in civilian and military fields is increasingly widespread. Traditionally, unmanned vehicles primarily rely on Global Positioning Systems (GPSs) for positioning; however, GPS signals can be limited or completely lost in conditions such as building obstructions, indoor environments, or electronic interference. In addition, countries are actively developing GPS jamming and deception technologies for military applications, making precise positioning and navigation of unmanned vehicles in GPS-denied or constrained environments a critical issue that needs to be addressed. In this work, authors propose a method based on Visual–Inertial Odometry (VIO), integrating the extended Kalman filter (EKF), an Inertial Measurement Unit (IMU), optical flow, and feature matching to achieve drone localization in GPS-denied environments. The proposed method uses the heading angle and acceleration data obtained from the IMU as the state prediction for the EKF, and estimates relative displacement using optical flow. It further corrects the optical flow calculation errors through IMU rotation compensation, enhancing the robustness of visual odometry. Additionally, when re-selecting feature points for optical flow, it combines a KAZE feature matching technique for global position correction, reducing drift errors caused by long-duration flight. The authors also employ an adaptive noise adjustment strategy that dynamically adjusts the internal state and measurement noise matrices of the EKF based on the rate of change in heading angle and feature matching reliability, allowing the drone to maintain stable positioning in various flight conditions. According to the simulation results, the proposed method is able to effectively estimate the flight trajectory of drones without GPS. Compared to results that rely solely on optical flow or feature matching, it significantly reduces cumulative errors. This makes it suitable for urban environments, forest areas, and military applications where GPS signals are limited, providing a reliable solution for autonomous navigation and positioning of drones. Full article

Background: The pathological response to neoadjuvant chemotherapy (NAC) is an established predictor of long-term outcomes in breast cancer. However, conventional binary assessment based solely on pathological complete response (pCR) fails to capture prognostic heterogeneity across molecular subtypes. This study aimed to develop an interpretable deep learning model that integrates multiple clinical and pathological variables to predict both recurrence and metastasis development following NAC treatment. Methods: We conducted a retrospective analysis of 832 breast cancer patients who received NAC between 2013 and 2022. The analysis incorporated five key variables: tumor size changes, nodal status, Ki-67 index, Miller–Payne grade, and molecular subtype. A Multi-Layer Perceptron (MLP) model was implemented on the PyTorch platform and systematically benchmarked against SVM, Random Forest, and XGBoost models using five-fold cross-validation. Model performance was assessed by calculating the area under the curve (AUC), accuracy, precision, recall, and F1-score, and by analyzing confusion matrices. Results: The MLP model achieved AUC values of 0.86 (95% CI: 0.82–0.93) for HER2-positive cases, 0.82 (95% CI: 0.70–0.92) for triple-negative cases, and 0.76 (95% CI: 0.66–0.82) for HR+/HER2-negative cases. SHAP analysis identified post-NAC tumor size, Ki-67 index, and Miller–Payne grade as the most influential predictors. Notably, patients who achieved pCR still had a 12% risk of developing recurrence, highlighting the necessity for ongoing risk assessment beyond binary response evaluation. Conclusions: The proposed deep learning system provides precise and interpretable risk assessment for NAC patients, facilitating individualized treatment approaches and post-treatment monitoring plans. Full article

Accurate target recognition and localization remain significant challenges for robotic fruit harvesting in unstructured orchard environments characterized by branch occlusion and leaf clutter. To address the difficulty in identifying and locating apples under such visually complex conditions, this paper proposes an improved YOLOv5-based visual recognition algorithm incorporating an efficient channel attention (ECA) module. The ECA module is strategically integrated into specific C3 layers (C3-3, C3-6, C3-9) of the YOLOv5 network architecture to enhance feature representation for occluded targets. During operation, the system simultaneously acquires apple pose information and achieves precise spatial localization through coordinate transformation matrices. Comprehensive experimental evaluations demonstrate the effectiveness of the proposed system. The custom-designed six-degree-of-freedom (6-DOF) robotic arm exhibits a wide operational range with a maximum working angle of 120°. The ECA-enhanced YOLOv5 model achieves a confidence level of 90% and an impressive in-range apple recognition rate of 98%, representing a 2.5% improvement in the mean Average Precision (mAP) compared to the baseline YOLOv5s algorithm. The end-effector positioning error is consistently controlled within 1.5 mm. The motion planning success rate reaches 92%, with the picking completed within 23 s per apple. This work provides a novel and effective vision recognition solution for future development of harvesting robots. Full article

Synchronization in complex networks mainly considers positive (attractive) couplings to guarantee network stability. However, in many real-world systems or processes, negative (repulsive) interactions exist, and this poses a challenging problem. In this proposal, we present an algorithm to design stable signed Laplacian matrices with mixed attractive and repulsive couplings that ensure stability in both complete and in-phase synchronization. The main result is established through a constructive theorem that guarantees a single zero eigenvalue, while all other eigenvalues are negative, thereby preserving the diffusivity condition. The algorithm allows control over the spectral properties of the matrix by adjusting two parameters, which can be interpreted as a pole placement strategy from control theory. The approach is validated through numerical examples involving the synchronization of a network of chaotic Lorenz systems and a network of Kuramoto oscillators. In both cases, full synchronization is achieved despite the presence of negative couplings. Full article

The double wishbone suspension (DWS) system is widely used in automotive engineering because of its favorable kinematic properties, which affect vehicle dynamics, handling, and ride comfort; hence, it is important to have an accurate 3D model, simulation, and analysis of the system in order to optimize its design. This requires efficient computational tools for parametric study. The development of effective computational tools that support parametric exploration stands as an essential requirement. Our research demonstrates a complete Wolfram Mathematica system that creates parametric 3D kinematic models and conducts simulations, performs analyses, and generates interactive visualizations of DWS systems. The system uses homogeneous transformation matrices to establish the spatial relationships between components relative to a global coordinate system. The symbolic geometric parameters allow designers to perform flexible design exploration and the kinematic constraints create an algebraic equation system. The numerical solution function NSolve computes linkage positions from input data, which enables fast evaluation of different design parameters. The integrated 3D visualization module based on Mathematica’s manipulate function enables users to see immediate results of geometric configurations and parameter effects while calculating exact 3D coordinates. The resulting robust, systematic, and flexible computational environment integrates parametric 3D design, kinematic simulation, analysis, and dynamic visualization for DWS, serving as a valuable and efficient tool for engineers during the design, development, assessment, and optimization phases of these complex automotive systems. Full article

Background: Sleep is a fundamental component of health, and deprivation has been linked to numerous adverse outcomes, including reduced academic and occupational performance, greater risk of accidents, and increased susceptibility to chronic diseases and premature mortality. Dietary and lifestyle behaviors are increasingly recognized as key determinants of sleep quality. Women are particularly susceptible to sleep disturbances due to hormonal fluctuations and psychosocial factors. However, women remain underrepresented in sleep research. This study aims to examine the associations between sleep quality, nutrition, and lifestyle in a large cohort of Spanish women. Methods: A cross-sectional study was conducted with 785 women aged 18–64. Participants completed the Pittsburgh Sleep Quality Index (PSQI) and the NutSo-HH questionnaire on dietary and lifestyle behaviors. Descriptive analyses, correlation matrices, Gaussian Graphical Models, and Principal Component Analyses were used to assess relationships between variables. Results: More than half of the participants rated their sleep quality as good or very good, although over 30% experienced frequent nighttime awakenings. Poor sleep quality was significantly associated with higher alcohol consumption, lower vegetable and white fish intake, and lower levels of physical activity. Diets rich in ultra-processed foods correlated moderately with subjective poor sleep and daytime dysfunction. However, no strong associations were found between stimulant consumption, late meals, or dietary patterns (e.g., Mediterranean diet) and sleep. Self-perceived health emerged as a protective factor, while nocturnal lifestyles were linked to longer sleep latency and fragmented sleep. Conclusions: In adult women, better sleep quality is linked to healthy dietary choices, regular physical activity, and a positive perception of general health. In contrast, alcohol use and irregular lifestyles are associated with poor sleep. Individual variability and cultural adaptation may moderate the impact of some traditionally harmful behaviors. Personalized, multidimensional interventions are recommended for promoting sleep health in women. Full article

Conventional horticultural fertilization frequently leads to nutrient loss and environmental contamination, driving interest in biodegradable controlled-release systems. This work developed soy protein isolate (SPI) matrices containing 5 wt.% FeSO₄·7H₂O using injection. The matrices were evaluated for crosslinking, mechanical properties, water uptake (WUC), soluble matter loss (SML), iron-release kinetics in water and soil, and biodegradability under composting conditions. Injection-molded samples achieved very high crosslinking with moderate rigidity and water absorption and delivered iron rapidly in water, while compression-molded samples exhibited slightly lower crosslinking but greater stiffness, higher WUC, minimal SML, and sustained iron release. Notably, both processing methods yielded comparable iron-release profiles in soil and complete biodegradation within 71 days. Overall, compression molding produces SPI-based matrices with superior mechanical strength and water retention, positioning them as an ideal solution for long-lasting, sustainable nutrient delivery in horticulture. Full article

The quantum description of relativistic orientable objects by a scalar field on the Poincaré group is considered. The position of the relativistic orientable object in Minkowski space is completely determined by the position of a body-fixed reference frame with respect to the position of the space-fixed reference frame, so that all the positions can be specified by elements q of the Poincaré group. Relativistic orientable objects are described by scalar wave functions $f\left(q\right)$, where the arguments $q=(x,z)$ consist of space–time points x and of orientation variables z from $SL(2,C)$ matrices. We introduce and study the double-sided representation $\mathit{T}\left(\mathit{g}\right)f\left(q\right)=f\left(g_l^{-1}q{g}_r\right)$, $\mathit{g}=(g_l,g_r)\in \mathit{M},$ of the group $\mathit{M}$. Here, the left multiplication by $g_l^{-1}$ corresponds to a change in a space-fixed reference frame, whereas the right multiplication by $g_r$ corresponds to a change in a body-fixed reference frame. On this basis, we develop a classification of orientable objects and draw attention to the possibility of connecting these results with particle phenomenology. In particular, we demonstrate how one may identify fields described by polynomials in z with known elementary particles of spins 0, ${\displaystyle \frac{1}{2}}$, and 1. The developed classification does not contradict the phenomenology of elementary particles and, in some cases, even provides a group-theoretic explanation for it. Full article

In this paper, we show that even-order Pascal tensors are positive-definite, and odd-order Pascal tensors are strongly completely positive. The significance of these is that our induction proof method also holds for some other families of completely positive tensors, whose construction satisfies certain rules, such that the inherence property holds. We show that for all tensors in such a family, even-order tensors would be positive-definite, and odd-order tensors would be strongly completely positive, as long as the matrices in this family are positive-definite. In particular, we show that even-order generalized Pascal tensors would be positive-definite, and odd-order generalized Pascal tensors would be strongly completely positive, as long as generalized Pascal matrices are positive-definite. We also investigate even-order positive-definiteness and odd-order strong complete positivity for fractional Hadamard power tensors. Furthermore, we study determinants of Pascal tensors. We prove that the determinant of the mth-order two-dimensional symmetric Pascal tensor is equal to the mth power of the factorial of $m-1$. Full article

There is currently a critical need for understanding how the response and activity of whole-cell bacterial reporters positioned in a complex biological or environmental matrix are impacted by the physicochemical properties of their micro-environment. Accordingly, a comprehensive analysis of the bioluminescence response of Cd(II)-inducible PzntA-luxCDABE Escherichia coli biosensors embedded in silica-based hydrogels is reported to decipher how metal bioavailability, cell photoactivity and ensuing light bioproduction are impacted by the hydrogel environment and the associated matrix effects. The analysis includes the account of (i) Cd speciation and accumulation in the host hydrogels, in connection with their reactivity and electrostatic properties, and (ii) the reduced bioavailability of resources for the biosensors confined (deep) inside the hydrogels. The measurements of the bioluminescence response of the Cd(II) inducible-lux biosensors in both hydrogels and free-floating cell suspensions are completed by those of the constitutive rrnB P1-luxCDABE E. coli so as to probe cell metabolic activity in these two situations. The approach contributes to unraveling the connections between the electrostatic hydrogel charge, the nutrient/metal bioavailabilities and the resulting Cd-triggered bioluminescence output. Biosensors are hosted in hydrogels with thickness varying between 0 mm (the free-floating cell situation) and 1.6 mm, and are exposed to total Cd concentrations from 0 to 400 nM. The partitioning of bioavailable metals at the hydrogel/solution interface following intertwined metal speciation, diffusion and Boltzmann electrostatic accumulation is addressed by stripping chronopotentiometry. In turn, we detail how the bioluminescence maxima generated by the Cd-responsive cells under all tested Cd concentration and hydrogel thickness conditions collapse remarkably well on a single plot featuring the dependence of bioluminescence on free Cd concentration at the individual cell level. Overall, the construction of this master curve integrates the contributions of key and often overlooked processes that govern the bioavailability properties of metals in 3D matrices. Accordingly, the work opens perspectives for quantitative and mechanistic monitoring of metals by biosensors in environmental systems like biofilms or sediments. Full article

This paper deals with the design of a complete state control for input unsigned, rank deficient matrix parameters of a linear system with system dynamics defined by ostensible structures of Metzler matrices. The proposed solution is based on the principle of diagonal stabilization of positive systems and uses a stabilizing additional component over the decomposition of the Metzler matrix in solving the incomplete internal positivity of such linear system structures. The novelty of the proposed approach is the unified representation of the parametric constraints of the Metzler matrix and the structurally constrained system inputs using linear matrix inequalities, which guarantees that the closed-loop system will be asymptotically stable. Despite the complexity of the constraint conditions on this class of linear continuous systems, the design conditions are formulated using sharp linear matrix inequalities only. A detailed design process is presented using a system-linearized mathematical model to verify the superiority and practicality of the proposed method. Full article

Plant extracts rich in phenolic compounds have been reported to exert different bioactive properties. Despite the fact that there are plant extracts with completely different phenolic compositions, many of them have been reported to have similar beneficial properties. Thus, the structure–bioactivity relationship mechanisms are not yet known in detail for specific classes of phenolic compounds. In this context, this work aims to demonstrate the relationship of extracts with different phenolic compositions versus different bioactive targets. For this purpose, five plant matrices (Theobroma cacao, Hibiscus sabdariffa, Silybum marianum, Lippia citriodora, and Olea europaea) were selected to cover different phenolic compositions, which were confirmed by the phytochemical characterization analysis performed by HPLC-ESI-qTOF-MS. The bioactive targets evaluated were the antioxidant potential, the free radical scavenging potential, and the inhibitory capacity of different enzymes involved in inflammatory processes, skin aging, and neuroprotection. The results showed that despite the different phenolic compositions of the five matrices, they all showed a bioactive positive effect in most of the evaluated assays. In particular, matrices with very different phenolic contents, such as T. cacao and S. marianum, exerted a similar inhibitory power in enzymes involved in inflammatory processes and skin aging. It should also be noted that H. sabdariffa and T. cacao extracts had a low phenolic content but nevertheless stood out for their bioactive antioxidant and anti-radical capacity. Hence, this research highlights the shared bioactive properties among phenolic compounds found in diverse matrices. The abundance of different phenolic compound families highlights their elevated bioactivity against diverse biological targets. Full article

In the competitive landscape of online learning, developing robust and effective learning resource recommendation systems is paramount, yet the field faces challenges due to high-dimensional, sparse matrices and intricate user–resource interactions. Our study focuses on geometric matrix completion (GMC) and introduces a novel approach, graph-based truncated norm regularization (GBTNR) for problem solving. GBTNR innovatively incorporates truncated Dirichlet norms for both user and item graphs, enhancing the model’s ability to handle complex data structures. This method synergistically combines the benefits of truncated norm regularization with the insightful analysis of user–user and resource–resource graph relationships, leading to a significant improvement in recommendation performance. Our model’s unique application of truncated Dirichlet norms distinctively positions it to address the inherent complexities in user and item data structures more effectively than existing methods. By bridging the gap between theoretical robustness and practical applicability, the GBTNR approach offers a substantial leap forward in the field of learning resource recommendations. This advancement is particularly critical in the realm of online education, where understanding and adapting to diverse and intricate user–resource interactions is key to developing truly personalized learning experiences. Moreover, our work includes a thorough theoretical analysis, complete with proofs, to establish the convergence property of the GMC-GBTNR model, thus reinforcing its reliability and effectiveness in practical applications. Empirical validation through extensive experiments on diverse real-world datasets affirms the model’s superior performance over existing methods, marking a groundbreaking advancement in personalized education and deepening our understanding of the dynamics in learner–resource interactions. Full article

Consider a sequence ${\left(X_n\right)}_{n\ge 1}$ of i.i.d. $2\times 2$ stochastic matrices with each $X_n$ distributed as $\mu$. This $\mu$ is described as follows. Let ${(C_n,D_n)}^T$ denote the first column of $X_n$ and for a given real r with $0<r<1$, let $r^{-1}{C}_n$ and $r^{-1}{D}_n$ each be Bernoulli distributions with parameters $p_1$ and $p_2$, respectively, and $0<p_1,p_2<1$. Clearly, the weak limit of the sequence ${\mu }^n$, namely $\lambda$, is known to exist, whose support is contained in the set of all $2\times 2$ rank one stochastic matrices. In a previous paper, we considered $0<r\le \frac{1}{2}$ and obtained $\lambda$ explicitly. We showed that $\lambda$ is supported countably on many points, each with positive $\lambda$-mass. Of course, the case $0<r\le \frac{1}{2}$ is tractable, but the case $r>\frac{1}{2}$ is very challenging. Considering the extreme nontriviality of this case, we stick to a very special such r, namely, $r=\frac{\sqrt{5}-1}{2}$ (the reciprocal of the golden ratio), briefly mention the challenges in this nontrivial case, and completely identify $\lambda$ for a very special situation. Full article