Disturbance of iron homeostasis in both the brain and blood is linked to cognitive impairment and neurodegenerative diseases. Investigation of the associations between changes in blood measurements and the accompanying structural changes in the brain, represented by susceptibility-weighted imaging, would generate evidence of a shared physiopathology between modalities, however linear approaches using derived imaging measures have not detected this relationship. Brain SWI of 4436 participants from UK Biobank at the first imaging visit were used to build a convolutional neural network (CNN) predicting the haemoglobin level. Haemoglobin concentration was grouped in three sets percentiles creating prediction classes. Attention maps were extracted from the CNN using Integrated Gradients to identify which brain regions allowed the CNN to correctly classify haemoglobin. The model predicted haemoglobin level >99% accuracy for three classes and >85% accuracy for ten classes. Using attention maps to understand the model's decision process, we identified regions of interest for the 3-class model. Four out of the nine regions were located in the cerebral cortex. Changes in the right thalamus and the white matter tracts between the temporal gyrus, putamen, hippocampus, and thalamus regions were associated to high haemoglobin class. Our model accurately predicted haemoglobin levels from SWI suggesting an association between the iron measures. Furthermore, it was possible to identify brain areas used by the CNN for decision-making. Using this integrated model of iron modalities will give a more comprehensive view of iron homeostasis and the potential associations with neurodegenerative diseases such as dementia.

The inverse dynamic games problem is to model expert demonstrations by identifying the underlying cost functions of multiple agents from observed trajectories of their dynamic game interactions. This article investigates discrete-time, finite-horizon linear-quadratic (LQ) problems where both the state weight matrix and input weight matrix are unknown, with the presence of both process noise and observation noise. In addition, each player's cost function incorporates a player-specific, unknown linear term with respect to the state. Under this framework, first, sufficient conditions are established for the solvability of the weight matrices. Subsequently, it is proved that the inverse dynamic games problem involving heterogeneous unknown target states is structurally identifiable, unaffected by process noise. Building on the necessary conditions for Nash equilibrium solutions in forward problems, the estimation of the cost function parameters is formulated as a nontrivial solution to a homogeneous linear estimation problem, which can be implemented in a distributed manner. Furthermore, the proposed estimator achieves statistical consistency under the influence of observation noise. The effectiveness is illustrated through a multivehicle spring-coupled dynamic game and an interactive steering control scenario.

State elimination in polynomial models: A linear algebraic approach

Predictive modeling of yarn properties in sustainable fiber blends: A fuzzy linear regression approach

Environmental benefits and impacts forecasting for three-phase induction motors operations in marine applications: A multiple linear regression approach

Optimization of corona ring design dimensions for extra high-voltage substation bus post insulators using a linear regression approach

In our rapidly digitizing world, data is being collected at a never-before-seen pace from dynamic global sectors like healthcare, manufacturing, sales, IoT devices, the web, smart gadgets, social media, and organizations on a regular basis. The properties of this type of data are high dimensionality, large volume, redundant features, and noise. As such, representing and processing such a large amount of complex, heterogeneous data becomes much more challenging. In many sectors where there is a lot of data with lots of columns or classes, dimensionality reduction techniques are crucial. One essential method for evaluating and understanding high-dimensional data is the application of dimensionality reduction techniques.These methods collect a wealth of interesting data properties, including dynamical structure, correlation between data sets, covariance, and input-output interactions. This paper reviews several types of algorithms used in dimension reduction in order to provide readers with a thorough and lucid summary of the field and to give them a sense of how to assess its increasing importance over the past few years. Since linear dimensionality reduction techniques have straightforward geometric interpretations and generally appealing computing features, they are fundamental to the analysis of high dimensional data. Principal Component Analysis (PCA), Singular Value Decomposition (SVD), linear discriminant analysis(LDA), and Independent Component Analysis (ICA) are the four linear dimensionality reduction approaches that will be examined and compared in this study. The purpose of this study is to examine and contrast the benefits and drawbacks of dimensionality reduction in several widely applied mathematical concepts and techniques.

BackgroundIn 2022, nearly 60,000 prostate cancer (CaP) cases were reported among Black men, who face an estimated lifetime risk of approximately 1 in 6, compared with 1 in 8 among White men. The disproportionate burden of CaP in Black men has been attributed to a combination of health-system factors, variations in care processes, and individual- or patient-level determinants. Addressing these multilevel contributors will require innovative strategies to advance prostate health equity.ObjectiveIn this paper, we discuss the protocol for assessing the feasibility of establishing a Community Living Lab (CoLLab) Learning Health System in Black communities and the impact on facilitating access to prostate health Resources, Education, Amenities, and Community Health (REACH) services for Black men.MethodsThe proposed research design for the study is a pragmatic clinical trial. The research setting is Northeast Florida, with the intervention based in 3 American Legion Posts (ALPs) and the control in one ALP. The development of the CoLLab REACH intervention was guided by the Intervention Mapping Framework, focusing on program design, adoption, implementation, and monitoring and evaluation plan. The intervention was cocreated with community members and is being provided by community health workers. The primary outcomes are improvement of Black men’s CaP awareness, knowledge, attitude, health beliefs, perceived control, intentions, cues to action, and clinical trials’ awareness. Generalized linear mixed regression approaches will be used to determine the differences between the study variables for the intervention and control groups.ResultsThe CoLLab study was awarded in September 2023 and received institutional ethics committee approval on March 20, 2024. CoLLab REACH intervention includes the following services: (1) Wellness RX program, implemented to address financial health needs and grocery distribution to address food deserts of the communities around the posts; (2) social determinants of health Navigation Services that include food, housing, transportation, employment aid, legal aid, and financial support based on zip code; (3) educational resources and videos on CaP prevention, screening, detection, treatment, and survivorship; (4) The Clinical Trials Matching Services to match participants to Mayo Clinic clinical trials and biomedical research; and (5) CaP Advocacy training program. Baseline recruitment for the intervention arm was completed in June 2025, with 183 Black men enrolled at ALPs. By June 2025, 11 participants were enrolled at the control ALP, and recruitment for the control arm remains ongoing. Data analysis will be conducted upon completion of follow-up assessments at months 4, 8, and 12.ConclusionsWe successfully co-designed the CoLLab REACH services at 3 ALPs. Using a longitudinal pretest-posttest design, we will assess the intervention’s impact at both the individual and community levels to evaluate the feasibility of this community-based, culturally informed approach. In addition, we will examine the feasibility of replicating the CoLLab Learning Health System in underserved communities nationwide.

Predicting language lateralization using functional MRI (fMRI) in patients with cerebral vascular malformations close to language areas is essential for treatment decision-making and patient outcomes. Functional MRI-based prediction is challenged because of potential remodeling processes and hemodynamic phenomena. However, there is a lack of possible factors influencing laterality prediction. The authors hypothesized that there might be an impact of lesion type and location on language lateralization. This retrospective study included 24 patients with arteriovenous malformations (AVMs), 11 patients with cavernomas, and 15 healthy controls. Participants performed a subvocal verb-generation task during fMRI. Data analysis in Statistical Parametric Mapping (SPM) 12 involved realignment, coregistration, and smoothing for preprocessing. The authors conducted a whole brain analysis using the general linear model approach at the individual level and calculated the lateralization indices (LIs) using the LI toolbox implemented in SPM independently based on the frontal, temporal, and parietal lobes. The mean absolute LIs were above 0.2 in all groups. Distribution between groups varied significantly (p = 0.032, f = 0.34). A significant difference was found between patients with AVMs and healthy controls (p = 0.038, r = 0.628). Specifically, patients with frontal AVMs showed significantly lower frontal LIs than did healthy controls (p = 0.032, r = 0.435). In contrast, LIs in cavernoma patients did not differ significantly from controls (p = 0.313). No significant difference was observed between language-adjacent and language-distant lesions (p = 0.14). The results of this study suggest that lesion type and location influence language lateralization prediction. Frontal AVMs exhibit significantly lower LIs, requiring caution and experience in interpreting results to ensure patient safety. Cavernomas did not influence LI. Further research with larger cohorts is necessary to understand the underlying causality and neuroplastic changes involved.

Objective. Electrocortical and hemodynamic signals measured by electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) can be used to provide complementary information about neural activity underpinning auditory processing. However, the causal interactions between these signals remain underexplored, partly due to several factors, including the complexity of disentangling cortical and systemic physiological components captured in fNIRS modality. Recent developments in the ability to estimate the influence of non-cortical sources on fNIRS signals provide new opportunities to address these confounds. In this study, we investigated causal interactions between vascular-hemodynamic and electrocortical dynamics through simultaneous recording of fNIRS and EEG during auditory processing.Approach. We employed multimodal multivariate Granger causal analysis to investigate causal interactions between EEG and fNIRS signals. To explore the role of systemic physiology on the obtained causal interactions, a temporally-embedded canonical correlation analysis-general linear model-based preprocessing approach was applied to fNIRS signals to correct for potential systemic physiology confounds.Main results. Our results showed significantly stronger causal interactions originating from fNIRS in the right auditory cortex to frontocentral EEG in the 38-42 Hz frequency range during the auditory task relative to rest before correction. We additionally observed broad connectivity in the direction originating from fNIRS variables to EEG across frequency bands within participants throughout the dataset. These results suggest complex relationships between cortical vascular-hemodynamics, electrocortical oscillations, and systemic physiological components with roles in both task-related and spontaneous neural activity.Significance. This study highlights the complex causal interplay between electrical, cerebral hemodynamic, and systemic physiology components underpinning spontaneous electro-vascular dynamics associated with auditory processing.

In genomic prediction, it remains unclear whether increasingly complex or ensemble models improve prediction over established linear approaches, and why prediction accuracy varies among traits. Here, we evaluated a comprehensive suite of genomic prediction models, including linear mixed models, Bayesian variable selection, kernel methods, machine learning algorithms, graph attention networks, and stacked ensembles, in mango (Mangifera indica L.). Across five traits, prediction accuracy converged across linear, Bayesian, kernel, and ensemble models, with only marginal gains derived from stacking and no systematic advantage of machine learning approaches. Ensemble ablation and weight analyses revealed that predictive signal was dominated by additive and smooth kernel components, while more complex learners contributed little or negatively upon performance. To explain these trait-dependent patterns in predictability, we quantified the phylogenetic signal using genome-wide marker-based trees. All traits showed a significant phylogenetic signal, with the magnitude varying widely and strongly associated with prediction accuracy (r ≈ 0.71). Traits with strong phylogenetic structure achieved the highest prediction accuracies, whereas traits with a weaker signal were consistently harder to predict, regardless of model choice. Together, these results confirm that, in mango, genomic prediction accuracy is determined more by evolutionary structure and trait architecture rather than increasing model complexity. Aligning prediction strategies with the evolutionary basis of trait variation may therefore be more effective than adopting increasingly complex models.

Abstract We present a low-frequency accelerometer calibration system based on rotation in the gravitational field. Example characterizations of three accelerometers from (0.01 to 1.5) Hz with an uncertainty analysis demonstrate k = 2 magnitude and phase uncertainty of <0.1 % and <0.2 ◦ , respectively. This rotational system complements the linear shakers in the NIST Primary Vibration Calibration Laboratory by improving uncertainty in the range of overlap and extending accelerometer calibration capability to lower frequencies. We demonstrate a magnitude comparison between the linear and rotational approaches, showing agreement to within the rotational calibration uncertainty of <0.1 %.

We present a spin-dependent extension of the non-orthogonal generalized Wannier function (NGWF) formalism within the framework of linear-scaling density functional theory (LS-DFT) as implemented in the ONETEP code. In traditional LS-DFT representations, both spin channels are constrained to share a common variational basis, which limits the accuracy for systems that are spin-polarized or exhibit magnetic order. Our approach allows NGWFs to vary independently for each spin channel, enabling a more accurate representation of spin-polarization in the electronic density. We demonstrate the efficacy of this method through a series of test cases, including localized magnetic defects in two-dimensional hBN, transition metal complexes, two-dimensional van der Waals magnetic materials, and both bulk and nanocluster ferromagnetic Co. In each scenario, the incorporation of spin-dependent NGWFs results in enhanced accuracy for total energy calculations, improved localization of spin density, and accurate predictions of magnetic ground states. This improvement is particularly notable when combined with DFT+U and DFT+U+J corrections. In this study, we take the opportunity to describe the combination of DFT+U+J and the projector-augmented wave (PAW) formalism within the LS-DFT framework, including how PAW participates in the ionic Pulay force and in the minimum-tracking linear response approach for computing parameters in situ. Our findings demonstrate that spin-dependent NGWFs are a crucial and computationally efficient advancement in the linear-scaling DFT simulation of spin-polarized materials.

Adolescent cyberbullying victimization has emerged as a pressing global concern, with complex risk factors spanning across multiple ecological levels. However, few studies have systematically examined multilevel determinants using large-scale, cross-national data. Guided by socioecological theory, this study examined associations between adolescent cyberbullying victimization and a range of individual, family, school, and community-level factors using both linear regression and machine learning approaches. The study drew on data from the 2019 Organization for Economic Co-operation and Development Survey on Social and Emotional Skills, involving 22,478 adolescents (Mage = 15.45, SD = 0.55; 51.77% female) from 10 cities worldwide. Both analytical approaches consistently identified a set of variables associated with cyberbullying victimization, including gender, grade, smoking, drinking, sleep problems, family material well-being, family conflict, parent-child relationships, parental punishment, school bullying, school climate, school connectedness, teacher expectations, teacher support, peer deviance, friendship quality, neighborhood connectedness, and neighborhood safety. In addition, the Shapley additive explanation analysis highlighted several variables that were less prominent in the traditional regression models, including body mass index and school anxiety. Findings underscore the multifaceted and cross-system nature of cyberbullying victimization among adolescents. Effective prevention efforts should adopt a multilevel approach that addresses risk and protective factors across individual, family, school, and community contexts. (PsycInfo Database Record (c) 2026 APA, all rights reserved).

Femoral bone metastases represent a frequent and severe complication in patients with advanced solid tumours. Although fracture risk assessment commonly relies on Mirels' score, its limited specificity often leads to unnecessary surgical interventions. Patient-specific finite element (FE) models have shown improved accuracy; however, current approaches vary widely in methodology and rarely capture the full fracture process. This study investigates for the first time the application of a linear FE approach based on an incremental element deletion technique to simulate both fracture initiation and propagation in femurs with lytic metastases. Twenty-four patients with femoral lytic lesions were retrospectively analyzed, and model outcomes were compared with clinical results. The proposed approach was evaluated for its ability to stratify patients according to pathological fracture risk, in comparison with conventional simulation methods and the clinically accepted Mirels' score. In addition, a new failure threshold parameter derived from the work required to fracture the femur was investigated. The simulations successfully replicated clinically observed fracture paths and demonstrated strong capability in differentiating high- and low-risk patients. The failure criterion based on the last applied load during simulated crack propagation provided the highest diagnostic performance, achieving excellent sensitivity and specificity. The fracture initiation parameter showed comparable performance, while the work-based parameter appeared more affected by variability in femur visibility. The proposed modeling framework offers the advantage of predicting fracture paths, providing clinically valuable insight, and represents a further step toward improved stratification methods for the clinical management of femoral metastases.

Although various reduced-order models (ROM) have been developed for the dynamics of multibody systems modeled based on the Absolute Nodal Coordinate Formulation (ANCF), there still remain issues concerning the efficiency of model reduction. The traditional Proper Orthogonal Decomposition (POD) method can effectively reduce the dimensionality of the ANCF dynamic equations, but it necessitates mapping the generalized nodal coordinates back to their original dimension at each time step to calculate the nonlinear terms of the elastic internal forces. This leads to limited time savings in equation solving despite the dimensionality reduction achieved by the POD method. To enhance the efficiency of equation solving using the POD method, this study proposes the POD-L model reduction approach. Firstly, we apply POD to the collected snapshot matrix and retain dominant POD modes to construct a linear basis. Next, leveraging local linearization, we approximate the system stiffness matrix as constant within a limited displacement range. And iteratively obtain the elastic internal forces using the improved local linearization method, reducing error accumulation. The computed forces are then incorporated into the dynamic equations, which are projected onto the linear basis to reduce their dimensionality. The POD-L approach not only significantly reduces the dimensionality of ANCF dynamic equations and simplifies the computation of nonlinear elastic internal forces, but also reduces the error accumulation associated with the local linearization approach. It demonstrates good accuracy in prolonged simulations. It is well suited for ANCF dynamics problems involving planar triangular element modeling. The effectiveness and accuracy of the POD-L method are validated through three numerical examples.

A multi-dimensional modeling framework integrating metabolomics analysis and process modeling for bioprocess characterization.

Inter-species variability in bivalve purification processes: Towards evidence-based optimization.

At the molecular level, selection pressures often act on protein structural features, yet most evolutionary analyses remain confined to linear sequences. Early structure-informed approaches improved interpretability by mapping single-site metrics onto protein structures, and later methods introduced three-dimensional (3D) sliding windows to capture spatially clustered signals missed by linear window approaches. These frameworks, however, are restricted to predefined statistics and narrowly defined 3D window types, limiting the scope of questions that can be addressed. We developed an R package, evo3D, as a new framework for structure-informed evolutionary analysis that supports a wide range of downstream statistics and scales from simple to complex structures. evo3D extracts structure-informed multiple sequence alignment subsets (spatial haplotypes), making the structure-informed unit of analysis directly available to users. The framework supports fixed-count and fixed-distance spatial windows, introduces residue and codon analysis modes, and extends to multimers, interfaces, and multiple structural models through a single wrapper, run_evo3d(). We demonstrate evo3D's utility by performing an epitope-level diversity scan of hepatitis C virus E1/E2 complex, identifying conserved spatial neighborhoods missed by linear sliding windows, and by evaluating evo3D's scalability on the octameric Chikungunya virus E1/E2 assembly. Importantly, evo3D formalizes the core components of structure-informed analysis of molecular evolution and removes technical barriers. As a result, the framework streamlines the evaluation of evolutionary patterns directly within 3D structural contexts, and we anticipate its wide application in molecular evolution studies. The package is available at github.com/bbroyle/evo3D.

The increasing global demand for hemp seed oil (HSO) necessitates robust analytical methods to monitor pesticide residues and ensure compliance with increasingly stringent international regulations. This study presents a rapid and efficient multiresidue method for the determination of 148 pesticides in HSO using liquid chromatography-tandem mass spectrometry (LC-MS/MS), coupled with a linear retention index (LRI) approach to enhance compound identification confidence. The method features a streamlined extraction protocol with reduced solvent consumption (5 mL ACN per sample) and achieves chromatographic separation within 12 min. Method validation demonstrated excellent sensitivity (LOQs, 0.03-22.29 ng g⁻1), accuracy (71-129%), and precision (RSD ≤ 19.9%). Integration of the LRI system enabled unambiguous differentiation of isomeric compounds. Application of the method to 12 commercial HSO samples revealed the presence of 21 pesticides, with malathion and proquinazid exceeding Canadian maximum residue limits (MRLs) in two samples. This robust and eco-efficient method provides a robust solution for pesticide monitoring in HSO, aligning with EU, Canadian, and Californian regulatory frameworks while introducing LRI as a valuable tool for pesticide analysis in complex lipid matrices.