Gaussian process modelling of infectious diseases using the Greta software package and GPUs.

Nonparametric model and response analysis of the complex uncertain pipeline-casing system

An accurate underwater camera calibration method based on a nonparametric distortion model

A novel nonparametric Bayesian model for time series clustering: Application to electricity load profile characterization

<b>Aims</b> : High-resolution spectrographs are central to modern exoplanet research and are particularly effective for detecting Earth-like planets whose radial velocity (RV) signals can be only a few tens of centimeters per second. Achieving this level of precision requires highly accurate wavelength calibration. A key factor in this process is the modeling of the instrumental profile (IP), which describes the response of the spectrograph to incoming light. The true IP of a high-resolution instrument is often complex. It may show asymmetry or extended wings and change across the detector because of optical aberrations, variations in fiber illumination, and environmental effects. These features lead to systematic errors in the measured line centers when traditional parametric models such as Gaussian functions are used, and they limit the achievable RV precision.<br> <b>Methods:</b> This work introduces a non-parametric IP modeling method based on Gaussian Process Regression (GPR). The IP is treated as a smooth function with a flexible covariance structure instead of being constrained by a predefined analytic form. GPR learns both the global structure and small-scale features of the line shape directly from the data. Since the IP varies slowly across the detector, the method divides each spectral order into several consecutive spatial segments. Each segment is fitted independently, capturing local variations. The model includes measurement uncertainties and provides a probabilistic description of the IP. Adjacent segments are linked with smooth interpolation to ensure a continuous IP across the entire order. Model performance is evaluated using reduced chi-squared and root mean square error (RMSE), allowing quantitative assessment and comparison with traditional approaches.<br> <b>Results:</b> The method is tested with laser frequency comb (LFC) exposures from the fiber-fed High Resolution Spectrograph (HRS) on the 2.16 m telescope at Xinglong Observatory. The LFC produces a dense and highly stable set of emission lines and is well suited for validating IP reconstruction. Three experiments show clear and consistent improvements. Using odd-numbered lines to predict evennumbered ones within a single exposure reduces the RMSE by 35.6% compared with a Gaussian model, showing better determination of line centers. Applying an IP model trained on one exposure to a later exposure reduces the RMSE by 42.5%, demonstrating improved stability when the model is transferred between exposures. A comparison between two channels in the same exposure shows a 37.1% improvement in calibration consistency, indicating reduced channel-tochannel systematics.<br> <b>Conclusions:</b> The results show that GPR provides a more accurate description of the instrumental profile and its spatial variation than traditional parametric models. The improved reconstruction of the IP leads to more accurate line center measurements and a more stable and precise wavelength solution. This capability is important for pushing the RV precision of high-resolution spectrographs toward the centimeter-per-second level. GPR offers a promising approach for modeling instrumental profiles and supports the precision required for detecting Earth-like exoplanets.

Gaussian process latent variable model and Bayesian inference for non-parametric failure modeling applied to ship engine

Landscape fragmentation and habitat loss are major drivers of global biodiversity decline. Understanding how animals adjust their behavior in response to these threats and adapt to human-altered environments is essential for developing effective conservation strategies. In this study, we apply Bayesian models to a unique movement dataset to explore how landscape transformation influences behavioral patterns in the giant anteater (Myrmecophaga tridactyla; Pilosa; Mammalia), a vulnerable species with limited physiological thermoregulatory capacity and strong behavioral responses to environmental changes. We used an extensive GPS tracking dataset from 41 giant anteaters in the Brazilian savanna-a biodiversity hotspot facing extensive landscape alteration due to land-use and land-cover change. We used the Time-Explicit habitat selection model to investigate giant anteater sex-based differences in habitat selection. Additionally, we evaluated how habitat types influence the species' activity patterns using a nonparametric Bayesian Hidden Markov model. Our study reveals that giant anteaters change their movement patterns in response to land use and land cover change. Individuals spent more time in and selected native habitats compared to human-modified habitats, regardless of sex. The time spent on, and selection for, native habitats are likely due to greater resources such as food, shelter, and protection from predation. Additionally, giant anteaters are more likely to rest in native habitats while being more active in human-disturbed areas. Our research reveals the negative impact of human-disturbed landscapes on animal movement. Consistent with our expectations, giant anteaters spend more time in native habitats and tend to avoid areas with high human disturbance, underscoring the urgent need for conservation efforts in degraded landscapes. Understanding how different land use and land cover classes influence animal habitat selection and activity patterns is essential for assessing the species' adaptability and ecological requirements in human-modified environments. Our findings can guide the prioritization of critical areas for conservation and restoration, offering valuable insights for policymakers and supporting the effective management of this vulnerable species and the Brazilian savanna.

Background: Eosinophilic granulomatosis with polyangiitis (EGPA) is a rare vasculitis, and contemporary population data from Central and Eastern Europe are limited. Aim: To describe hospital-based incidence, patient characteristics and comorbidities among EGPA hospitalizations in Poland (2014–2023), including differences by age, sex and place of residence. Methods: This retrospective, population-based study used nationwide hospital discharge records with an EGPA diagnosis. First EGPA-coded hospitalizations were used to estimate annual incidence per 1,000,000 inhabitants. Demographics, duration of stay and accompanying comorbidities (cardiovascular disease, pulmonary disease, and asthma) were analyzed for all EGPA hospitalizations. Place of residence was classified as urban or rural. Group differences and temporal trends were assessed using appropriate parametric and non-parametric tests and regression models, with a two-sided p value < 0.05 considered statistically significant. Results: Between 2014 and 2023, 911 patients had a first EGPA-coded hospitalisation in Poland, corresponding to a mean annual hospital-based incidence of 2.38 per 1,000,000 inhabitants (range 1.28–3.38); incidence declined significantly from 2014 to 2019 (p < 0.001) and was disrupted during the COVID-19 period. Overall, 3524 EGPA hospitalisations were recorded, and women were more frequently hospitalised than men (54.5% vs. 45.2%; p < 0.001). Mean age at hospitalisation increased over time, with patients in 2023 being about 5–6 years older than in 2014 (p ≤ 0.009). Median length of stay was 8 days for first admissions and 5 days for all EGPA stays and shortened significantly over the study period (p < 0.001). Cardiovascular disease, pulmonary disease and asthma were present in 23.6%, 35.3% and 32.3% of patients, respectively. Cardiovascular disease was more common in men and in rural residents (both p < 0.001) and was associated with older age (p < 0.001), whereas pulmonary disease was associated with younger age (p < 0.001). Among women, the proportions with pulmonary disease and asthma decreased over time (p = 0.009 and p = 0.025). Conclusions: EGPA in Poland is rare, with hospital-based incidence comparable to other European and Asian populations. The hospitalized EGPA population is aging and cardiovascular comorbidity is increasingly prominent, especially in older and rural patients, while recorded pulmonary disease and asthma in women are decreasing.

Background: Metabolic syndrome (MetS) is linked to brain degeneration and Alzheimer’s disease (AD). Women, especially during menopausal transition, show increased susceptibility to AD-related brain changes. This study investigated the sex-specific neurostructural impact of MetS on brain regions vulnerable to AD. Methods: This cross-sectional study analyzed data from 500 participants (303 women, 197 men) from the Baependi Heart Study cohort, Brazil. High-resolution T1-weighted MRI scans were used for volumetric analysis of AD-related regions of interest (ROIs). Non-parametric quantile regression models compared ROI volumes between MetS and Non-MetS groups, stratified by sex and age (median split), adjusting for age and education. Results: No significant differences in ROI volume were observed between the MetS and Non-MetS groups in men. In women, findings were age-dependent. The younger cohort (≤48 years) with MetS exhibited significantly smaller left hippocampal volume (p = 0.02) and a trend toward smaller left middle temporal gyrus volume (p = 0.05) compared to Non-MetS. The older cohort (>48 years) with MetS showed a significantly larger right amygdala volume (p < 0.001). Furthermore, age-related volume decline in the hippocampus and middle temporal gyrus was significant in Non-MetS women but not in women with MetS, suggesting that MetS may be a confounding factor in age-related neurodegeneration. Conclusions: MetS is associated with sex-specific alterations in AD-vulnerable brain structures. In women, MetS may influence medial temporal lobe atrophy pre-menopause, and is linked to amygdala enlargement post-menopause. These exploratory results generate the hypothesis that MetS may uniquely predispose women to AD-related neurodegeneration, which requires critical longitudinal confirmation.

Navigating the world requires accurate categorization of objects around us, which often involves processing multiple sources of information. The predictiveness of a source plays an important role in accurate categorization. This study aims to investigate how the predictiveness of features modulates the processing strategies of two features that are generally considered more integral than separable: color and luminance. Participants categorized a set of visual stimuli, created by varying levels of color and luminance, into two categories defined by logical rules. The stimulus-category mapping was 100% in Experiment 1, but it was reduced to 95% in Experiment 2. In both experiments, the predictiveness of both features was equal. Lastly, in Experiment 3, we introduced unequal predictiveness such that color was more predictive for some participants, while luminance was more predictive for others. These manipulations were designed to test whether, as predicted by the strong version of the relative saliency hypothesis, even integral features such as color and luminance could be processed serially if one were made more predictive of the category. Across the three experiments, we employed both system factorial technology (SFT) and computational modeling to infer processing strategies in nonparametric and parametric manners, respectively. Although some variability existed at the individual subject level, both non-parametric and parametric modeling revealed robust evidence for coactive processing for the aggregated group data, regardless of the varied stimulus-category mapping and feature predictiveness. These findings suggest that the processing of color and luminance within an object involves obligatory coactive processing, thereby challenging the strategic adjustment relative saliency hypothesis.

Childhood trauma is associated with positive subjective responses to acute opioids in the laboratory. Yet, its impact on behavioural economic demand for opioids remains unclear. Because demand can predict future use and misuse, we investigated whether individuals with childhood trauma also display greater demand for opioids in the laboratory. This secondary analysis used data from a double-blind, randomised, controlled, counterbalanced trial. Across two sessions, individuals with (n = 26) and without (n = 21) childhood trauma received a high (0.15mg/kg) and low (0.01mg/kg) dose of intramuscular morphine 120min before a hypothetical purchase task. Participants also regularly reported their subjective desire for the study drug. Area under the consumption and expenditure curves, and specific demand indices (Omax, Pmax, breakpoint, intensity, elasticity), were analysed with non-parametric mixed models. Consumption and expenditure for low dose morphine was lower in the childhood trauma group when compared to high dose morphine in this group, and low dose morphine in the control group (ps ≤ 0.03). Consistent results were obtained for specific demand indices Omax, Pmax, and breakpoint, but not intensity or elasticity. Positive correlations between subjective desire and consumption and expenditure were only observed in the childhood trauma group (Tau-c = 0.24-0.34, ps ≤ 0.03). Under laboratory conditions, desires to use opioids among individuals with childhood traumas may not manifest in strong intentions to acquire these drugs. Diverging results from psychometric measures of subjective and economic opioid value highlight the need for future research to explore contextual and resilience factors to understand how problematic opioid use emerges after childhood traumas.

Movement disorders occur in early-onset hereditary spastic paraplegia (HSP), but their prevalence, genotype associations, and clinical impact are not well defined. To delineate the spectrum and frequency of movement disorders across childhood-onset HSP genotypes and assess associations with clinician- and caregiver-reported outcomes. We performed a cross-sectional analysis of 428 children and young adults with molecularly confirmed HSP enrolled in a multicenter natural history study. Standardized clinical phenotyping and video examinations were reviewed. Movement and motor disorders (dystonia, ataxia, parkinsonism, tremor, choreoathetosis) were identified using predefined criteria. Associations with SPATAX-EUROSPA disability stage (SPATAX), Spastic Paraplegia Rating Scale (SPRS; total and spasticity subscore), Modified Ashworth Scale, and Caregiver Priorities and Child Health Index of Life with Disabilities (CPCHILD) quality-of-life scores were analyzed with nonparametric tests and multivariable linear models adjusted for age and sex. Movement disorders were present in 27.6% (118/428) of participants; 22.8% of these had ≥2 movement disorders. Dystonia (16.4%) and ataxia (10.0%) predominated. Distinct genotype-specific patterns were observed: dystonia in SPG4, SPG3A, and AP-4-HSP; parkinsonism in SPG11; and ataxia in SPG15, SPG76, SPG7, SPG5a, and SPG46. Presence of any movement disorder correlated with greater disability and motor burden and lower quality of life. In adjusted models, dystonia and parkinsonism were associated with higher SPATAX and SPRS scores, whereas ataxia and tremor correlated with lower scores; dystonia showed the largest decrement in CPCHILD. Movement disorders are common, genotype-specific, and clinically impactful features of childhood-onset HSP. Routine screening and genotype-tailored management of movement disorders, especially dystonia, may improve functional outcomes and quality of life. © 2025 International Parkinson and Movement Disorder Society.

We conducted simulations of common bean (Phaseolus vulgaris L.) breeding programs to better understand the interplay between different choices a breeder must make when launching a genomic selection (GS) pipeline. We complement preceding studies on optimizing model parameters and training set makeup by exploring the practical implementation of GS in a common bean breeding program aimed at increasing seed yield. We simulated 24 GS implementation pathways on (1) what generation to train a new prediction model, (2) what generation to select parents for the next cycle, (3) which generation to collect training data, and (4) whether to use a parametric (ridge regression best linear unbiased predictor) or a nonparametric model (artificial neural network) for estimating breeding values. We found that early generation parent selections (also called rapid-cycle GS) generally resulted in higher gain over three breeding cycles compared to late-generation parent selections. When implementing a new parametric genomic prediction model, training data should be as diverse as possible, while also matching testing data in terms of genetic makeup and allele frequency. Parametric models showed more consistent genomic estimated breeding value prediction accuracy, while nonparametric models fluctuated, showing both the highest and the lowest prediction accuracy across all pathways. Despite the trade-off between gains and genetic variance, nonparametric models showed greater balance of allelic diversity and gains. We observed that the key to sustained gains over time is the renewal of genetic variance. Our results indicate a potential for the use of nonparametric models, but more investigation will be required to stabilize their performance.

One-pass online learning under evolving feature data streams: A non-parametric model

Nonparametric online modeling for differential-driven unmanned surface vehicle based on fuzzy C-means and PD-like event-triggered strategy

This study examined whether a brief measure combining subjective cognitive concerns and financial interference, termed Subjective Cognitive Decline-Financial (SCD-F), is associated with excess spending behavior in older adults. Community-dwelling older adults, some with early memory loss and some with no cognitive complaints (N = 150, M age = 72.6) provided 12 months of checking account statements and participated in interviews to clarify aspects of their personal financial behaviors. SCD-F was defined by asking if memory decline was interfering with financial decision making or transactions. A 3-point SCD-F measure was created. Excess spending was determined by whether checking account expenditures exceeded all sources of income. Nonparametric tests (Kruskal-Wallis and Mann-Whitney U) and multiple regression models assessed group differences and predictors. Group differences in excess spending were pronounced (H(2) = 15.75, p < .001). Those in the high SCD-F group had a significantly greater likelihood of excess spending (Z = -4.11; r = .43) and higher excess spending percentages (Z = -4.11; r = .43) compared to those with no memory loss. Regression analyses indicated that SCD-F was the strongest predictor of excessive spending (β = .40, t = 5.43, p < .001), even after controlling for age, gender, race, and education (R2 = .235, F(5,144) = 8.86, p < .001). A brief self-report measure, SCD-F, effectively identifies older adults at risk of financial mismanagement, even absent formal cognitive impairment. Monitoring subjective cognitive concerns together with financial interference could enable early intervention. This brief measure may be useful in clinical settings as a screening tool, and in large national surveys.

Waste generation exceeding landfill capacity highlights the urgency of realizing its economic value. This study analyzes the effect of Quality of Facilities and Infrastructure (X1) and Use of Waste Banks (X2) on Waste Management-Based 3R (Y1) and Waste Economic Value Utilization (Y2) using a truncated spline nonparametric path model. This study evaluates the performance of a nonparametric path analysis model based on truncated spline combined with a double resampling. Data were collected using a Likert scale questionnaire on community perceptions of waste’s economic benefits in Batu City. Simulation results show that the Jackknife-Bootstrap method achieves the lowest average bias (0.058), outperforming single resampling approaches such as Single-Bootstrap (0.178) and Single-Jackknife (0.176). Empirical findings indicate that improvements in the Quality of Facilities and Infrastructure (X1) and Waste Bank Use (X2) significantly enhance Waste Management Based 3R (Y1) and Utilization of Waste Economic Value (Y2). The truncated spline model reveals a saturation effect, where the marginal benefits of X1 and X2 decrease beyond a threshold. Furthermore, Y1 positively affects Y2, emphasizing the importance of efficient waste management in enhancing economic value. The results support policies promoting balanced infrastructure development, community empowerment, and institutional innovation for sustainable circular economy implementation.

Scarce evidence exists on the impact of ozone (O3) on atherosclerotic cardiovascular disease (ASCVD) morbidity in tropical urban settings, and whether temperature modifies this effect remains unclear. To bridge this gap, we assessed the association between ambient O3 and ASCVD outpatient visits, and the potential effect modification by temperature, in Haikou, China. A time-series analysis was performed on data from 163,348 daily hospital outpatient visits for ASCVD collected between 1 January 2020 and 31 December 2022. The association between O3 exposure and daily visits was evaluated with an over-dispersed Poisson generalized additive model (GAM), and the modifying effect of temperature was scrutinized using a nonparametric bivariate response surface model. A 10 μg/m3 increment in the daily maximum 8 h average concentration of O3 was associated with a 1.35% (95% CI: 0.63, 2.07) increase in ASCVD outpatient visits at lag0. Stratified analyses revealed that the association between O3 and ASCVD visits was only significant during the warm season, with stronger effects observed above 30 °C, peaking at 34 °C (lag06). The combined exposure to high temperature and O3 concentrations significantly amplified ASCVD outpatient visits. Ambient O3 exposure was associated with increased ASCVD outpatient visits in the tropical city, and this risk was enhanced under high temperatures. These results highlighted the importance of considering temperature interaction in O3-related risk assessments.

We analyzed the spatially resolved and global star formation histories (SFHs) for a sample of 25 TNG50- SKIRT Atlas galaxies to assess the feasibility of reconstructing accurate SFHs from Euclid-like data. This study provides a proof of concept for extracting the spatially resolved SFHs of local galaxies with Euclid, highlighting the strengths and limitations of SFH modeling in the context of next-generation galaxy surveys. We used the spectral energy distribution (SED) fitting code Prospector to model both spatially resolved and global SFHs using parametric and nonparametric configurations. The input consisted of mock ultraviolet--near-infrared photometry derived from the TNG50 cosmological simulation and processed with the radiative transfer code SKIRT We show that nonparametric SFHs provide a more effective approach to mitigating the outshining effect by recent star formation, offering improved accuracy in the determination of galaxy stellar properties. Also, we find that the nonparametric SFH model at resolved scales closely recovers the stellar mass formation times (within 0.1 dex) and the ground truth values from TNG50, with an absolute average bias of $0.03$ dex in stellar mass and $0.01$ dex in both specific star formation rate and mass-weighted age. In contrast, larger offsets are estimated for all stellar properties and formation times when using a simple τ-model SFH, at both resolved and global scales, highlighting its limitations. These results emphasize the critical role of nonparametric SFHs in both global and spatially resolved analyses, as they better capture the complex evolutionary pathways of galaxies and avoid the biases inherent in simple parametric models.

The comoving cosmic star formation rate density peaks at z ∼ 2-3, with dusty star-forming galaxies being significant contributors to this peak. These galaxies are characterized by their high star formation rates and substantial infrared (IR) luminosities. The formation mechanisms remain an open question for these galaxies, particularly with respect to how such intense levels of star formation are triggered and maintained. We aim to resolve CO(3--2) emission toward two strongly lensed galaxies, SPT,0125-47 and SPT,2134-50, at z ∼ 2.5-2.8 to determine their morphology and physical properties. We used high-resolution ALMA band 3 observations of CO(3--2) emission toward both sources to investigate their properties. We performed parametric and nonparametric lens modeling using the publicly available lens modeling software PyAutoLens . We divided the CO(3--2) emission line into two bins corresponding to the red and blue portions of the emission line and nonparametrically modeled the source plane emission for both bins. We found that both sources are well described by a single Sérsic profile in both the parametric and nonparametric models of the source plane emission, in contrast to what was previously found for SPT,0125-47. Parametric lens modeling studies of the red and blue bins have reported distinctive differential magnification across the line spectrum. We performed a basic analysis of the morphology and kinematics in the source plane using nonparametric lens modeling of the red and blue bins. We found tentative evidence of a velocity gradient across both sources and no evidence of any clumpy structure, companions, or ongoing mergers. The previously calculated high star formation rates and low depletion times of both SPT,0125-47 and SPT,2134-50 suggest that these galaxies are undergoing a dramatic phase in their evolution. Given the lack of evidence of ongoing interactions or mergers in our source plane models, we suggest that the intense star formation was triggered by a recent interaction and/or merger. We also consider the possibility that these galaxies might be in the process of settling into disks.