Universal Kriging Model Research Articles

Modified Universal Kriging-based clearance error optimization for orthogonal robot

Abstract Aiming at the problem that the clearance error of the traditional tire laser engraving robot is too high, and thus the resulting positioning accuracy cannot meet the accuracy requirements needed for engraving at this stage, this paper takes the homemade orthogonal robot as the object of research, and researches the optimization method of the positioning error of the orthogonal robot based on the modified Kriging. In this paper, we use spinor theory to construct a robot clearance error model, establish an objective function with orthogonal robot clearance error as the optimization objective, and use a particle swarm algorithm to modify universal kriging (MUK) to optimize the positioning error. The optimization method is simulated and analyzed, and MUK is selected for comparison with other optimization methods. The simulation results show that 52% relative reduces the number of iterations of the MUK to the universal kriging model (UK), 79% relative to the particle Swarm Optimization (PSO), and 81% relative to the ordinary kriging model (OK), which makes the algorithm better applicable in comparison. At the same time, a tire laser engraving orthogonal robot experimental platform was built for physical verification experiments, and the experimental results show that the maximum error before and after the optimization of the integrated error of orthogonal robot positioning accuracy has been reduced by 89.91%; the average positioning error has been reduced by 86.90%; and the root-mean-square error has been reduced by 83.01%, which further proves the practicability of the proposed method.

Construction of orthogonal-MaxPro Latin hypercube designs

Orthogonal Latin hypercube designs (LHDs) and maximum projection (MaxPro) LHDs are widely used in computer experiments. They are efficient for estimating the trend part and the Gaussian process part of the universal Kriging (i.e., the Gaussian process) model, respectively, especially when only some of the factors are active. Yet, the orthogonality and the MaxPro criteria often do not agree with each other. In this work, we propose a new class of optimal designs, called orthogonal-MaxPro LHDs, optimizing a well-defined multi-objective criterion combining the correlation and the MaxPro metrics. An efficient parallel algorithm via level permutations and expansions is developed, whose efficiency is guaranteed by theories. Numerical results are presented to show that the construction is fast and the obtained designs are attractive, especially for large computer experiments.

Expert's experience-informed hierarchical kriging method for aerodynamic data modeling

Data-driven models, such as kriging, have gained popularity in aerospace engineering due to their capability of predicting multidimensional, nonlinear aerodynamic characteristics of an aircraft. However, they are still suffering from the problem associated with poor extrapolation capability and physical interpretability, which in turn has great impacts on aerodynamic performance and flight safety. To address this problem, this article proposes to incorporate an empirical aerodynamic model obtained from expert's understanding of aerodynamics in the fitting process of a kriging model. First, empirical aerodynamic models based on expert's understanding and experience are derived. Second, the regression term of a kriging model is replaced by the expert's experience-informed model so that the global trend can be consistent with physical laws and provides extra knowledge in the subregion(s) without training data, especially for the extrapolation region(s). Finally, the expert's experience-informed hierarchical kriging (EEI-HK) model is built in a sequential way. The proposed method is validated with analytical test examples and demonstrated by aerodynamic data modeling of an AGARD-B missile and an FDL-5A hypersonic flight vehicle. Results show that, compared with ordinary and universal kriging models, the proposed EEI-HK model can dramatically improve the prediction accuracy in the extrapolation domain and slightly enhance the interpolation accuracy, with assistance from physical information provided by an empirical model. In consequence, it can be a promising approach for aerodynamic data extrapolation and saving the cost of establishing an aerodynamic database.

Spatial and temporal analysis of African swine fever front-wave velocity in wild boar: implications for surveillance and control strategies.

The front-wave velocity of African swine fever (ASF) virus spread is depicted through a retrospective spatial and temporal analyses of wild boar outbreaks from Jan. 2014 to Jan. 2022 in Estonia, Latvia, Lithuania and Eastern Poland-regions responsible for more than 50% of all wild boar cases in the EU. The study uses empirical semivariograms in a universal kriging model to assess spatial autocorrelation in notification dates and identifies a discernable large-scale spatial trend. The critical parameter of ASF front-wave velocity was identified (Mean = 66.33 km/month, SD = 163.24) in the whole study area, and explored the variations across countries, wild boar habitat suitability, seasons, and the study period. Statistical differences in front-wave velocity values among countries and temporal clusters are explored, shedding light on potential factors influencing ASF transmission dynamics. The implications of these findings for surveillance and control strategies are discussed.

Associations between air pollution exposure and empirically derived subtypes of mild cognitive impairment in older women

AbstractBackgroundExposure to air pollution in older adulthood is a novel risk factor for mild cognitive impairment (MCI). MCI is a heterogenous construct with different subtypes likely reflecting different disease etiologies. Understanding if long‐term exposure to air pollution is differentially associated with MCI subtypes may provide insight into the neurotoxic effects of air pollution exposure on the aging brain.MethodParticipants included women from the Women’s Health Initiative Memory Study. Data from 651 women without dementia who screened positive for cognitive impairment on the modified mini‐mental state examination (3MS) and a randomly selected comparison sample of 2,560 women who were cognitively intact based on the 3MS were utilized. Participants were administered the 3MS annually (1996‐2008). Those who screened positive for cognitive impairment on the 3MS completed the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) cognitive battery and the Trailmaking Test parts A and B. Average annual levels of particulate matter of aerodynamic diameter<2.5 (PM2.5) and nitrogen dioxide (NO2) were estimated at the participant’s residence using regionalized national universal kriging models and averaged over the 3‐year period before the cognitive assessment. Latent class structural equation models (SEM) were constructed to identify MCI subtypes. We then examined associations between air pollution exposure and latent class membership using multinomial regression while adjusting for demographic, lifestyle, and clinical confounders.ResultFour significant MCI subtypes were identified (Figure 1): multidomain MCI (n = 147; 23%), amnestic MCI (n = 141; 22%), dysnomic/dysexecutive MCI (n = 200; 31%), and subtle MCI (n = 163; 25%). Women living in locations with higher levels of PM2.5 (per interquartile range [IQR] = 3.93 µg/m3; OR = 1.28; p = .04) or NO2 (per IQR = 9.25 ppb; OR = 1.36; p = .02) were more likely to be classified as having subtle MCI compared to normal comparisons. One IQR increase in NO2 was also associated with increased likelihood of being classified with dysnomic/dysexecutive MCI (OR = 1.38; p = .003) compared to normal comparisons.ConclusionOur study illustrates that long‐term exposure to PM2.5 and NO2 may be associated with specific subtypes of MCI. Our findings provide further evidence that air pollution may exert neurotoxic effects on brain areas underlying executive function and language ability in older women.

Statistical evaluation of stability margin of a multi-stage compressor with geometric variability using adaptive polynomial chaos-Kriging model

Compressed air energy storage systems must promptly adapt to power network demand fluctuations, necessitating a high surge margin in the compression system to ensure safety. It is challenging to completely eliminate blade geometric variations caused by limited machining precision, the important effects of which should be considered during aerodynamic shape design and production inspection. The present paper explores the uncertainty impact of geometric deviations on the stability margin of a multi-stage axial compressor at a low rotational speed. Initially, an adaptive polynomial chaos expansion-based universal Kriging model is introduced, and its superior response performance in addressing high-dimensional uncertainty quantification problems is validated through rigorous analytical and engineering tests. Then, this model is used to statistically evaluate the stability margin improvement (SMI) of the compressor due to the Gaussian and realistic geometric variabilities separately. The results show that the mean and standard deviation of SMI are −0.11% and 0.5% under the Gaussian geometric variability, while those are 0.33% and 0.39% under the realistic variability. For both the geometric variabilities, the stagger angle and maximum thickness deviations of the first-stage rotor are the most influential parameters controlling the uncertainty variations in the stability margin. Finally, the underlying impact mechanism of the influential geometric deviations is investigated. The variation in the stability margin caused by the geometric deviations primarily results from the alteration of inlet incidences, affecting the size of the tip leakage vortex blockage and boundary-layer separation regions near the blade tip of the first-stage rotor.

Modified Penalized Blind Kriging for efficiently selecting a global trend model

The Kriging model has been more and more widely used to replace complicated engineering models as surrogate models in a variety of areas within the engineering field. It is important to select a global trend for the universal Kriging model. Several regularization model selection strategies, such as the Penalized Blind Kriging (PBK) and the Lasso Kriging (LK), are the main approaches to solving this problem. LK avoids the iterative computation of PBK, but it does not use the correlation between sample points in selecting variables, which could not guarantee the effect of optimizing the prediction accuracy. In this paper, the Modified Penalized Blind Kriging (MPBK) is proposed to effectively overcome the above problems. Simulation and practical engineering case studies are used to show that the MPBK is an effective approach in both prediction accuracy and time efficiency.

Bayesian fixed-domain asymptotics for covariance parameters in a Gaussian process model

Gaussian process models typically contain finite-dimensional parameters in the covariance function that need to be estimated from the data. We study the Bayesian fixed-domain asymptotics for the covariance parameters in a universal kriging model with an isotropic Matérn covariance function, which has many applications in spatial statistics. We show that when the dimension of domain is less than or equal to three, the joint posterior distribution of the microergodic parameter and the range parameter can be factored independently into the product of their marginal posteriors under fixed-domain asymptotics. The posterior of the microergodic parameter is asymptotically close in total variation distance to a normal distribution with shrinking variance, while the posterior distribution of the range parameter does not converge to any point mass distribution in general. Our theory allows an unbounded prior support for the range parameter and flexible designs of sampling points. We further study the asymptotic efficiency and convergence rates in posterior prediction for the Bayesian kriging predictor with covariance parameters randomly drawn from their posterior distribution. In the special case of one-dimensional Ornstein–Uhlenbeck process, we derive explicitly the limiting posterior of the range parameter and the posterior convergence rate for asymptotic efficiency in posterior prediction. We verify these asymptotic results in numerical experiments.

A Spatial Model for Repairing of the Dam Safety Monitoring Data Combining the Variable Importance for Projection (VIP) and Cokriging Methods

The safe operation of dams is related to the lifeline of the national economy, the safety of the people, and social stability, and dam safety monitoring plays an essential role in scientifically controlling the safety of dams. Since the effects of environmental variables were not considered in conventional monitoring data repairing methods (such as the single time series model and spatial interpolation model), a spatial model for repairing monitoring data combining the variable importance for projection (VIP) method and cokriging was put forward in this paper. In order to improve the accuracy of the model, the influence of different combinations of covariates on it was discussed, and the VIPj value greater than 0.8 was proposed as the threshold of covariates. The engineering verification shows that the VIP-cokriging spatial model had the advantages of high precision and strong applicability compared with the inverse distance weighting (IDW) model, the ordinary kriging model, and the universal kriging model, and the overall error can be reduced by more than 60%, which could better realize the expansion of the monitoring effect variable to the whole area of the dam space. The engineering application of the PBG dam showed that the model scientifically correlated the existing monitoring points with the spatial location of the dam, and reasonably repaired the measured values of the stopping and abnormal measured points, effectively ensuring that the spatial regular of the monitoring data could truly reflect the actual safety and operational status of the dam.

Association between late‐life air pollution exposure and medial temporal lobe atrophy in older women

AbstractBackgroundAmbient air pollution exposures increase the risk of Alzheimer’s disease (AD) and related dementias, which may be driven by structural brain alterations in the medial temporal lobe (MTL). The MTL consists of subregions critical for memory and vulnerable to AD neuropathology, including the hippocampus, parahippocampal gyrus (PHG), entorhinal cortex (ERC), and amygdala. However, previous structural MRI studies examining the hypothesized adverse effects of exposure on MTL were cross‐sectional and largely focused on hippocampus which yielded mixed results, failing to provide insights on the underlying neuropathological processes affected by air pollution exposures.MethodWe conducted a longitudinal study including 653 cognitively‐normal community‐dwelling women from the Women’s Health Initiative Memory Study with two MRI brain scans (MRI‐1: 2005‐6; MRI‐2: 2010‐11; Meanage at MRI‐1 = 77.3±3.5). Using regionalized universal kriging models, exposures at residential locations were estimated as 3‐year annual averages of fine particulate matter (PM2.5) and nitrogen dioxide (NO2; traffic‐related pollution) prior to MRI‐1. Volumes of the total MTL and MTL subregions were derived using multi‐atlas region segmentation with ensembled registration. We used linear regressions to estimate exposure effects on 5‐year standardized volume changes, adjusting for intracranial volume, sociodemographic, lifestyle, and clinical characteristics. We also examined if associations differed between APOE4 carriers (n = 123) and non‐carriers (n = 463).ResultMTL volume decreased by 0.53±1.00cm3 over 5 years. For each interquartile increase of PM2.5 (3.26µg/m3) and NO2 (6.77ppb), adjusted MTL volume had greater shrinkage by 0.32cm3 (95%CI = [‐0.43,‐0.21]) and 0.12cm3 (95%CI = [‐0.23,‐0.01]) over 5 years, equivalent to respectively 17% (95%CI = [11%, 24%]) and 6% (95%CI = [0.6%, 12%]) increased dementia risk. For PHG, we found loss of 0.24cm3 (95%CI = [‐0.30,‐0.19]) and 0.09cm3 (95%CI = [‐0.14,‐0.04]) associated with PM2.5 and NO2 respectively. There was loss of 0.06cm3 (95%CI = [‐0.10,‐0.01]) in ERC associated with PM2.5, but not with NO2. No associations were found with amygdala or hippocampal atrophy. The observed adverse exposure effects on MTL atrophy did not differ by APOE4 genotype.ConclusionMTL atrophy in older women was associated with late‐life exposures to PM2.5 and NO2, and the putative adverse effects varied across brain subregions. The PHG and ERC—the MTL subregions where AD neuropathologies likely begin, may be preferentially vulnerable to air pollution neurotoxicity.

An improvement in data interpretation to estimate residual stresses and mechanical properties using instrumented indentation: A comparison between machine learning and Kriging model

The instrumented indentation method has been introduced as an effective means of estimating surface residual stresses. This technique has also been widely utilized to measure mechanical properties of the materials. Most studies in recent years have been conducted for the nano and micro-indentation. However, macro-indentation technique can result in interesting results in terms of residual stresses as well as mechanical properties. The instrumented indentation is a rapid and non-destructive method with acceptable precision which is desirable for a variety of different applications. This technique is applicable not only for large samples or those under service but also for small material volumes. In the author’s previous work, a portable indentation apparatus was examined for extracting the p–h curve. The fuzzy neural networks were then employed to determine the residual stresses as well as the plastic mechanical properties of materials. The current research uses different data interpretation methods to extract equi-biaxial residual stresses and mechanical properties. A comparative study between two methods, supervised machine learning and Kriging model, was conducted to estimate the unknown parameters. The performance accuracies of machine learning and Kriging were then compared. The results indicated that supervised machine learning using kNN algorithms can perform slightly better than the universal Kriging model. The findings were later compared with the experimental results for an aluminum plate.

Air quality improvement and cognitive decline in community-dwelling older women in the United States: A longitudinal cohort study.

Late-life exposure to ambient air pollution is a modifiable risk factor for dementia, but epidemiological studies have shown inconsistent evidence for cognitive decline. Air quality (AQ) improvement has been associated with improved cardiopulmonary health and decreased mortality, but to the best of our knowledge, no studies have examined the association with cognitive function. We examined whether AQ improvement was associated with slower rate of cognitive decline in older women aged 74 to 92 years. We studied a cohort of 2,232 women residing in the 48 contiguous US states that were recruited from more than 40 study sites located in 24 states and Washington, DC from the Women's Health Initiative (WHI) Memory Study (WHIMS)-Epidemiology of Cognitive Health Outcomes (WHIMS-ECHO) study. They were predominantly non-Hispanic White women and were dementia free at baseline in 2008 to 2012. Measures of annual (2008 to 2018) cognitive function included the modified Telephone Interview for Cognitive Status (TICSm) and the telephone-based California Verbal Learning Test (CVLT). We used regionalized universal kriging models to estimate annual concentrations (1996 to 2012) of fine particulate matter (PM2.5) and nitrogen dioxide (NO2) at residential locations. Estimates were aggregated to the 3-year average immediately preceding (recent exposure) and 10 years prior to (remote exposure) WHIMS-ECHO enrollment. Individual-level improved AQ was calculated as the reduction from remote to recent exposures. Linear mixed effect models were used to examine the associations between improved AQ and the rates of cognitive declines in TICSm and CVLT trajectories, adjusting for sociodemographic (age; geographic region; race/ethnicity; education; income; and employment), lifestyle (physical activity; smoking; and alcohol), and clinical characteristics (prior hormone use; hormone therapy assignment; depression; cardiovascular disease (CVD); hypercholesterolemia; hypertension; diabetes; and body mass index [BMI]). For both PM2.5 and NO2, AQ improved significantly over the 10 years before WHIMS-ECHO enrollment. During a median of 6.2 (interquartile range [IQR] = 5.0) years of follow-up, declines in both general cognitive status (β = -0.42/year, 95% CI: -0.44, -0.40) and episodic memory (β = -0.59/year, 95% CI: -0.64, -0.54) were observed. Greater AQ improvement was associated with slower decline in TICSm (βPM2.5improvement = 0.026 per year for improved PM2.5 by each IQR = 1.79 μg/m3 reduction, 95% CI: 0.001, 0.05; βNO2improvement = 0.034 per year for improved NO2 by each IQR = 3.92 parts per billion [ppb] reduction, 95% CI: 0.01, 0.06) and CVLT (βPM2.5 improvement = 0.070 per year for improved PM2.5 by each IQR = 1.79 μg/m3 reduction, 95% CI: 0.02, 0.12; βNO2improvement = 0.060 per year for improved NO2 by each IQR = 3.97 ppb reduction, 95% CI: 0.005, 0.12) after adjusting for covariates. The respective associations with TICSm and CVLT were equivalent to the slower decline rate found with 0.9 to 1.2 and1.4 to 1.6 years of younger age and did not significantly differ by age, region, education, Apolipoprotein E (ApoE) e4 genotypes, or cardiovascular risk factors. The main limitations of this study include measurement error in exposure estimates, potential unmeasured confounding, and limited generalizability. In this study, we found that greater improvement in long-term AQ in late life was associated with slower cognitive declines in older women. This novel observation strengthens the epidemiologic evidence of an association between air pollution and cognitive aging.

Association of improved air quality with lower dementia risk in older women

Late-life ambient air pollution is a risk factor for brain aging, but it remains unknown if improved air quality (AQ) lowers dementia risk. We studied a geographically diverse cohort of older women dementia free at baseline in 2008 to 2012 (n = 2,239, aged 74 to 92). Incident dementia was centrally adjudicated annually. Yearly mean concentrations of fine particulate matter (PM2.5) and nitrogen dioxide (NO2) were estimated using regionalized national universal kriging models and averaged over the 3-y period before baseline (recent exposure) and 10 y earlier (remote exposure). Reduction from remote to recent exposures was used as the indicator of improved AQ. Cox proportional hazard ratios (HRs) for dementia risk associated with AQ measures were estimated, adjusting for sociodemographic, lifestyle, and clinical characteristics. We identified 398 dementia cases during follow up (median = 6.1 y). PM2.5 and NO2 reduced significantly over the 10 y before baseline. Larger AQ improvement was associated with reduced dementia risks (HRPM2.5 0.80 per 1.78 μg/m3, 95% CI 0.71-0.91; HRNO2 0.80 per 3.91 parts per billion, 95% CI 0.71-0.90), equivalent to the lower risk observed in women 2.4 y younger at baseline. Higher PM2.5 at baseline was associated with higher dementia risk (HRPM2.5 1.16 per 2.90 μg/m3, 95% CI 0.98-1.38), but the lower dementia risk associated with improved AQ remained after further adjusting for recent exposure. The observed associations did not substantially differ by age, education, geographic region, Apolipoprotein E e4 genotypes, or cardiovascular risk factors. Long-term AQ improvement in late life was associated with lower dementia risk in older women.

On the classification error rates in terms of semivariograms for Gaussian universal kriging models

Bayes multiclass classification of spatial Gaussian data following the universal kriging model is considered. The closed-form expressions for the maximum likelihood (ML) estimator of regression parameters and the actual error rate (AER) in terms of semivariograms are derived.

Ambient Air Pollution and Increasing Depressive Symptoms in Older Women: The Mediating Role of The Prefrontal Cortex

Exposure to air pollution may accelerate brain aging and increase risk of late-life depressive symptoms (DS). Brain structures underlying these associations are unknown. Longitudinal data from 829 community-dwelling women without dementia (baseline age 81.6 ± 3.6 years old) who participated in both the Women’s Health Initiative Memory Study Magnetic Resonance Imaging study (WHIMS-MRI; 2005-06) and the WHIMS-Epidemiology of Cognitive Health Outcomes (2008-16) were analyzed to examine whether volumetric measures of brain structures mediated associations between long-term exposure to ambient air pollutants and annual increases in DS (as measured by annually assessed 15-item Geriatric Depression Scale). Annual PM2.5 (fine particulate matter of aerodynamic diameter <2.5) and NO2 were estimated at the participants’ residence using regionalized universal kriging models and aggregated to the 3-year average prior to the WHIMS-MRI. Structural equation models were constructed to estimate associations between exposure, structural brain variables, and trajectories of DS (standardized on baseline mean and SD). Living in locations with higher NO2 (standardized β = 0.023; 95% Confidence Interval (CI) = 0.004, 0.042) or PM2.5 (standardized β = 0.021; 95% CI = 0.004, 0.038) was associated with larger annual increases in DS (~60% larger annual increase in DS). Higher NO2, but not PM2.5, was associated with smaller prefrontal cortical volumes (standardized β = -0.431; 99% CI = -0.518; -0.344). Prefrontal cortical volume explained 30.4% of the total association between annual DS increase and NO2. These findings underscore the importance of the prefrontal cortex in associations between NO2 exposure and increasing DS in later-life.

Air Quality Improvement Is Associated With Decreasing Depressive Symptoms in Older Women

Air pollution exposure is an environmental risk factor in brain aging and may also be associated with late-life depressive symptoms (DS). It is unknown if air quality (AQ) improvement is associated with reductions in DS in later life. Longitudinal data from 917 cognitively intact women with no prior history of depression (baseline age 66.4 ± 1.6 years old) participating in the Women’s Health Initiative Memory Study of Younger Women (WHIMSY; 2008-2016) were analyzed to examine whether AQ improvement over the 5-years prior to WHIMSY baseline was associated with trajectories of DS (measured by 15-item Geriatric Depression Scale). Annual PM2.5 (fine particulate matter of aerodynamic diameter <2.5) and NO2 were estimated at the participants’ residence using regionalized universal kriging models. Estimates were aggregated to the 3-year average at 5 years (remote) and immediately (recent) before WHIMSY baseline. Associations between AQ improvement (difference between remote to recent exposure) and trajectories of DS were estimated using linear mixed effect models, adjusting for sociodemographic, lifestyle, and clinical covariates. AQ improved prior to baseline (PM2.5: 1.62 ± 1.45 μg/m3 and NO2: 3.70 ± 2.81 ppb). Women residing in locations with greater improvement in NO2 (per IQR = 4.34 ppb) or PM2.5 (per IQR = 2.30 μg/m3) reported significant annual reductions in DS (βNO2=3.1%, p=.046; βPM25=1.6%, p=.046), similar to the effect of engaging in moderate to vigorous physical activity four times or more a week. These findings suggest that improving air quality may reduce depressive symptoms in older women.

Associations between air pollution exposure and subtypes of cognitive performance in older women

AbstractBackgroundAir pollution exposure is a novel environmental risk factor in brain aging. It remains unclear whether long‐term exposure in older adulthood is associated with profiles of cognitive performance across multiple domains.MethodParticipants included 2,185 women (aged 66‐83 years) without dementia from the Women’s Health Initiative Study of Cognitive Aging. Baseline neuropsychological assessment included tests of verbal episodic memory (California Verbal Learning Test), figural memory (Benton’s Visual Recognition Test), language (letter and category fluency), attention (Digit Span Forward and Backwards), and visuospatial ability (Card Rotations Test). Testes were adjusted for demographic age, education, and race/ethnicity. Average annual levels of particulate matter of aerodynamic diameter&lt;2.5 (PM2.5) and nitrogen dioxide (NO2) were estimated at the participant’s residence using regionalized national universal kriging models and averaged over the 3‐year period before the baseline assessment. Latent class structural equation models (SEM) were constructed to identify latent classes of women who had similar patterns of performance across cognitive tests. After identifying the number of latent classes extract in the model, we constructed a second SEM to examine associations between air pollution exposure and latent class membership using multinomial regression while adjusting for demographic, lifestyle, and clinical confounders.ResultFive significant latent classes were identified (Figure 1): multidomain cognitive impairment (n=247; 11%), amnestic cognitive impairment (n=263; 12%), superior memory (n=407; 19%), superior attention (n=351; 16%), and average cognitive ability (n=917; 42%). Women living in locations with higher levels of PM2.5 (per interquartile range [IQR] = 3.63 μg/m3) or NO2 (per IQR = 10.33) were more likely to be classified as having amnestic cognitive impairment compared to having average cognitive ability (OR=1.27; p=.021). One IQR increase in NO2 was also associated with decreased likelihood of having superior memory (OR=.81; p=.024) compared to average cognitive ability.ConclusionOur cross‐sectional study illustrates that long‐term exposure to PM2.5 and NO2 may be associated with specific patterns of cognitive performance; mainly worse episodic memory relative to performance in other domains. Our findings provide further evidence that air pollution may exert neurotoxic effects on brain areas underlying episodic memory in older women.

Heterogeneous associations of air quality improvement with domain‐specific cognitive function in older women

AbstractBackgroundAir pollution is associated with increased risk of cognitive decline and dementia. However, whether and how air quality (AQ) improvement is associated with better cognitive performance across multiple domains is unknown.MethodParticipants included 2,114 women (aged 74‐92 at the enrollment) from the US‐based Women’s Health Initiative Memory Study‐Epidemiology of Cognitive Health Outcomes. Cognitive domains evaluated annually included episodic memory (EM; sum of scores by immediate and delayed recalls of East Boston Memory Test and the word list of modified Telephone Interview for Cognitive Status), language (Verbal Fluency‐Animals), working memory (WM, sum of scores from forward and backward Digit Span Tests), and attention/executive function (AEF, log‐transformed ratio of the time spent in Part A over Part B for Oral Trails Making Test), all standardized using corresponding baseline measure. Annual PM2.5 (fine particulate matter) and NO2 were estimated using regionalized universal kriging models and aggregated to 3‐year average at 10 years (remote) and immediately (recent) before enrollment. AQ improvement was defined as reduction from remote to recent exposures. We examined associations between AQ improvement and domain‐specific cognitive trajectories using linear mixed effect models, adjusting for sociodemographic, lifestyle, clinical covariates, and time‐varying propensity scores to account for selective attrition.ResultsAQ improved significantly over 10 years for both PM2.5 (13.3±2.7µg/m3 to 10.6±2.0µg/m3) and NO2 (15.7±7.1ppb to 10.4±4.9ppb). Over a median of 6.2 (inter‐quartile‐range [IQR]=5.1) years of follow‐up, cognitive performance declined significantly in all domains. Women residing in locations with greater PM2.5 improvement (per IQR=1.77µg/m3) had better performance in EM (βPM2.5=0.069, p=0.002), WM (βPM2.5=0.050, p=0.01), and AEF (βPM2.5=0.044, p=0.01), but not in language (βPM2.5=0.013, p=0.57). Similar findings with stronger associations were observed with improved NO2 (βNO2=0.065∼0.073 per IQR=3.89ppb for EM, WM and AEF, all p≤0.002). Slower declines in cognitive trajectories across EM, Language and AEF were found with improved AQ, but only the association between improved NO2 and slower EM decline was statistically significant (βNO2=0.010/year, p&lt;0.001).ConclusionsImproved AQ of PM2.5 and NO2 was associated with better performance of EM, WM and AEF, but not language. In addition, improved NO2 was associated with slower decline of EM over time in older women.

Association of lower dementia risk with improved air quality in older women

AbstractBackgroundLate‐life ambient air pollution is a novel risk factor for brain aging. Air quality improvement has been associated with improved cardiopulmonary health and reduced mortality; however, whether it is associated with reduced dementia risk is unknown.MethodWe studied a geographically‐diverse cohort from the Women’s Health Initiative Memory Study‐Epidemiology of Cognitive Health Outcomes (WHIMS‐ECHO). Participants included 2239 older women (aged 74‐92) without dementia at WHIMS‐ECHO enrollment in 2008‐12. Incident dementia cases, based on DSM‐IV criteria, were centrally‐adjudicated annually (2008‐18). Yearly mean concentrations of ambient PM2.5 (fine particulate matter) and NO2 at residential locations were estimated using validated regionalized universal kriging models. These annual estimates were then averaged over 3‐year periods at both immediately preceding (recent exposure) and 10 years prior to (remote exposure) WHIMS‐ECHO enrollment. Individual‐level improved air quality was defined as reduction from remote to recent exposures. We used Cox proportional hazard models to estimate the hazard ratios (HR) for dementia risk associated with air quality measures, adjusting for sociodemographic, lifestyle, and clinical characteristics. We incorporated inverse‐probability weighting to account for differential attrition during the follow up. Models were replicated using multiple imputation to include those with missing air pollution and covariate data as sensitivity analyses.ResultsWe identified 398 incident dementia cases during median 6.1 years of follow up (interquartile range [IQR]=5.2 years). PM2.5 and NO2 reduced significantly over the 10 years before WHIMS‐ECHO enrollment. Greater improvement in air quality was significantly associated with reduced dementia risks (HRPM2.5=0.80 per IQR=1.78µg/m3, 95% confidence interval [CI]=[0.71, 0.91]; HRNO2=0.80 per IQR=3.91ppb, 95%CI=[0.71, 0.90]), equivalent to the lower risk observed in women 2.4 years younger at baseline. There was some evidence that recent PM2.5 exposure was associated with higher dementia risk (HRPM2.5=1.16 per IQR=2.90µg/m3, 95%CI=[0.98, 1.38]), but the lower dementia risk associated with improved air quality remained after further adjusting for the recent exposure effects. The observed associations did not substantially differ by age, education, geographic region, APOE e4 genotypes, or cardiovascular risk factors. Similar results were found in the sensitivity analyses using multiple imputation.ConclusionsLong‐term air quality improvement in late‐life was associated with lower dementia risk in older women.

Association of air quality improvement with slower decline of cognitive function in older women

AbstractBackgroundLate‐life exposure to ambient air pollution is a modifiable risk factor for dementia, but epidemiological studies have shown inconsistent evidence for cognitive decline. Improved air quality (AQ) is associated with decreased mortality and improved respiratory health, but no studies have examined the association with cognitive function.MethodsParticipants included 2,232 U.S. community‐dwelling older women (aged 74‐92) from the Women’s Health Initiative Memory Study‐Epidemiology of Cognitive Health Outcomes (WHIMS‐ECHO) without dementia at enrollment. Measures of annual (2008‐18) cognitive function included general cognitive status (assessed by the Telephone Interview for Cognitive Status‐modified, TICSm) and episodic memory (assessed by the telephone‐based California Verbal Learning Test, CVLT). Regionalized universal kriging models were used to estimate annual exposures (1996‐2012) to fine particulate matter (PM2.5) and nitrogen dioxide (NO2) at residential locations, which were averaged over the 3‐year periods immediately preceding (recent exposure) and 10 years prior to (remote exposure) WHIMS‐ECHO enrollment. Individual‐level improvements in AQ were calculated as the reduction from remote to recent exposures. We used linear mixed effect models to examine the associations between improved AQ and the rates of cognitive decline in TICSm and CVLT trajectories, adjusting for sociodemographic, lifestyle, and clinical characteristics.ResultsAQ improved significantly for both PM2.5 and NO2 over the 10 years before enrollment. During a median of 6.2 (inter‐quartile‐range [IQR]=5.0) years of follow‐up, general cognitive status (βTICSm=‐0.42/year, 95% CI: ‐0.44, ‐0.40) and episodic memory (βCVLT=‐0.59/year, 95% CI: ‐0.64, ‐0.54) significantly declined. Greater AQ improvement was associated with slower decline in TICSm (βPM2.5=0.026/year per IQR=1.79 µg/m3, 95% CI: 0.001, 0.05; βNO2=0.034/year per IQR=3.92 ppb, 95% CI: 0.01, 0.06) and CVLT (βPM2.5=0.070/year per IQR=1.79 µg/m3, 95% CI: 0.02, 0.12; βNO2=0.060/year per IQR=3.97 ppb, 95% CI: 0.005, 0.12). The respective associations with TICSm and CVLT were equivalent to the slower declines found in women 0.9‐1.2 and 1.4‐1.6 years younger, and did not significantly differ by age, region, education, ApoE e4 genotypes, or cardiovascular risk factors. The benefits were attenuated after excluding incident dementia cases but remained after excluding prevalent or incident stroke cases.ConclusionsLong‐term improvement in AQ in late‐life was associated with slower cognitive declines in older women.