Markov Chain Monte Carlo Simulation Research Articles

Bayesian analysis of physiologically based toxicokinetic (PBTK) modeling for pentachlorophenol exposure in pregnant women

Pentachlorophenol (PCP) is a persistent organic compound that is widely present in the environment. The estimation of internal exposure levels for a given external exposure using toxicokinetic models is key to the human health risk assessment of PCP. The present study developed a physiologically based multicompartmental pharmacokinetic (PBTK) model to describe and predict the behavior of pentachlorophenol (PCP) in an organism. The model consists of stomach, intestines, adipose tissue, kidneys and fast- and poorly perfused tissues that are interconnected via blood circulation. We constructed a PBTK model of PCP in rats and extrapolated it to human dietary PCP exposure. The toxicokinetic data of PCP in human tissues and excreta were obtained from the published literature. Based on the collected PCP dietary survey and internal exposure data of pregnant women in Shanghai, Bayesian statistical analysis was performed for the model using Markov chain Monte Carlo (MCMC) simulation. The posterior distributions of the sensitive parameters were estimated, and the model was parameter optimized and validated using the pregnant women's test dataset. The results showed that the root mean square error (RMSE) improved 37.3% compared to the original model, and a systematic literature search revealed that the optimized model achieved acceptable prediction results for other datasets in China. A PCP metabolism model based on the exposure characteristics of pregnant women in China was constructed in the present study. The model provides a theoretical basis for the study of PCP toxicity and risk assessment.

NPEX: Never give up protein exploration with deep reinforcement learning

Elucidating unknown structures of proteins, such as metastable states, is critical in designing therapeutic agents. Protein structure exploration has been performed using advanced computational methods, especially molecular dynamics and Markov chain Monte Carlo simulations, which require untenably long calculation times and prior structural knowledge. Here, we developed an innovative method for protein structure determination called never give up protein exploration (NPEX) with deep reinforcement learning. The NPEX method leverages the soft actor-critic algorithm and the intrinsic reward system, effectively adding a bias potential without the need for prior knowledge. To demonstrate the method's effectiveness, we applied it to four models: a double well, a triple well, the alanine dipeptide, and the tryptophan cage. Compared with Markov chain Monte Carlo simulations, NPEX had markedly greater sampling efficiency. The significantly enhanced computational efficiency and lack of prior domain knowledge requirements of the NPEX method will revolutionize protein structure exploration.

System-Level Offshore Wind Energy and Hydrogen Generation Availability and Operations and Maintenance Costs

With the current trends of wind energy already playing a major part in the Scottish energy supply, the capacity of wind farms is predicted to grow exponentially and reach further depths offshore. However, a key challenge that presents itself is the integration of large producing assets into the current UK grid. One potential solution to this is green hydrogen production, which is being heavily researched in industry, with many concepts being investigated for large-scale purposes. However, the operations and maintenance (O&M) costs and availability of green hydrogen systems need to be quantified to ensure economical and technical viability, which is sparse in the available literature. The study presented in this paper investigated the availability and O&M costs of coupled wind–hydrogen systems by attempting to quantify the failure rates, repair times, repair costs and number of technicians required for key green hydrogen components. This study also utilised an O&M model created by the University of Strathclyde, which uses Monte Carlo Markov chain simulations to produce the O&M outputs. A number of assumptions were made throughout the study in relation to the O&M model inputs, and the baseline availability for the coupled wind–hydrogen system was 85.24%. Whilst the wind turbine still contributed a major part to the downtime seen in the simulations, the combined hydrogen system also contributed a significant amount, a total of 37%, which could have been due to the technology readiness levels of some the components included in the hydrogen system.

Searching for magnetar binaries disrupted by core-collapse supernovae

ABSTRACT Core-collapse supernovae (CCSNe) are considered the primary magnetar formation channel, with 15 magnetars associated with supernova remnants (SNRs). A large fraction of these should occur in massive stellar binaries that are disrupted by the explosion, meaning that $\sim 45~{{\ \rm per\ cent}}$ of magnetars should be nearby high-velocity stars. Here, we conduct a multiwavelength search for unbound stars, magnetar binaries, and SNR shells using public optical (uvgrizy bands), infrared (J, H, K, and Ks bands), and radio (888 MHz, 1.4 GHz, and 3 GHz) catalogues. We use Monte Carlo analyses of candidates to estimate the probability of association with a given magnetar based on their proximity, distance, proper motion, and magnitude. In addition to recovering a proposed magnetar binary, a proposed unbound binary, and 13 of 15 magnetar SNRs, we identify two new candidate unbound systems: an OB star from the Gaia catalogue we associate with SGR J1822.3−1606, and an X-ray pulsar we associate with 3XMM J185246.6 + 003317. Using a Markov Chain Monte Carlo simulation that assumes all magnetars descend from CCSNe, we constrain the fraction of magnetars with unbound companions to $5\lesssim f_u \lesssim 24~{{\ \rm per\ cent}}$, which disagrees with neutron star population synthesis results. Alternate formation channels are unlikely to wholly account for the lack of unbound binaries as this would require $31\lesssim f_{nc} \lesssim 66~{{\ \rm per\ cent}}$ of magnetars to descend from such channels. Our results support a high fraction ($48\lesssim f_m \lesssim 86~{{\ \rm per\ cent}}$) of pre-CCSN mergers, which can amplify fossil magnetic fields to preferentially form magnetars.

Nonstationary frequency analysis of extreme precipitation: Embracing trends in observations

Knowledge of the recurrence intervals of precipitation extremes is vital for infrastructure design, risk assessment, and insurance planning. However, trends and shifts in rainfall patterns globally pose challenges to the application of extreme value analysis (EVA) which relies critically on the assumption of stationarity. In this paper, we explore: (1) the suitability of nonstationary (NS) models in the presence of statistically significant trends, and (2) their potential in modeling out-of-sample data to improve frequency analysis of extreme precipitation. We analyze the benefits of using a nonstationary Generalized Extreme Value (GEV) model for annual extreme precipitation records from Southern Brazil. The location of the GEV distribution is allowed to change with time. The unknown GEV model parameters are estimated using Bayesian techniques coupled with Markov chain Monte Carlo simulation. Next, we use GAME sampling to compute the evidence (and their ratios, the so-called Bayes factors) for stationary and nonstationary models of annual maximum precipitation. Our results show that the presence of a statistically significant trend in annual maximum precipitation alone does not justify the use of a NS model. The location parameter of the GEV distribution must also be well defined, otherwise, stationary models of annual maximum precipitation receive more support by the data. These findings reiterate the importance of accounting for GEV model parameters and predictive uncertainty in frequency analysis and hypothesis testing of annual maximum precipitation data records. Furthermore, a meaningful EVA demands detailed knowledge about the origin and persistence of observed changes.

Bayesian data-driven framework for structural health monitoring of composite structures under limited experimental data

Ultrasonic-guided waves can be used to monitor the health of thin-walled structures. However, the run of experimental damage tests on materials like carbon fiber-reinforced plastics can be impractical and costly. Instead, numerical models can be used to create hybrid datasets to train machine learning algorithms, integrating data from numerical and experimental tests. This paper presents a Bayesian-driven framework to compensate for limited experimental data regarding Lamb wave propagation in composite plates. Using Bayesian inference, the framework updates a numerical finite element model, considering observed uncertainties by sampling posterior probability density functions for input parameters using Markov–Chain Monte Carlo simulations with the Metropolis-Hastings algorithm. A neural network surrogate model speeds-up these simulations, leading to a model that replicates the uncertain experimental setup. This model then generates data to augment true experimental data. Finally, a one-dimensional convolutional neural network is trained on a three different datasets to analyze Lamb wave signals and assess damage. Comparing training strategies shows the hybrid dataset augmented by samples generated by the updated FE model gives the most accurate damage size predictions.

First ALMA observations of the HD 105211 debris disc: A warm dust component close to a gigayear-old star

Most debris discs consist of a gas-poor, cold dust belt located tens to hundreds of astronomical units away from the host star. Many cold dust belts exhibit distinct structures attributed to the dynamic interaction of planetary systems. Moreover, in a few systems, additional warm components can be found closer to the central star, resembling the asteroid belt or zodiacal dust in our Solar System. In this work, we investigate the structure of the disc surrounding the nearby F2V star HD 105211 (eta Cru, HIP 59072), which has a warm excess (seen with Spitzer /MIPS at 24 mu m) and a potential asymmetry in the cold belt (seen in the Herschel /PACS images). We applied the CASA pipeline to obtain the ALMA 1.3 mm continuum images of HD105211. Then we constructed the spectral energy distribution (SED) of the system and performed Markov Chain Monte Carlo (MCMC) simulations to fit a model to the ALMA visibility data. To characterise the disc asymmetry, we analysed the ALMA images of two individual observation blocks (OB1, conducted on 2017 March 28, and OB2, conducted on 2018 May 9) and compared them to the previous Herschel images. Our modelling reveals that the disc around HD105211 is a narrow ring (23.6pm 4.6 au) with low eccentricity ($e positioned at a distance of 133.7pm 1.6 au from the central star, which differs from the broad disc ($100 au) starting at an inner edge of $87 au, inferred from the Herschel images. We found that both observation blocks show excess emission at the stellar position ($>3 while OB1 shows an offset between the star and the phase centre ($ and OB2 shows brightness clumps ($ We used a two-temperature model to fit the infrared SED and used the ALMA detection to constrain the warm component to a nearly pure blackbody model. The relatively low ratio of actual radius to blackbody radius of the HD105211 debris disc indicates that this system is depleted in small grains, which could indicate that it is dynamically cold. The excess emission from the stellar position suggests that there should be a warm millimetre-sized dust component close to the star, for which we suggest two possible origins: in situ asteroid belt or comet delivery.

On State Transition Probability and Performance of Direct Acyclic Graph Based Ledgers

Direct acyclic graph (DAG) based ledgers with multi-chain structures aim to solve the technical bottlenecks associated with classical blockchain technologies in the Internet of Things (IoT). The basic working principle of DAG-based ledgers is to validate new transactions by previous transactions in order to be added to the system. During the tip selection process in the unsteady regime, the state transition probability refers to the probability of a transaction changing from the initial state to an arbitrary state. The state transition probability plays an indispensable role in the performance and security analysis of the IoT relying on DAG-based ledgers. In this paper, we derive the exact expression and an approximate expression of the state transition probability, which both are in closed form. In addition, we propose and analyze three performance metrics, i.e., the expected cumulative weight, the expected number of steps, and the confirmation failure probability, which are derived from the state transition probability and greatly enrich the performance analysis and evaluation of the IoT. Markov chain Monte Carlo (MCMC) simulations are carried out to verify the derived analytical results and provide insight into the IoT using DAG-based ledgers.

A Bayesian approach based on discounting factor for consistency assessment in multi-regional clinical trial

ABSTRACT Multi-regional clinical trial (MRCT) has become an increasing trend for its supporting simultaneous global drug development. After MRCT, consistency assessment needs to be conducted to evaluate regional efficacy. The weighted Z-test approach is a common consistency assessment approach in which the weighting parameter W does not have a good practical significance; the discounting factor approach improved from the weighted Z-test approach by converting the estimation of W in original weighted Z-test approach to the estimation of discounting factor D. However, the discounting factor approach is an approach of frequency statistics, in which D was fixed as a certain value; the variation of D was not considered, which may lead to un-reasonable results. In this paper, we proposed a Bayesian approach based on D to evaluate the treatment effect for the target region in MRCT, in which the variation of D was considered. Specifically, we first took D random instead of fixed as a certain value and specified a beta distribution for it. According to the results of simulation, we further adjusted the Bayesian approach. The application of the proposed approach was illustrated by Markov Chain Monte Carlo simulation.

Relative Efficacies of Interventions to Improve the Quality of Screening-Related Colonoscopy: A Systematic Review and Network Meta-Analysis of Randomized Controlled Trials

Significant variability exists in colonoscopy quality indicators, including adenoma detection rate (ADR). We synthesized evidence from randomized trials in a network meta-analysis on interventions to improve colonoscopy quality. We included trials from database inceptions to September 25, 2023, of patients undergoing screening-related colonoscopy and presented efficacies of interventions within domains (periprocedural parameters, endoscopist-directed interventions, intraprocedural techniques, endoscopic technologies, distal attachment devices, and additive substances) compared to standard colonoscopy. The primary outcome was ADR. We used a Bayesian random-effects model using Markov-chain Monte Carlo simulation, with 10,000 burn-ins and 100,000 iterations. We calculated odds ratios with 95% credible intervals and present surface under the cumulative ranking (SUCRA) curves. We included 124 trials evaluating 37 interventions for the primary outcome. Nine interventions resulted in statistically significant improvements in ADR compared to standard colonoscopy (9-minute withdrawal time, dual observation, water exchange, i-SCAN [Pentax Ltd], linked color imaging, computer-aided detection, Endocuff [Olympus Corp], Endocuff Vision [Olympus Corp], and oral methylene blue). Dual observation (SUCRA, 0.84) and water exchange (SUCRA, 0.78) ranked highest among intraprocedural techniques; i-SCAN (SUCRA, 0.95), linked color imaging (SUCRA, 0.85), and computer-aided detection (SUCRA, 0.78) among endoscopic technologies; WingCap (A&A Medical Supply LLC) (SUCRA, 0.87) and Endocuff (SUCRA, 0.85) among distal attachment devices and oral methylene blue (SUCRA, 0.94) among additive substances. No interventions improved detection of advanced adenomas, and only narrow-band imaging improved detection of serrated lesions (odds ratio, 2.94; 95% credible interval, 1.46-6.25). Several interventions are effective in improving adenoma detection and overall colonoscopy quality, many of which are cost-free. These results can inform endoscopists, unit managers, and endoscopy societies on relative efficacies.

Uncertainty analysis of greenhouse gas emissions of monorail transit during the construction.

This paper examines the uncertainty of greenhouse gas (GHG) emissions during monorail construction. Firstly, a deterministic analysis is conducted. Subsequently, the obtained data are evaluated using the data quality indicator (DQI), and a Markov chain Monte Carlo (MCMC) simulation method is employed to assume different parameter distributions. The results of the deterministic calculation indicate that the calculated emissions per unit area of the station amount to 1.97 ton CO2e/m2, while the calculated emissions per unit section length reach 7.55 ton CO2e/m2. To simulate parameter distribution, we utilize a Beta distribution with good shape applicability. Furthermore, we establish scenarios involving system boundary reduction, low-emission factors, and reduced material and energy inputs in order to analyze scenario uncertainties. Regarding model uncertainty, this paper assumes that the material and energy quantity data conform to the normal, log-normal, uniform, and triangular distributions, respectively, subsequently analyzing the uncertainty distributions. This paper analyzes the GHG emission uncertainty evaluation of 16 monorail stations and sections during the construction period, which is divided into parameter, scenario, and model uncertainty. We provide a concrete framework for studying uncertainties related to GHG emissions at stations and sections during the monorail construction period. The scenario analysis results will help to make decisions about the choice of parameters, system boundaries, and other settings. It provides new guidance for emission reduction policies, such as reducing the use of steel-related products or using alternative environmentally friendly materials, considering emission reduction factors more comprehensively and setting emission reduction factors according to uniform distribution principle as far as possible.

A new selection criteria to optimize growth in animal breeding programs

Pedigree records and longitudinal measurements of live weight from 2628 buffaloes were analyzed. The aim of this research was to propose a new selection criteria, the Area Under the Growth Curve (AUGC), derived from a growth curve-based model. A hierarchical Bayesian approach with two levels was employed. In the first level, the growth trajectory was modeled using a fourth-degree polynomial, while in the second level, each parameter of the polynomial function was treated as a dependent variable influenced by environmental and genetic effects. The animal model included sex, dams’ parity and contemporary group (herd-year-season) as fixed effects, and relationships among animals as a random effect. Inference was conducted using Markov Chain Monte Carlo (MCMC) simulation algorithm. The proposed AUGC is interesting for use in selection programs because it allows breeders to identify heavier animals with lower risk in the production system. Additionally, that trait showed moderate to high heritabilities from weaning onwards, providing a useful new tool for cattle selection in the post-weaning phases.

Probabilistic characterisation of rock mass deformation modulus from P-wave velocity based on information criteria

ABSTRACT Determination of rock mass deformation modulus (Erm ) is an important but very challenging task in rock mechanics and rock engineering. Direct measurements of Erm by in-situ tests are expensive, time-consuming, and sometimes even impossible. Since the nondestructive and easy geophysical method is commonly used to measure the P-wave velocity (Vp ) in the field, the study has developed an empirical correlation between Erm and Vp based on 834 collected datasets by considering both inherent and transformation uncertainties. The information criteria are employed to select the best model for deriving a general empirical correlation. The maximum likelihood estimation (MLE) method and computational intelligence techniques like particle swarm optimisation (PSO), artificial bee colony (ABC), and differential evolution (DE) algorithms are used to specify the incorporated parameters of each candidate model. Based on the derived general empirical equation, a Bayesian equivalent sample approach using Markov Chain Monte Carlo (MCMC) simulation is applied to conduct probabilistic characterisation of Erm by combining prior knowledge and project-specific observed Vp data. Comparative studies of the results from theory and in-situ measured Erm values suggest that the proposed method is effective and validate the applicability of the derived general empirical correlation.

OLYMPUS-POPGEN: A synthetic population generation model to represent urban populations for assessing exposure to air quality.

With the new individual- and activity-based approaches to simulating exposure to air pollutants, exposure models must now provide synthetic populations that realistically reflect the demographic profiles of individuals in an urban territory. Demographic profiles condition the behavior of individuals in urban space (activities, mobility) and determine the resulting risks of exposure and environmental inequalities. In this context, there is a strong need to determine the relevance of the population modeling methods to reproduce the combinations of socio-demographic parameters in a population from the existing databases. The difficulty of accessing complete, high-resolution databases indeed proves to be very limiting for the ambitions of the different approaches. This work proposes to evaluate the potential of a statistical approach for the numerical modeling of synthetic populations, at the scale of dwellings and including the representation of coherent socio-demographic profiles. The approach is based on and validated against the existing open databases. The ambition is to be able to build upon such synthetic populations to produce a comprehensive assessment of the risk of environmental exposure that can be cross-referenced with lifestyles, indicators of social, professional or demographic category, and even health vulnerability data. The approach implemented here is based on the use of conditional probabilities to model the socio-demographic properties of individuals, via the deployment of a Monte Carlo Markov Chain (MCMC) simulation. Households are assigned to housing according to income and house price classes. The resulting population generation model was tested in the Paris region (Ile de France) for the year 2010, and applied to a population of almost 12 million individuals. The approach is based on the use of census and survey databases. Validation, carried out by comparison with regional census data, shows that the model accurately reproduces the demographic attributes of individuals (age, gender, professional category, income) as well as their combination, at both regional and sub-municipal levels. Notably, population distribution at the scale of the model buildings remains consistent with observed data patterns. The outcomes of this work demonstrate the ability of our approach to create, from public data, a coherent synthetic population with broad socio-demographic profiles. They give confidence for the use of this approach in an activity-based air quality exposure study, and thus for exploring the interrelations between social determinants and environmental risks. The non-specific nature of this work allows us to consider its extension to broader demographic profiles, including health indicators, and to different study regions.

Hospitalizations for Cardiovascular Diseases Sensitive to Primary Health Care in Paraná State, Brazil: A Bayesian Spatiotemporal Model.

To analyze spatiotemporal trends in hospitalizations for cardiovascular diseases (CVD) sensitive to primary health care (PHC) among individuals aged 50-69 years in Paraná State, Brazil, from 2014 to 2019 and investigate correlations between PHC services and the Social Development Index. We conducted a cross-sectional ecological study using publicly available secondary data to analyze the municipal incidence of hospitalizations for CVD sensitive to PHC and to estimate the risk of hospitalization for this group of diseases and associated factors using hierarchical Bayesian spatiotemporal modeling with Markov chain Monte Carlo simulation. There was a 5% decrease in the average rate of hospitalizations for PHC-sensitive CVD from 2014 to 2019. Regarding standardized hospitalization rate (SHR) according to population size, we found that no large municipality had an SHR >2. Likewise, a minority of these municipalities had SHR values of 1-2 (33%). However, many small and medium-sized municipalities had SHR values >2 (47% and 48%, respectively). A greater Social Development Index value served as a protective factor against hospitalizations, with a relative risk of 0.957 (95% credible interval, 0.929-0.984). The annual risk of hospitalization decreased over time; however, small municipalities had the greatest rates of hospitalization, indicating an increase in health inequity. The inverse association between social development and hospitalizations for CVD sensitive to PHC raises questions about intersectionality in health care.

Inference for Bayesian Nonparametric Models with Binary Response Data via Permutation Counting

Since the beginning of Bayesian nonparametrics in the early 1970s, there has been a wide interest in constructing models for binary response data. Such data arise naturally in problems dealing with bioassay, current status data and sensitivity testing, and are equivalent to left and right censored observations if the inputs are one-dimensional. For models based on the Dirichlet process, inference is possible via Markov chain Monte Carlo (MCMC) simulations. However, there exist multiple processes based on different principles, for which such MCMC-based methods fail. Examples include logistic Gaussian processes and quantile pyramids. These require MCMC for posterior inference given exact observations, and thus become intractable when the data comprise both left and right censored observations. Here we present a new importance sampling algorithm for nonparametric models given exchangeable binary response data. It can be applied to any model from which samples can be generated, or even only approximately generated. The main idea behind the algorithm is to exploit the symmetries introduced by exchangeability. Calculating the importance weights turns out to be equivalent to evaluating the permanent of a certain class of (0,1)-matrix, which we prove can be done in polynomial time by deriving an explicit algorithm.

Probabilistic fatigue stiffness variation of angle-ply GFRP composites considering stress ratio effect

Fiber-reinforced polymer (FRP) composites are frequently used in bridges, wind power installations, and marine floating structures, and there is a growing interest in understanding their fatigue behavior. However, due to a considerable dispersion of the fatigue properties of FRP composites, reliable tools for probabilistic prediction of the fatigue stiffness are required. Thus, in this work, it is proposed a method that uses Bayesian inference and Markov Chain Monte Carlo (MCMC) simulations to estimate the degradation of the stiffness in FRP composites. This methodology is suggested not only to predict the fatigue residual stiffness, but also to consider the reliability of such predictions. Currently, the existing models for stiffness degradation in FRP composites need to fit the model parameters separately under different stress ratios. To address this, a correction term is proposed to describe the relationship between the stress level, stress ratio, fatigue cycles, and residual stiffness. The results achieved validate and prove the effectiveness of the suggested stiffness model, which incorporates four material parameters to calculate the residual fatigue stiffness of FRP composites.

Domain-decomposed Bayesian inversion based on local Karhunen-Loève expansions

In many Bayesian inverse problems the goal is to recover a spatially varying random field. Such problems are often computationally challenging especially when the forward model is governed by complex partial differential equations (PDEs). The challenge is particularly severe when the spatial domain is large and the unknown random field needs to be represented by a high-dimensional parameter. In this paper, we present a domain-decomposed method to attack the dimensionality issue and the method decomposes the spatial domain and the parameter domain simultaneously. On each subdomain, a local Karhunen-Loève (KL) expansion is constructed, and a local inversion problem is solved independently in a parallel manner, and more importantly, in a lower-dimensional space. After local posterior samples are generated through conducting Markov chain Monte Carlo (MCMC) simulations on subdomains, a novel projection procedure is developed to effectively reconstruct the global field. In addition, the domain decomposition interface conditions are dealt with an adaptive Gaussian process-based fitting strategy. Numerical examples are provided to demonstrate the performance of the proposed method.

Long‐term scenario generation of renewable energy generation using attention‐based conditional generative adversarial networks

AbstractLong‐term scenario generation of renewable energy is regarded as an important part of the optimal planning of renewable energy systems. This study proposes a scenario generation method for generating long‐term correlated scenarios of wind and photovoltaic outputs from historical renewable energy data. The generation of scenarios was divided into two processes: long‐term yearly sequence generation and intraday scenario generation of wind‐solar energy. In the long‐term yearly sequence generation process, the k‐means clustering algorithm and Markov chain Monte Carlo simulation method were developed to capture the seasonal and long‐term features of wind and photovoltaic energies. Furthermore, an attention‐based conditional generative adversarial network (ACGAN) was proposed to capture short‐term features. An attention structure and conditional classifiers were developed to capture features in the generated scenarios. To accelerate the convergence process and improve the quality of the generated scenarios, a gradient penalty was included in the ACGAN model. Numerical case studies were conducted to verify the validity of the proposed method using a real‐world dataset.

Mortality in Women with Coronary Artery Disease in Paraná State, Brazil: A Bayesian Spatiotemporal Analysis.

Mortality resulting from coronary artery disease (CAD) among women is a complex issue influenced by many factors that encompass not only biological distinctions but also sociocultural, economic, and healthcare-related components. Understanding these factors is crucial to enhance healthcare provisions. Therefore, this study seeks to identify the social and clinical variables related to the risk of mortality caused by CAD in women aged 50 to 79 years old in Paraná state, Brazil, between 2010 and 2019. This is an ecological study based on secondary data sourced from E-Gestor, IPARDES, and DATASUS. We developed a model that integrates both raw and standardized coronary artery disease (CAD) mortality rates, along with sociodemographic and healthcare service variables. We employed Bayesian spatiotemporal analysis with Markov Chain Monte Carlo simulations to assess the relative risk of CAD mortality, focusing specifically on women across the state of Paraná. A total of 14,603 deaths from CAD occurred between 2010 and 2019. Overall, temporal analysis indicates that the risk of CAD mortality decreased by around 22.6% between 2010 (RR of 1.06) and 2019 (RR of 0.82). This decline was most prominent after 2014. The exercise stress testing rate, accessibility of cardiology centers, and IPARDES municipal performance index contributed to the reduction of CAD mortality by approximately 4%, 8%, and 34%, respectively. However, locally, regions in the Central-West, Central-South, Central-East, and Southern regions of the Central-North parts of the state exhibited risks higher-than-expected. In the last decade, CAD-related deaths among women in Paraná state decreased. This was influenced by more exercise stress testing, better access to cardiology centers, improved municipal performance index. Yet, elevated risks of deaths persist in certain regions due to medical disparities and varying municipal development. Therefore, prioritizing strategies to enhance women's access to cardiovascular healthcare in less developed regions is crucial.