Model Predictive Approach Research Articles

A two-stage spatial prediction modeling approach based on graph neural networks and neural processes

Spatial prediction models hold significant application value in fields such as environmental science, economic development, and geological exploration. With advancements in deep learning technology, graph neural networks (GNNs) offer a powerful and scalable solution for spatial data modeling. In these models, each spatial location is represented as a vertex, with the explanatory variables at each point converted into vertex features, and the vertex label corresponding to the target value at that point. GNNs utilize this representation to predict vertex labels, thereby performing spatial prediction. However, the residuals of GNN predictions are found to still exhibit spatial autocorrelation, indicating that traditional GNNs only achieve suboptimal results in terms of capturing complex spatial relationships. In this paper, we propose a two-stage spatial prediction method called Location Embedded Graph Neural Networks-Residual Neural Processes (LEGNN-RNP) to address this challenge. In the first stage, we employ LEGNNs, a new framework that integrates location features into GNNs through an attention mechanism, enhancing the learning of complex spatial models by considering both spatial context and attribute features. In the second stage, we model the residuals from LEGNN predictions to further extract spatial patterns from the data. Specifically, we introduce RNP, a neural process-based approach to model the Gaussian Process (GP) distribution of residuals. The distribution parameters (i.e., mean and covariance) are parameterized by a neural network, where the mean estimates the residual and the variance quantifies the prediction uncertainty. Experiments on four datasets demonstrate that our proposed two-stage method achieves state-of-the-art results by effectively extracting spatial relationships and significantly improving spatial prediction accuracy.

Risk-aware scheduling and dispatch of flexibility events in buildings

Residential and commercial buildings, equipped with systems such as heat pumps (HPs), hot water tanks, or stationary energy storage, have a large potential to offer their consumption flexibility as grid services. In this work, we leverage this flexibility to react to consumption requests related to maximizing self-consumption and reducing peak loads. We employ a data-driven virtual storage modeling approach for flexibility prediction in the form of flexibility envelopes for individual buildings. The risk-awareness of this prediction is inherited by the proposed scheduling algorithm. A Mixed-integer Linear Program (MILP) is formulated to schedule the activation of a pool of buildings in order to best respond to an external aggregated consumption request. This aggregated request is then dispatched to the active individual buildings, based on the previously determined schedule. The effectiveness of the approach is demonstrated by coordinating up to 500 simulated buildings using the Energym Python library and observing about 1.5 times peak power reduction in comparison with a baseline approach while maintaining comfort more robustly. We demonstrate the scalability of the approach by solving problems with 2000 buildings in about 21 s, with solving times being approximately linear in the number of considered assets.

TEMPRO: nanobody melting temperature estimation model using protein embeddings

Single-domain antibodies (sdAbs) or nanobodies have received widespread attention due to their small size (~ 15 kDa) and diverse applications in bio-derived therapeutics. As many modern biotechnology breakthroughs are applied to antibody engineering and design, nanobody thermostability or melting temperature (Tm) is crucial for their successful utilization. In this study, we present TEMPRO which is a predictive modeling approach for estimating the Tm of nanobodies using computational methods. Our methodology integrates various nanobody biophysical features to include Evolutionary Scale Modeling (ESM) embeddings, NetSurfP3 structural predictions, pLDDT scores per sdAb region from AlphaFold2, and each sequence’s physicochemical characteristics. This approach is validated with our combined dataset containing 567 unique sequences with corresponding experimental Tm values from a manually curated internal data and a recently published nanobody database, NbThermo. Our results indicate the efficacy of protein embeddings in reliably predicting the Tm of sdAbs with mean absolute error (MAE) of 4.03 °C and root mean squared error (RMSE) of 5.66 °C, thus offering a valuable tool for the optimization of nanobodies for various biomedical and therapeutic applications. Moreover, we have validated the models’ performance using experimentally determined Tms from nanobodies not found in NbThermo. This predictive model not only enhances nanobody thermostability prediction, but also provides a useful perspective of using embeddings as a tool for facilitating a broader applicability of downstream protein analyses.

Organizational and Individual Contributing Factors to Safety Climate in Healthcare Industries - Bayesian Network Predictive Modeling Approach.

The current study aims to identify individual and joint drivers that significantly influence the safety climate in healthcare industries by using Bayesian Network simulations for an in-depth analysis. Survey data were collected from 452 employees from two branches of one hospital in China for a study about workplace safety. The original English surveys were translated into Chinese using the back-translation procedure recommended by Brislin. Employees were asked to complete two online surveys with one month in between each administration. The sample was 42% doctors and 58% nurses. A Bayesian Network (BN) model, based on theory, was updated and complemented with expert knowledge. A graphical model based on expert knowledge and data-driven machine learning approaches was used to refine the BN structure, representing interrelationships among all studied variables. The BN model was employed to identify the best key drivers and joint strategies for safety climate improvement. The Bayesian Network model demonstrated a good overall fit. The Euclidean distance metric was used to assess the influence between connected variables, with interpersonal trust and locus of control having the strongest independent effects on safety climate among the five contributing factors. Joint strategies, particularly joint optimization of error disclosure culture and interpersonal trust, as well as error disclosure culture and self-efficacy, were most effective in promoting a safe climate. The findings suggest that hospital safety climate can be improved by providing a psychologically safe error disclosure culture and enhancing interpersonal trust among employees and their self-efficacy.

Predictive modeling of combined cycle power plant performance using a digital twin-based neural ODE approach

This study presents a novel digital twin-based predictive modeling approach to enhance the operation, maintenance, and management of combined cycle power plants. The research aimed to accurately forecast the combined cycle power (CCP) output, a critical performance indicator, to support intelligent decision-making for plant operators and engineers. The proposed framework leverages a neural ordinary differential equation (NODE) model integrated within a digital twin architecture to capture the complex, nonlinear dynamics of combined cycle gas turbine unit (CCGU) system. The predictive model was trained and validated using multivariate time-series data collected from a CCGU, achieving high accuracy with an R2_score of 0.993, a mean absolute percentage error of 0.325 %, and a root mean squared error of 1.518. Importantly, the NODE model demonstrated superior forecasting capabilities compared to traditional machine learning techniques. The digital twin (DT) concept enables seamless real-time data integration, physics-based models, and advanced data analytics to facilitate comprehensive CCGU lifecycle management. This novel approach provides operators with reliable, real-time insights to optimize plant performance, reduce maintenance costs, and improve overall sustainability. The generalization of the NODE model to an independent CCGU unit validated its applicability across diverse power plant configurations, highlighting the originality and practical value of the research. The key contribution of this work is the development of a digital twin-based predictive modeling framework centered on the innovative use of neural ordinary differential equations to enhance the operation and maintenance of critical energy infrastructure. This research advances the state-of-the-art in intelligent asset management for the built environment.

Towards proactive corrosion management: A predictive modeling approach in pipeline industrial applications

Corrosion is the most significant factor affecting pipeline integrity management. It may lead to facility damage, production interruption, environmental pollution and even explosion. Corrosion management developed in recent years is a critical means to ensure the safety and reliability of pipeline systems. However, developing proactive corrosion management is difficult due to limited availability of data on target pipelines and corrosion factors. The motivation of this study is to develop a proactive corrosion management approach based on a predictive framework. The proposed framework integrates multiple corrosion factors, diverse working conditions, multi-phase degradation, and probabilistic analysis to offer an efficient analysis tool. A numerical analysis of corrosion prediction is conducted using the corrosion data of the training pipeline and the target pipeline, and the corrosion rate, corrosion depth, and remaining useful life (RUL) of the target pipeline are evaluated. Based on the RUL information, the maintenance time of the pipeline is further analyzed to maximize the benefits of maintenance. This study offers a comprehensive approach to pipeline corrosion prediction, facilitating proactive maintenance strategies and improving asset management in the pipeline industry.

Individualized prediction of online shopping addiction from whole-brain functional connectivity

Online shopping addiction (OSA) is defined as a behavioral addiction where an individual exhibits an unhealthy and excessive attachment to shopping on the Internet. Since the OSA shown its adverse impacts on individuals' daily life and social functions, it is important to examine the neurobiological underpinnings of OSA that could be used in clinical practice to identify individuals with OSA. The present study addressed this question by employing a connectome-based prediction model approach to predict the OSA tendency of healthy subjects from whole-brain resting-state functional connectivity. The OSA connectome - a set of connections across multiple brain networks that contributed to predict individuals' OSA tendency was identified, including the functional connectivity between the frontal-parietal network (FPN) and cingulo-opercular network (CON) (i.e., positive network), as well as the functional connectivity within default mode network (DMN) and that between FPN and DMN (i.e., negative network). Key nodes that contributed to the prediction model included the middle frontal gyrus, inferior frontal gyrus, anterior cingulate cortex, and inferior temporal gyrus, which have been associated with impulsivity and emotional processing. Notably, this connectome has shown its specific role in predicting OSA by controlling for the influence of general Internet addiction. Moreover, the strength of the negative network mediated the relationship between OSA and impulsivity, highlighting that the negative network underlies the impulsivity characteristic of OSA. Together, these findings advanced our understanding of the neural correlates of OSA and provided a promising framework for diagnosing OSA.

Utilizing machine learning to classify persistent organic pollutants in the serum of pregnant women: a predictive modeling approach.

Polychlorinated biphenyls (PCBs), organochlorine pesticides (OCPs), polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDD/Fs), and per- and poly-fluoroalkyl substances (PFAS) are persistent organic pollutants (POPs) that remain detrimental to critical subpopulations, namely pregnant women. Required tests for biomonitoring are quite expensive. Moreover, statistical models aiming to discover the relationships between pollutants levels and human characteristics have their limitations. Therefore, the objective of this study is to use machine learning predictive models to further examine the pollutants' predictors, while comparing them. Levels of 33 congeners were measured in the serum of 269 pregnant women, from whom data was collected regarding sociodemographic, dietary, environmental, and anthropometric characteristics. Several machine learning algorithms were compared using "Python" for each pollutant: support vector machine (SVM), random forest, XGBoost, and neural networks. The aforementioned characteristics were included in the model as features. Prediction, accuracy, precision, recall, F1-score, area under the ROC curve (AUC), sensitivity, and specificity were retrieved to compare the models between them and among pollutants. The highest performing model for all pollutants was Random Forest. Results showed a moderate to acceptable performance and discriminative power among all POPs, with OCPs' model performing slightly better than all other models. Top related features for each model were also presented using SHAP analysis, detailing the predictors' negative or positive impact on the model. In conclusion, developing such a tool is of major importance in a context of limited financial and research resources. Nevertheless, machine learning models should always be interpreted with caution by exploring all evaluation metrics.

Corporate Governance and Innovation: A Predictive Modeling Approach with Machine Learning

Corporate Governance and Innovation: A Predictive Modeling Approach with Machine Learning

Impact of peripheral lymphocyte subsets on prognosis for patients after acute ischemic stroke: A potential disease prediction model approach.

To investigate the relationship between peripheral blood lymphocyte subsets and prognosis in patients with acute ischemic stroke (AIS). We enrolled 294 patients with AIS and collected peripheral blood samples for analysis of lymphocyte subsets. Prognosis was assessed at 3 months using the modified Rankin Scale (mRS). Association between lymphocyte count and poor outcomes (mRS score >2) was assessed using logistic regression. Individualized prediction models were developed to predict poor outcomes. Patients in the mRS score ≤2 group had higher T-cell percentage (odds ratio [OR] = 0.947; 95% confidence interval [CI]: 0.899-0.998; p = 0.040), CD3+ T-cell count (OR = 0.999; 95% CI: 0.998-1.000; p = 0.018), and CD4+ T-cell count (OR = 0.998; 95% CI: 0.997-1.000; p = 0.030) than those in the mRS score >2 group 1-3 days after stroke. The prediction model for poor prognosis based on the CD4+ T-cell count showed good discrimination (area under the curve of 0.844), calibration (p > 0.05), and clinical utility. Lower T cell percentage, CD3+, and CD4+ T-cell counts 1-3 days after stroke were independently associated with increased risk of poor prognosis. Individualized predictive model of poor prognosis based on CD4+ T-cell count have good accuracy and may predict disease prognosis.

Impact of silver nanoparticles active packaging on the behavior of Listeria monocytogenes and other microbial groups during ripening and storage of Canastra cheeses

This study aimed to assess the behavior of endogenous microbiota (total mesophilic and psychrotrophic bacteria) during the ripening period and of L. monocytogenes and lactic acid bacteria (LAB) during storage at 5, 10, and 20 °C of artisanal Canastra cheeses packed in a silver nanoparticles packaging system using a predictive modeling approach. The tested packaging system did not influence the initial values of pH (around 5.0) or aw (around 0.95) over the storage. Based on the mathematical parameters obtained by the Weibull predictive model, the active packaging system tested did not inactivate L. monocytogenes during the storage period of artisanal Canastra cheeses. Nonetheless, the findings indicate that the polyethylene packaging system containing silver nanoparticles (3000 ppm active based on silicon dioxide and silver) did not affect endogenous bacteria of Canastra cheese during ripening and storage, which opens the opportunity for the use of this material without compromising the role of these bacteria during cheese ripening, contributing to characteristics of the product's identity.

Comparison of predictive modeling approaches to estimate soil erosion under spatially heterogeneous field conditions

The accuracy of soil erosion models in agroecosystems with heterogeneous field conditions is challenging due to uncertainties from soil water fluxes and crop growth. In this study, we coupled two modeling methods (Freebairn and Rose) to represent soil erosion with a process-based crop and runoff models within the SIMPLACE framework. Their accuracy was compared to a statistical model developed using 16 erosion plots (each of 625 cm2) within the same field. Uncertainty analysis showed that runoff and slope angle were the most critical components for predicting sediment yield in both models, followed by soil erodibility in the Freebairn model and entrainment efficiency in the Rose model. However, due to plot size constraints, slope-length effects were not examined. The Freebairn model had a slightly higher accuracy (RMSE = 0.69 t ha−1 d−1) of sediment yield predictions than the Rose model (RMSE = 0.83 t ha−1 d−1). Both models are effective for predicting soil loss with appropriate parameter values.

Corporate governance and innovation: a predictive modeling approach using machine learning

The examination of the associations between internal corporate governance (CG) mechanisms and innovation faces challenges due to nonlinear patterns and complex interactions. Consequently, existing literature rarely reaches a consensus on the directions or strengths of these relationships. Furthermore, to investigate the CG–innovation association, prior research has predominantly relied on explanatory modeling, which involves applying statistical models to data to test correlational or causal hypotheses about theoretical constructs. These are the reasons why it remains unclear whether internal CG mechanisms, when considered collectively as an extensive array of interconnected variables, offer valuable insights for accurately predicting innovation. To address this gap, we analyze a dataset of research and development (R&D) projects from the Brazilian electricity sector by employing predictive modeling, which entails using statistical models or data mining algorithms to predict new observations, particularly using supervised machine learning (ML) methods. Our study demonstrates that a comprehensive set of variables representing internal CG mechanisms significantly enhances the predictive capabilities of ML algorithms for innovation. Furthermore, we illustrate how ML can illuminate nonlinear and non‐monotonic patterns, and interactions among variables, in the CG–innovation relationship. Our contribution to the literature encompasses three key aspects: introducing a predictive modeling approach to the discourse on the role of CG in innovation attainment through R&D endeavors, which can complement and enrich existing explanatory research; investigating non‐linear and non‐monotonic relationships, as well as interactions, in innovation prediction; and affirming the emerging body of literature that recognizes supervised ML as a valuable tool accessible to management researchers.

A predictive modeling approach for Taiwanese diagnosis-related groups medical costs: A focus on laparoscopic appendectomy

Abstract Background: In the Taiwanese diagnosis-related groups (Tw-DRGs) system, effective management of medical resources is vital to ensure the sustainability of hospital operations. Objectives: The aim is to create a predictive model to estimate the medical costs linked to a specific Tw-DRGs item, utilizing laparoscopic appendectomy without complications or comorbidities (DRG16701) as a representative case. Materials and Methods: We employed a dataset comprising 248 surgical cases performed at a regional teaching hospital between January 2017 and December 2019. These cases were classified based on the difference between the Tw-DRGs payment standard and actual medical costs. Two experiments were conducted: one without feature selection and one with feature selection. We utilized random forest (RF) and principal component analysis in each experiment. Each experiment applied the following four predictive models: decision tree, RF, logistic regression, and backpropagation neural network. The models were evaluated by measuring the accuracy, F1-score, and area under the receiver operating characteristic curve (AUROC). Results: The RF model demonstrated satisfactory performance, achieving an accuracy and F1-score of 0.920 on the testing set, with an AUROC ranging from 0.92 to 0.95. Feature selection methods enhanced model performance, particularly for the RF model. Critical features included premeal glucose levels, age, body mass index, weight, potassium, activated partial thromboplastin time, C-reactive protein level, and height. Conclusion: On average, each laparoscopic appendectomy case resulted in a deficit of NTD 3173.6. Cost prediction proved feasible using routine blood test data obtained upon admission or before surgery. The RF model and feature selection emerged as the most suitable predictive model for this specific purpose.

Data Analysis and Prediction of Of Various app

Abstract – This study focuses on the exploratory data analysis (EDA) and prediction of ratings for various apps available on the Google Play Store. As the number of mobile applications continues to grow exponentially, understanding the factors that influence app ratings can provide valuable insights for developers and stakeholders. This research involves a detailed examination of a dataset containing information about numerous apps, including their categories, sizes, user ratings, number of installs, and more. The EDA process involves summarizing the main characteristics of the data, identifying patterns, and uncovering anomalies. Key techniques such as descriptive statistics, visualization, and correlation analysis are utilized to explore relationships between app ratings and various attributes. Insights gained from EDA include the distribution of app ratings, the influence of app category and size on ratings, and trends in user feedback over time. Following EDA, a predictive modeling approach is employed to forecast app ratings based on identified influential features. Various machine learning algorithms, such as linear regression, decision trees, and random forests, are applied and compared to determine the most effective model. Performance metrics like Mean Squared Error (MSE) and R-squared are used to evaluate and validate the models. The results of this study highlight significant predictors of app ratings, offering practical recommendations for app developers to enhance user satisfaction. Additionally, the predictive models provide a framework for anticipating app success in the market, enabling more informed decision-making. Overall, this research contributes to a deeper understanding of the app ecosystem on the Google Play Store and demonstrates the value of data-driven approaches in optimizing app development and marketing strategies. IndexTerms – Car price prediction, Machine learning, Regression analysis, Automobile market, Predictive modeling, Vehicle valuation, Data analysis, Price determinants, Automotive industry, Market efficiency

Exploratory Data Analysis and Prediction of Rating Of Various app on Google Play Store

Abstract – This study focuses on the exploratory data analysis (EDA) and prediction of ratings for various apps available on the Google Play Store. As the number of mobile applications continues to grow exponentially, understanding the factors that influence app ratings can provide valuable insights for developers and stakeholders. This research involves a detailed examination of a dataset containing information about numerous apps, including their categories, sizes, user ratings, number of installs, and more. The EDA process involves summarizing the main characteristics of the data, identifying patterns, and uncovering anomalies. Key techniques such as descriptive statistics, visualization, and correlation analysis are utilized to explore relationships between app ratings and various attributes. Insights gained from EDA include the distribution of app ratings, the influence of app category and size on ratings, and trends in user feedback over time. Following EDA, a predictive modeling approach is employed to forecast app ratings based on identified influential features. Various machine learning algorithms, such as linear regression, decision trees, and random forests, are applied and compared to determine the most effective model. Performance metrics like Mean Squared Error (MSE) and R-squared are used to evaluate and validate the models. The results of this study highlight significant predictors of app ratings, offering practical recommendations for app developers to enhance user satisfaction. Additionally, the predictive models provide a framework for anticipating app success in the market, enabling more informed decision-making. Overall, this research contributes to a deeper understanding of the app ecosystem on the Google Play Store and demonstrates the value of data-driven approaches in optimizing app development and marketing strategies. IndexTerms – Car price prediction, Machine learning, Regression analysis, Automobile market, Predictive modeling, Vehicle valuation, Data analysis, Price determinants, Automotive industry, Market efficiency

Repurposing plastic waste: Experimental study and predictive analysis using machine learning in bricks

The construction industry significantly contributes to environmental degradation and resource depletion. This paper investigates the use of waste plastic from post-consumer and post-industrial sources as a sustainable alternative in brick manufacturing. By substituting plastic waste for traditional materials like clay or cement, the ecological footprint of brick production is reduced, and plastic pollution is mitigated by diverting waste from landfills and oceans. This study employs dual approach of experimentation as well as machine learning for assessing the strength behaviour of bricks enhanced with plastic, emphasising compressive strength, durability, thermal insulation, and moisture resistance. Utilization of Chlorinated Polyvinyl Chloride (CPVC) for Waste plastic bricks (WPB) was done by replacing cement and sand in different proportions i.e. 10 %, 20 %, 30 %, 40 %, and 50 % of fly ash bricks The best performance was observed at 30 % replacement of sand with waste CPVC. A Deep Neural Network (DNN) model also predicted compressive strength and water absorption, identifying key material components influencing brick properties through SHAP value analysis. This combined experimental and predictive modelling approach demonstrates the potential of waste plastic in creating lightweight, sustainable building materials for structural applications, contributing to sustainable construction practices.

Experimental and numerical characterization of the under-vane pressure in balanced vane pumps

Vane motion is a critical aspect of balanced vane pumps as related to efficiency, durability, and noise-vibration-harshness performance. Assessing the behavior of the under-vane pockets in those machines is relevant since the related pressure forces are recognized as important contributors to vane dynamics. Within this context, this work addresses a numerical-experimental approach for the under-vane pressure characterization of balanced vane pumps. The methodology is numerically based on transient three-dimensional computational fluid dynamics models that simulate the fluid domains of the entire machine, under-vane pockets included. The experimental approach for under-vane monitoring is based on a remote pressure transducer linked to an under-vane pocket through an auxiliary conduit. The methodology is applied to an automotive machine, and a virtual-experimental campaign concerning dedicated samples is presented for pump characterization and model validation. The numerical results reveal a notable agreement with experiments, showing the suitability of the proposed modeling approach for under-vane pressure prediction. Special attention is devoted to the limitations of the described experimental method due to the frequency response behavior of the auxiliary conduit as underlined by the numerical results. The latter topic is further investigated through acoustic Finite Element Method and Helmholtz resonator lumped parameter models.

Advanced Prognostic Models for Bearing Health: A Comparative Analysis of BiLSTM and ANFIS

Bearings play a critical role in the operation of rotary machines, serving as essential components. Their failure often leads to unexpected shutdowns, posing a significant risk to the entire system. To mitigate these risks, it is imperative to implement proactive maintenance measures and strategic planning to prevent system breakdowns. This article introduces a comparative analysis between two predictive modelling approaches: Bidirectional Long Short-Term Memory (Bi-LSTM) and Adaptive Neuro Fuzzy Inference System (ANFIS) networks, aiming to enhance bearing prognostics. The proposed methodology involves a two-step process. Firstly, data undergoes pre-processing through wavelet packet decomposition (WPD). Subsequently, a degradation model is employed for predicting the remaining useful life (RUL). To validate the accuracy of the proposed approach, extensive testing is conducted using a bearing's life dataset obtained from a run-to-failure experiment. The results demonstrate that the ANFIS model exhibits remarkable capabilities in learning and accurately estimating the system's RUL, achieving this with minimal computation time compared to alternative methods, thus presenting a more efficient and precise solution.

The interplay between white adipose tissue, adipokines, and structural gray matter changes

AbstractThe growing global obesity issue emphasizes the importance of understanding its health implications. Previous research has identified consistent alterations in gray matter (GM) volume in connection with obesity. Given the various implications of distinct fat compartments and the potential role of adipose tissue‐derived adipokines in brain health, a more detailed investigation of adiposity is required. This study investigates a sample of 65 males with varying body mass indices to explore the relationship between various fat compartments, adipokine levels, and volumetric GM variations, aiming to provide a deeper understanding of the interplay between adiposity, brain structure, and metabolic signals. Whole‐body magnetic resonance imaging (MRI) was used to assess total, visceral, and subcutaneous adipose tissue, while MR spectroscopy was performed to capture liver fat content. For the assessment of adipokine levels leptin and adiponectin concentrations were measured, and structural brain images underwent cortical and subcortical segmentation for GM volume and thickness. A predictive modeling approach with leave‐one‐out cross‐validation was used to predict body composition metrics and adipokine levels based on structural GM data. Our investigation revealed diminished GM volume and thickness correlated with elevated leptin levels in areas crucial for appetite regulation, decision‐making, and cognitive control, including the anterior insula, orbitofrontal cortex, and anterior cingulate cortex. These findings suggest a potential adverse impact of heightened leptin concentrations on brain health and eating habits. Contrary to expectations, our investigation found no significant relationship between GM volume and any of the measured fat compartments. This result prompts the need for further research to elucidate the relationship between obesity, adipokines, and brain structure.