Decision Support Research Articles

A decision support system to increase the compliance of diagnostic imaging examinations with imaging guidelines: focused on cerebrovascular diseases

Abstract Objectives Diagnostic imaging decision support (DI-DS) system has emerged as an innovative evidence-based solution to decrease inappropriate diagnostic imaging. The aim of the present study was to design and evaluate a DI-DS system for cerebrovascular diseases. Methods The present study was an applied piece of research. First, the conceptual model of the DI-DS system was designed based on its functional and non-functional requirements. Afterwards, to create the system’s knowledge base, cerebrovascular diseases diagnostic imaging algorithms were extracted from the American College of Radiology Appropriateness Criteria (ACR-AC). Subsequently, the system was developed based on the obtained conceptual model and the extracted algorithms. The software was programmed by means of the C#. After debugging the system, it was evaluated regarding its performance and also the users’ satisfaction with it. Results Assessing the users’ satisfaction with the system demonstrated that all the evaluation criteria met the acceptable threshold (85 %). The retrospective evaluation of the system’s performance indicated that from among 76 imaging examinations, which had previously been performed for 30 patients, 12 (15.78 %) were deemed inappropriate. And, the system accurately identified all the inappropriate physicians’ decisions. The concurrent evaluation of the system’s performance indicated that the system’s recommendations helped the physicians remove 100 % (4 out of 4) of the inappropriate and 40 % (2 out of 5) of the inconclusive imaging examinations from their initial choices. Conclusions A DI-DS system could increase the compliance of the physicians’ decisions with diagnostic imaging guidelines, and also improve treatment outcomes through correct diagnosis and providing timely care.

From analytics to empathy and creativity: Charting the AI revolution in marketing practice and education

The rapid advancement of artificial intelligence (AI) increasingly demands an understanding of its impact on marketing practice and education. Our hybrid literature review synthesizes 312 peer-reviewed articles on AI in marketing and consumer behavior, using scientometrics and the TCCM (Theory, Context, Characteristics, Methodology) framework. We identify five research areas: human–AI interaction in services, natural language processing (NLP) and computer vision for consumer insights, AI for e-commerce and decision support, marketing automation and creativity, and AI ethics. AI’s evolution is marked by a transition from analytical to empathetic and intuitive technologies like affective computing and generative AI. We highlight the changing dynamics between humans and AI, AI integration in marketing practices and education, and the transformed AI-enhanced marketing workplace. We underscore the significance of ethical considerations, the well-being of users, and the integration of generative AI tools. This review provides a comprehensive guide for forthcoming research, practical applications, and educational advancements in AI-enhanced marketing.

Novel statistically equivalent signature-based hybrid feature selection and ensemble deep learning LSTM and GRU for chronic kidney disease classification

Chronic kidney disease (CKD) involves numerous variables, but only a few significantly impact the classification task. The statistically equivalent signature (SES) method, inspired by constraint-based learning of Bayesian networks, is employed to identify essential features in CKD. Unlike conventional feature selection methods, which typically focus on a single set of features with the highest predictive potential, the SES method can identify multiple predictive feature subsets with similar performance. However, most feature selection (FS) classifiers perform suboptimally with strongly correlated data. The FS approach faces challenges in identifying crucial features and selecting the most effective classifier, particularly in high-dimensional data. This study proposes using the Least Absolute Shrinkage and Selection Operator (LASSO) in conjunction with the SES method for feature selection in CKD identification. Following this, an ensemble deep-learning model combining long short-term memory (LSTM) and gated recurrent unit (GRU) networks is proposed for CKD classification. The features selected by the hybrid feature selection method are fed into the ensemble deep-learning model. The model’s performance is evaluated using accuracy, precision, recall, and F1 score metrics. The experimental results are compared with individual classifiers, including decision tree (DT), Random Forest (RF), logistic regression (LR), and support vector machine (SVM). The findings indicate a 2% improvement in classification accuracy when using the proposed hybrid feature selection method combined with the LSTM and GRU ensemble deep-learning model. Further analysis reveals that certain features, such as HEMO, POT, bacteria, and coronary artery disease, contribute minimally to the classification task. Future research could explore additional feature selection methods, including dynamic feature selection that adapts to evolving datasets and incorporates clinical knowledge to enhance CKD classification accuracy further.

Collaborative Scheduling of Mass Rescue Operations at Sea

The continued growth of the global cruise market has posed a major challenge to maritime search and rescue, as available rescue resources are limited compared to the scale of passenger ship disasters. To bridge the gap between the number of passengers in distress and the availability of rescue resources, this paper develops a two-stage Mass Rescue Operation (MRO) decision support model (MRO model) to fully utilize the available multiple rescue resources. Based on the combinatorial optimization theory, in the MRO model we consider the rescue capacity of multiple rescue resources and the synergy between them, the accident types and the marine environment conditions to optimize two objectives (rescue time and number of rescue resources dispatched) in two stages. In the first stage, the objective is to minimize the rescue time by Classical Selection Sort Algorithm. In the second stage, the rescue time and the number of rescue resources are simultaneously optimized by Simulated Annealing Arithmetic (SAA) integrated with Genetic Algorithm (GA). Furthermore, considering the actual role of helicopters in MROs, the MRO model is enhanced to schedule helicopters mandatorily or non-mandatorily. Finally, the MRO model was verified by simulating accidents in the Taiwan Strait. The simulation results show that compared with the first stage, the rescue time in the second stage model is saved by up to 16.18% and the number of rescue resources is reduced by up to 45.16%.

Evaluating small-scale eucalyptus growers’ experience of the EVO mobile application: a comparative analysis of quantitative and qualitative assessment techniques

ABSTRACT Mobile applications (apps) have become widely available as decision support tools in agriculture and forestry worldwide. Owing to the above-mentioned development trend, identifying testing methods that can be easily implemented for evaluating mobile app development targeting rural livelihood-related topics is urgently required. This study examined user experiences with a previously launched and published mobile app (EVO) that small-scale eucalyptus growers can utilize to estimate their gross income and net profit from timber sales. The survey incorporated three user experience assessment methods. The study involved two rounds of data collection regarding the mobile app: the first round concerned version 1.0, whereas the second round referred to version 2.0. The aim was to identify the user experience of the EVO app for guiding further software development to improve its suitability among potential user groups, as well as to estimate the feasibility and usability of survey methods for evaluating mobile apps among smallholders and related stakeholders. The results indicated a remarkable improvement in user satisfaction between the two rounds. Using a combination of system usability scale, user acceptance testing, and quality questions is recommended for collecting valuable and useful feedback for the development of mobile apps.

Estimation of Understory Fine Dead Fuel Moisture Content in Subtropical Forests of Southern China Based on Landsat Images

The understory fine dead fuel moisture content (DFMC) is an important reference indicator for regional forest fire warnings and risk assessments, and determining it on a large scale is a critical goal. It is difficult to estimate understory fine DFMC directly from satellite images due to canopy shading. To address this issue, we used canopy meteorology estimated by Landsat images in combination with explanatory variables to construct random forest models of in-forest meteorology, and then construct random forest models by combining the meteorological factors and explanatory variables with understory fine DFMC obtained from the monitoring device to (1) investigate the feasibility of Landsat images for estimating in-forest meteorology; (2) explore the feasibility of canopy or in-forest meteorology and explanatory variables for estimating understory fine DFMC; and (3) compare the effects of each factor on model accuracy and its effect on understory fine DFMC. The results showed that random forest models improved in-forest meteorology estimation, enhancing in-forest relative humidity, vapor pressure deficit, and temperature by 50%, 34%, and 2.2%, respectively, after adding a topography factor. For estimating understory fine DFMC, models using vapor pressure deficit improved fit by 10.2% over those using relative humidity. Using in-forest meteorology improved fits by 36.2% compared to canopy meteorology. Including topographic factors improved the average fit of understory fine DFMC models by 123.1%. The most accurate model utilized in-forest vapor pressure deficit, temperature, topographic factors, vegetation index, precipitation data, and seasonal factors. Correlations indicated that slope, in-forest vapor pressure deficit, and slope direction were most closely related to understory fine DFMC. The regional understory fine-grained DFMC distribution mapped according to our method can provide important decision support for forest fire risk early warning and fire management.

Hybrid Oversampling and Undersampling Method (HOUM) via Safe-Level SMOTE and Support Vector Machine

The improvements in collecting and processing data using machine learning algorithms have increased the interest in data mining. This trend has led to the development of real-life decision support systems (DSSs) in diverse areas such as biomedical informatics, fraud detection, natural language processing, face recognition, autonomous vehicles, image processing, and each part of the real production environment. The imbalanced datasets in some of these studies, which result in low performance measures, have highlighted the need for additional efforts to address this issue. The proposed method (HOUM) is used to address the issue of imbalanced datasets for classification problems in this study. The aim of the model is to prevent the overfitting problem caused by oversampling and valuable data loss caused by undersampling in imbalanced data and obtain successful classification results. The HOUM is a hybrid approach that tackles imbalanced class distribution challenges, refines datasets, and improves model robustness. In the first step, majority-class data points that are distant from the decision boundary obtained via SVM are reduced. If the data are not balanced, SLS is employed to augment the minority-class data. This loop continues until the dataset becomes balanced. The main contribution of the proposed method is reproducing informative minority data using SLS and diminishing non-informative majority data using the SVM before applying classification techniques. Firstly, the efficiency of the proposed method, the HOUM, is verified by comparison with the SMOTE, SMOTEENN, and SMOTETomek techniques using eight datasets. Then, the results of the W-SIMO and RusAda algorithms, which were developed for imbalanced datasets, are compared with those of the HOUM. The strength of the HOUM is revealed through this comparison. The proposed HOUM algorithm utilizes a real dataset obtained from a project endorsed by The Scientific and Technical Research Council of Turkey. The collected data include quality control and processing parameters of yarn data. The aim of this project is to prevent yarn breakage errors during the weaving process on looms. This study introduces a decision support system (DSS) designed to prevent yarn breakage during fabric weaving. The high performance of the algorithm may encourage producers to manage yarn flow and enhance the HOUM’s efficiency as a DSS.

MIS Solutions During Natural Disaster Management: A Review On Responsiveness, Coordination, And Resource Allocation

This systematic review explores the impact of Management Information Systems (MIS) on enhancing disaster management by optimizing decision-making, resource allocation, and inter-agency communication. As the frequency and severity of natural disasters increase globally, there is a pressing need for more efficient disaster response mechanisms. This study systematically reviewed a total of 160 peer-reviewed articles published between 2010 and 2024, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure a comprehensive and transparent review process. The findings reveal that MIS technologies, such as decision support systems (DSS), geographic information systems (GIS), and predictive analytics, play a crucial role in improving the speed and accuracy of emergency responses. Specifically, DSS and predictive models were found to enhance situational awareness and optimize resource deployment, reducing response times by up to 30%. Additionally, GIS tools significantly improved spatial data analysis, enabling better-targeted relief efforts. However, challenges related to data interoperability, cybersecurity, and the integration of advanced technologies remain significant barriers to fully leveraging MIS in disaster management. Addressing these challenges through investments in infrastructure, standardized protocols, and specialized training will be essential for maximizing the potential of MIS in future disaster response efforts. The review concludes that the strategic use of MIS is vital for building more resilient and responsive disaster management systems, ultimately reducing the socio-economic impacts of emergencies.

Study protocol of the ALMA-CKD trial; an electronic triggering decision-support system to improve the detection, recognition, and management of patients with chronic kidney disease in primary care

Abstract Background Chronic kidney disease (CKD) is a global health problem affected by under-recognition and under-treatment in primary care settings. Electronic clinical decision support (CDS) triggering systems have the potential to improve detection and management of people with CKD by assisting clinicians in adhering to guideline recommendations. We aimed to test whether an electronic CDS triggering system would improve the detection, recognition, and management of patients with CKD in primary care. Method/Design This is a pragmatic cluster-randomized controlled trial where 66 primary healthcare centers from the Stockholm Region, Sweden were randomized 1:1 to receive either a new expanded CDS-triggering system offering kidney-specific advice or to continue with their current CDS-triggering system. The expanded CDS system reminds and provides practical facilitators of the processes of CKD screening, recognition with a diagnosis, management and referral to specialist care. The trial duration is 24 months and it is embedded into the Stockholm CREAtinine measurements (SCREAM) project, a repository of healthcare data from the region, which minimizes disturbances with healthcare praxis due to the trial and makes it fully pragmatic. The primary outcomes are the number of eligible patients screened for creatinine and albuminuria once annually and the re-testing of these labs within 6 months in patients with abnormal eGFR or albuminuria. Secondary outcomes are the proportions of issued clinical diagnoses among those fulfilling criteria, proportions of patients with significant albuminuria receiving prescribed nephroprotective medications, proportions of accepted referrals to nephrologist care among those fulfilling criteria and proportion of referrals for ultrasound of the kidneys. Discussion Prior pragmatic trials of CDS-systems in CKD has shown an improvement in quality indicators primarily in patients already diagnosed with CKD. This study expands this evidence by focusing on the process of screening, identification, monitoring and diagnostic work-up. Conclusion This pragmatic trial will assess the value of CDS for improved adherence to CKD guidelines in primary care. Clinicaltrials.gov registration: NCT06386172, submitted 2024-04-23.

Protocol for a mixed methods process evaluation for a randomised controlled trial to improve shared decision-making about, and uptake of, osteoporosis medicines: the iFraP study

Background High quality shared decision-making (SDM) conversations involve people with or at risk of osteoporosis and clinicians working together to decide, where appropriate, which evidence-based medicines best fit the person’s life, beliefs, and values. The improving uptake of Fracture Prevention drug treatments (iFraP) intervention comprises a computerised Decision Support Tool (DST), clinician training package and information resources, designed for use in UK Fracture Liaison Service (FLS) consultations. The iFraP intervention will be tested in a pragmatic, parallel-group, individual randomised controlled trial in patients referred to four FLSs in England. This mixed methods process evaluation aims to assess which components of iFraP were delivered and how (fidelity), whether iFraP results in a change in osteoporosis drug treatment initiation rates and how, and how context affects implementation of iFraP and outcomes. Methods We will collect quantitative data using (1) Case Report Forms completed by FLS clinicians; (2) self-reported questionnaires completed by patient participants; and (3) DST analytic data. We will collect qualitative data using (1) semi-structured interviews with patients who receive the iFraP intervention in their FLS appointment, FLS clinicians delivering iFraP appointments, and primary care clinicians that have consulted with a patient following their iFraP FLS appointment; and (2) FLS consultation recordings. A triangulation protocol will be used to integrate the quantitative and qualitative findings to generate novel insights about the intervention under evaluation. Discussion The process evaluation, alongside the trial, will help to understand what elements of the iFraP intervention were delivered and how, the mechanisms of impact and how context affected implementation and outcomes, and intervention acceptability. Mixed methods interpretation will lead to further insights about the implementation of SDM and DSTs in-practice. Trial registration ISRCTN: 10606407, 21/11/2022 https://doi.org/10.1186/ISRCTN10606407

Design of judicial public opinion supervision and intelligent decision-making model based on Bi-LSTM

Fuzzy preference modeling in intelligent decision support systems aims to improve the efficiency and accuracy of decision-making processes by incorporating fuzzy logic and preference modeling techniques. While network public opinion (NPO) has the potential to drive judicial reform and progress, it also poses challenges to the independence of the judiciary due to the negative impact of malicious public opinion. To tackle this issue within the context of intelligent decision support systems, this study provides an insightful overview of current NPO monitoring technologies. Recognizing the complexities associated with handling large-scale NPO data and mitigating significant interference, a novel judicial domain NPO monitoring model is proposed, which centers around semantic feature analysis. This model takes into account time series characteristics, binary semantic fitting, and public sentiment intensity. Notably, it leverages a bidirectional long short-term memory (Bi-LSTM) network (S-Bi-LSTM) to construct a judicial domain semantic similarity calculation model. The semantic similarity values between sentences are obtained through the utilization of a fully connected layer. Empirical evaluations demonstrate the remarkable performance of the proposed model, achieving an accuracy rate of 85.9% and an F1 value of 87.1 on the test set, surpassing existing sentence semantic similarity models. Ultimately, the proposed model significantly enhances the monitoring capabilities of judicial authorities over NPO, thereby alleviating the burden on public relations faced by judicial institutions and fostering a more equitable execution of judicial power.

Clinical Acceptability of a Quality Improvement Program for Reducing Cardiovascular Disease Risk in People With Chronic Kidney Disease in Australian General Practice: Qualitative Study

Abstract Background Future Health Today (FHT) is a technology program that integrates with general practice clinical software to provide point of care (PoC) clinical decision support and a quality improvement dashboard. This qualitative study looks at the use of FHT in the context of cardiovascular disease risk in chronic kidney disease (CKD). Objective This study aims to explore factors influencing clinical implementation of the FHT module focusing on cardiovascular risk in CKD, from the perspectives of participating general practitioner staff. Methods Practices in Victoria were recruited to participate in a pragmatic cluster randomized controlled trial using FHT, of which 19 practices were randomly assigned to use FHT’s cardiovascular risk in CKD program. A total of 13 semistructured interviews were undertaken with a nominated general practitioner (n=7) or practice nurse (n=6) from 10 participating practices. Interview questions focused on the clinical usefulness of the tool and its place in clinical workflows. Qualitative data were coded by 2 researchers and analyzed using framework analysis and Clinical Performance Feedback Intervention Theory. Results All 13 interviewees had used the FHT PoC tool, and feedback was largely positive. Overall, clinicians described engaging with the tool as a “prompt” or “reminder” system. Themes reflected that the tool’s goals and clinical content were aligned with clinician’s existing priorities and knowledge, and the tool’s design facilitated easy integration into existing workflows. The main barrier to implementation identified by 2 clinicians was notification fatigue. A total of 7 interviewees had used the FHT dashboard tool. The main barriers to use were its limited integration into clinical workflows, such that some participants did not know of its existence; clinicians’ competing clinical priorities; and limited time to learn and use the tool. Conclusions This study identified many facilitators for the successful use of the FHT PoC program, in the context of cardiovascular risk in CKD, and barriers to the use of the dashboard program. This work will be used to inform the wider implementation of FHT, as well as the development of future modules of FHT for other risk or disease states.

An Electronic Medical Record–Based Prognostic Model for Inpatient Falls: Development and Internal-External Cross-Validation

Background Effective fall prevention interventions in hospitals require appropriate allocation of resources early in admission. To address this, fall risk prediction tools and models have been developed with the aim to provide fall prevention strategies to patients at high risk. However, fall risk assessment tools have typically been inaccurate for prediction, ineffective in prevention, and time-consuming to complete. Accurate, dynamic, individualized estimates of fall risk for admitted patients using routinely recorded data may assist in prioritizing fall prevention efforts. Objective The objective of this study was to develop and validate an accurate and dynamic prognostic model for inpatient falls among a cohort of patients using routinely recorded electronic medical record data. Methods We used routinely recorded data from 5 Australian hospitals to develop and internally-externally validate a prediction model for inpatient falls using a Cox proportional hazards model with time-varying covariates. The study cohort included patients admitted during 2018-2021 to any ward, with no age restriction. Predictors used in the model included admission-related administrative data, length of stay, and number of previous falls during the admission (updated every 12 hours up to 14 days after admission). Model calibration was assessed using Poisson regression and discrimination using the area under the time-dependent receiver operating characteristic curve. Results There were 1,107,556 inpatient admissions, 6004 falls, and 5341 unique fallers. The area under the time-dependent receiver operating characteristic curve was 0.899 (95% CI 0.88-0.91) at 24 hours after admission and declined throughout admission (eg, 0.765, 95% CI 0.75-0.78 on the seventh day after admission). Site-dependent overestimation and underestimation of risk was observed on the calibration plots. Conclusions Using a large dataset from multiple hospitals and robust methods to model development and validation, we developed a prognostic model for inpatient falls. It had high discrimination, suggesting the model has the potential for operationalization in clinical decision support for prioritizing inpatients for fall prevention. Performance was site dependent, and model recalibration may lead to improved performance.

How to make land use policy decisions: Integrating science and economics to deliver connected climate, biodiversity, and food objectives

Land use change is crucial to addressing the existential threats of climate change and biodiversity loss while enhancing food security [M. Zurek et al. , Science 376 , 1416–1421 (2022)]. The interconnected and spatially varying nature of the impacts of land use change means that these challenges must be addressed simultaneously [H.-O. Pörtner et al. , Science 380 , eabl4881 (2023)]. However, governments commonly focus on single issues, incentivizing land use change via “Flat-Rate” subsidies offering constant per hectare payments, uptake of which is determined by the economic circumstances of landowners rather than the integrated environmental outcomes that will be delivered [G. Q. Bull et al. , Forest Policy Econ. 9 , 13–31 (2006)]. Here, we compare Flat-Rate subsidies to two alternatives: “Land Use Scenario” allocation of subsidies through consultation across stakeholders and interested parties; and a “Natural Capital” approach which targets subsidies according to expected ecosystem service response. This comparison is achieved by developing a comprehensive decision support system, integrating new and existing natural, physical, and economic science models to quantify environmental, agricultural, and economic outcomes. Applying this system to the United Kingdom’s net zero commitment to increase carbon storage via afforestation, we show that the three approaches result in significantly different outcomes in terms of where planting occurs, their environmental consequences, and economic costs and benefits. The Flat-Rate approach actually increases net carbon emissions while Land Use Scenario allocation yields poor economic outcomes. The Natural Capital targeted approach outperforms both alternatives, providing the highest possible social values while satisfying net zero commitments.

Comprehensive Symptom Prediction in Inpatients With Acute Psychiatric Disorders Using Wearable-Based Deep Learning Models: Development and Validation Study

Background Assessing the complex and multifaceted symptoms of patients with acute psychiatric disorders proves to be significantly challenging for clinicians. Moreover, the staff in acute psychiatric wards face high work intensity and risk of burnout, yet research on the introduction of digital technologies in this field remains limited. The combination of continuous and objective wearable sensor data acquired from patients with deep learning techniques holds the potential to overcome the limitations of traditional psychiatric assessments and support clinical decision-making. Objective This study aimed to develop and validate wearable-based deep learning models to comprehensively predict patient symptoms across various acute psychiatric wards in South Korea. Methods Participants diagnosed with schizophrenia and mood disorders were recruited from 4 wards across 3 hospitals and prospectively observed using wrist-worn wearable devices during their admission period. Trained raters conducted periodic clinical assessments using the Brief Psychiatric Rating Scale, Hamilton Anxiety Rating Scale, Montgomery-Asberg Depression Rating Scale, and Young Mania Rating Scale. Wearable devices collected patients’ heart rate, accelerometer, and location data. Deep learning models were developed to predict psychiatric symptoms using 2 distinct approaches: single symptoms individually (Single) and multiple symptoms simultaneously via multitask learning (Multi). These models further addressed 2 problems: within-subject relative changes (Deterioration) and between-subject absolute severity (Score). Four configurations were consequently developed for each scale: Single-Deterioration, Single-Score, Multi-Deterioration, and Multi-Score. Data of participants recruited before May 1, 2024, underwent cross-validation, and the resulting fine-tuned models were then externally validated using data from the remaining participants. Results Of the 244 enrolled participants, 191 (78.3%; 3954 person-days) were included in the final analysis after applying the exclusion criteria. The demographic and clinical characteristics of participants, as well as the distribution of sensor data, showed considerable variations across wards and hospitals. Data of 139 participants were used for cross-validation, while data of 52 participants were used for external validation. The Single-Deterioration and Multi-Deterioration models achieved similar overall accuracy values of 0.75 in cross-validation and 0.73 in external validation. The Single-Score and Multi-Score models attained overall R² values of 0.78 and 0.83 in cross-validation and 0.66 and 0.74 in external validation, respectively, with the Multi-Score model demonstrating superior performance. Conclusions Deep learning models based on wearable sensor data effectively classified symptom deterioration and predicted symptom severity in participants in acute psychiatric wards. Despite lower computational costs, Multi models demonstrated equivalent or superior performance than Single models, suggesting that multitask learning is a promising approach for comprehensive symptom prediction. However, significant variations were observed across wards, which presents a key challenge for developing clinical decision support systems in acute psychiatric wards. Future studies may benefit from recurring local validation or federated learning to address generalizability issues.

Investment Potential in Latin America under the Belt and Road Initiative

This study analyzes the current state, influencing factors, and potential of China's investments in Latin America under the Belt and Road Initiative (BRI). Through a systematic review of China's investments in sectors such as mining, energy, agriculture, infrastructure, and manufacturing in Latin America, the research explores the impact of geographical distance, institutional quality, infrastructure level, and cultural differences on Chinese enterprises' investments. The findings suggest that despite the vast geographical distance between China and Latin America, economic ties are strengthening through overseas infrastructure investments and financial cooperation. The study proposes policy recommendations to further promote bilateral investments through institutional reforms and localized strategies. The research provides theoretical support for Chinese enterprises' investment decisions in Latin America and serves as a valuable reference for policymakers.

Abstract 4140120: Artificial intelligence-driven morbidity prediction in acute kidney injury after acute type A aortic dissection surgery

Background: Acute kidney injury (AKI) often complicates acute type A aortic dissection (ATAAD), with elevated comorbidity rates and a significant tie to in-hospital mortality. Identifying risk factors early can mitigate AKI severity. Research Questions: This research endeavors to develop and corroborate predictive models leveraging Machine Learning (ML) techniques from Artificial Intelligence to forecast AKI occurrences in ATAAD-afflicted individuals. Methods: The study employed various machine learning (ML) algorithms including Gradient Boosting Machine (GBM), LightGBM, Random Forest (RF), K-Nearest Neighbors (KNN), Multi-Layer Perceptron Neural Network (MLP-NN), Naive Bayes (NB), Logistic Regression (LR), and ensemble methods (combining LR&LightGBM), employing tenfold cross-validation. Model performance was evaluated using SHapley Additive exPlanations (SHAP). A web-based tool for predicting AKI incidence was developed using Streamlit, based on the most effective model. The analysis involved 1350 ATAAD patients, among whom 586 (43.4%) developed post-operative AKI. Patients were divided into two cohorts: 85% for training and 15% for testing, with 126 features included in the predictive model. Results: Incorporating top 10 features, LightGBM (AUROC=0.886, 95% CI 0.841-0.930) excelled in predictive accuracy, calibration, and clinical utility, identifying key factors such as ventilation time in ICU, hourly urine output post-surgery, diuretic use, Scr, heart rate, urea, administration of recombinant human brain natriuretic peptide and ebrantil, MCHC, and blood glucose as associated with ATAAD-AKI. Conclusion(s): These ML models are robust tools for predicting AKI in ATAAD patients, with LightGBM's superior predictive ability standing out. They offer valuable support for clinical decision-making in ATAAD management, helping optimize postoperative strategies to minimize AKI occurrence after surgery.

Machine learning techniques for the smart faults detection and diagnosis of centrifugal compressor

In this work, we have conducted a comparative study among several machine learning techniques with the aim of selecting the best one for classifying faults affecting the compressor system to enable smart monitoring. This study encompasses various machine learning techniques, including Support Vector Machine, k-nearest neighbor, Decision Tree, Naive Bayes, AdaBoost, and Bag ensembles. To determine the optimal classification technique, we applied three distinct criteria: the confusion matrix, error histogram, and mean square error through cross-validation. Based on these criteria, the results indicate a tie for the top position between two classification models: Decision Tree and Bag ensemble. To solidify our choice of a single model, we employed the new AutoML technique to automatically identify the most suitable machine learning classification model for our case study. We evaluated this approach using process data obtained from an operational industrial centrifugal compressor. Consequently, the results presented in this work affirm that Decision Tree is the superior technique for classifying faults in the 3MCL compressor.

Hybrid Machine Learning Approaches for 5G Traffic Prediction

ABSTRACT Accurate traffic prediction poses great difficulties because of the continuously increasing scale and diversity of 5G network traffic, which is driven by user demands. Moreover, certain characteristics of 5G traffic are constantly changing; thus, simulations using traditional models often lead to incorrect estimations or inefficient utilization of available resources. Consequently, we propose a hybrid machine learning model that integrates support vector machine (SVM) and decision tree algorithms to enhance efficiency of 5G traffic prediction. The structure of the hybrid model dynamically adjusts by adding or removing hidden layers and units within the network to improve prediction performance. The efficacy of the proposed model is evaluated using metrics like mean squared error, mean absolute error, and root mean squared error (RMSE). Findings show that the hybrid model consistently achieves lower error rates than SVM alone. Further performance enhancement of the hybrid model in predicting 5G traffic is also supported by comparisons of R-squared values against signal-to-noise ratios. These outcomes show the potential of the proposed method to improve traffic prediction accuracy in 5G networks, serving as a powerful tool for network control.

Adrenal indeterminate nodules: CT-based radiomics analysis of different machine learning models for predicting adrenal metastases in lung cancer patients

ObjectiveThere is a paucity of research using different machine learning algorithms for distinguishing between adrenal metastases and benign tumors in lung cancer patients with adrenal indeterminate nodules based on plain and biphasic-enhanced CT radiomics.Materials and MethodsThis study retrospectively enrolled 292 lung cancer patients with adrenal indeterminate nodules (training dataset, 205 (benign, 96; metastases, 109); testing dataset, 87 (benign, 42; metastases, 45)). Radiomics features were extracted from the plain, arterial, and portal CT images, respectively. The independent risk radiomics features selected by least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression (LR) were used to construct the single-phase and combined-phase radiomics models, respectively, by support vector machine (SVM), decision tree (DT), random forest (RF), and LR. The independent clinical-pathological and radiological risk factors for predicting adrenal metastases selected by using univariate and multivariate LR were used to develop the traditional model. The optimal model was selected by ROC curve, and the models’ clinical values were estimated by decision curve analysis (DCA).ResultsIn the testing dataset, all SVM radiomics models showed the best robustness and efficiency, and then RF, LR, and DT models. The combined radiomics model had the best ability in predicting adrenal metastases (AUC=0.938), and then the plain (AUC=0.935), arterial (AUC=0.870), and portal radiomics model (AUC=0.851). Besides, compared to clinical-pathological-radiological model (AUC=0.870), the discriminatory capability of the plain and combined radiomics model were further improved. All radiomics models had good calibration curves and DCA showed the plain and combined radiomics models had more optimal clinical efficacy compared to other models, with the combined radiomics model having the largest net benefit.ConclusionsThe combined SVM radiomics model can non-invasively and efficiently predict adrenal metastatic nodules in lung cancer patients. In addition, the plain radiomics model with high predictive performance provides a convenient and accurate new method for patients with contraindications in enhanced CT.