RESUMO

The non-invasive detection of crack/cocaine and other bioactive compounds from its pyrolysis in saliva can provide an alternative for drug analysis in forensic toxicology. Therefore, a highly sensitive, fast, reagent-free, and sustainable approach with a non-invasive specimen is relevant in public health. In this animal model study, we evaluated the effects of exposure to smoke crack cocaine on salivary flow, salivary gland weight, and salivary composition using Attenuated total reflection-Fourier transform infrared (ATR-FTIR) spectroscopy. The exposure to crack cocaine was performed in an acrylic box apparatus with a burned activation of crack/cocaine 400 mg for 10 min for 14 consecutive days. Crack/cocaine exposure increased the salivary secretion without changes in parotid and submandibular weights. Hierarchical Clustering Analysis (HCA) was applied to depict subgrouping patterns in infrared spectra, and Principal components analysis (PCA) explained 83.2 % of the cumulative variance using 3 PCs. ATR-FTIR platforms were coupled to AdaBoost, Artificial Neural Networks, Naïve Bayes, Random Forest, and Support Vector Machine (SVM) algorithms tool to identify changes in the infrared salivary spectra of rats exposed to crack cocaine. The best classification of crack cocaine exposure using the salivary spectra was performed by Naïve Bayes, presenting a sensitivity of 100 %, specificity of 80 %, and accuracy of 90 % between crack cocaine and control rats. The SHAP features of salivary infrared spectra mostly indicate the vibrational modes at 1331 cm-1 and 2806 cm-1, representing CH2 wagging commonly linked in lipids and C-H stretch often attributed to the CH2 or CH3 groups in lipid molecules, respectively, as the main responsible vibrational modes for crack cocaine exposure discrimination. In summary, the present pre-clinical findings indicate the potential of the ATR-FTIR platform coupled with machine learning to effectively detect changes in salivary infrared spectra promoted by exposure to crack cocaine.

RESUMO

Ecological niche modeling (ENM) is a valuable tool for inferring suitable environmental conditions and estimating species' geographic distributions. ENM is widely used to assess the potential effects of climate change on species distributions; however, the choice of modeling algorithm introduces substantial uncertainty, especially since future projections cannot be properly validated. In this study, we evaluated the performance of seven popular modeling algorithms-Bioclim, generalized additive models (GAM), generalized linear models (GLM), boosted regression trees (BRT), Maxent, random forest (RF), and support vector machine (SVM)-in transferring ENM across time, using Mexican endemic rodents as a model system. We used a retrospective approach, transferring models from the near past (1950-1979) to more recent conditions (1980-2009) and vice versa, to evaluate their performance in both forecasting and hindcasting. Consistent with previous studies, our results highlight that input data quality and algorithm choice significantly impact model accuracy, but most importantly, we found that algorithm performance varied between forecasting and hindcasting. While no single algorithm outperformed the others in both temporal directions, RF generally showed better performance for forecasting, while Maxent performed better in hindcasting, though it was more sensitive to small sample sizes. Bioclim consistently showed the lowest performance. These findings underscore that not all species or algorithms are suited for temporal projections. Therefore, we strongly recommend conducting a thorough evaluation of the data quality-in terms of quantity and potential biases-of the species of interest. Based on this assessment, appropriate algorithm(s) should be carefully selected and rigorously tested before proceeding with temporal transfers.

RESUMO

BACKGROUND: Emergency department (ED) overcrowding is an important problem in many countries. Accurate predictions of ED patient arrivals can help management to better allocate staff and medical resources. In this study, we investigate the use of calendar and meteorological predictors, as well as feature-engineered variables, to predict daily patient arrivals using datasets from eleven different EDs across three countries. METHODS: Six machine learning (ML) algorithms were tested on forecasting horizons of 7 and 45 days. Three of them - Light Gradient Boosting Machine (LightGBM), Support Vector Machine with Radial Basis Function (SVM-RBF), and Neural Network Autoregression (NNAR) - were never before reported for predicting ED patient arrivals. Algorithms' hyperparameters were tuned through a grid-search with cross-validation. Prediction performance was assessed using fivefold cross-validation and four performance metrics. RESULTS: The eXtreme Gradient Boosting (XGBoost) was the best-performing model on both prediction horizons, also outperforming results reported in past studies on ED arrival prediction. XGBoost and NNAR achieved the best performance in nine out of the eleven analyzed datasets, with MAPE values ranging from 5.03% to 14.1%. Feature engineering (FE) improved the performance of the ML algorithms. CONCLUSION: Accuracy in predicting ED arrivals, achieved through the FE approach, is key for managing human and material resources, as well as reducing patient waiting times and lengths of stay.

RESUMO

Autonomous underwater vehicles (AUV) constitute a specific type of cyber-physical system that utilize electronic, mechanical, and software components. A component-based approach can address the development complexities of these systems through composable and reusable components and their integration, simplifying the development process and contributing to a more systematic, disciplined, and measurable engineering approach. In this article, we propose an architecture to design and describe the optimal performance of components for an AUV engineering process. The architecture involves a computing approach that carries out the automatic control of a testbed using genetic algorithms, where components undergo a 'physical-running' evaluation. The procedure, defined from a method engineering perspective, complements the proposed architecture by demonstrating its application. We conducted an experiment to determine the optimal operating modes of an AUV thruster with a flexible propeller using the proposed method. The results indicate that it is feasible to design and assess physical components directly using genetic algorithms in real-world settings, dispensing with the corresponding computational model and associated engineering stages for obtaining an optimized and tested operational scope. Furthermore, we have developed a cost-based model to illustrate that designing an AUV from a physical-running perspective encompasses extensive feasibility zones, where it proves to be more cost-effective than an approach based on simulation.

RESUMO

The wake effect is a relevant factor in determining the optimal distribution of wind turbines within the boundaries of a wind farm. This reduces the incident wind speed on downstream wind turbines, which results in a decrease in energy production for the wind farm. This paper proposes a novel approach for optimizing the distribution of wind turbines using a new Genetic Gray Wolf Optimizer (GGWO). The GGWO employs a teamwork model inspired by wolf prey hunting, guided by four leaders: Alpha, Beta, Delta, and Omicron wolves, each with different hierarchical weights. To improve the competitiveness of the wolves, GGWO utilizes genetic algorithm operators such as crossover blending, normal mutation, and a new genetic operator called Random Selective Mutation (RSM), which improves solution search efficiency. The proposed GGWO is compared to other algorithms such as Gray Wolf Optimizer (GWO), Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC), and Ant Colony Optimization (ACO). The studies examine varying wind speeds in magnitude and direction throughout the year, as well as different wind farm boundaries. The outcomes show that GGWO successfully identifies the ideal locations for wind turbines, scoring better scores in terms of total simulation duration and annual energy generation for the wind farm. It surpasses the performance of GWO, ABC, and PSO algorithms and exhibits comparable competitiveness with more intricate algorithms like ACO.

RESUMO

Nowadays, metaheuristic algorithms have been applied to optimize last lower-surface models. Also, the last lower-surface model has been adjusted through the computational algorithms to perform custom shoe lasts. Therefore, it is necessary to implement nature-inspired metaheuristic algorithms to perform the adjustment of last lower-surface model to the footprint topography. In this study, a metaheuristic genetic algorithm is implemented to adjust the last lower surface model to the footprint topography. The genetic algorithm is constructed through an objective function, which is defined through the last lower Bezier model and footprint topography, where a mean error function moves the last lower surface toward the footprint topography through the initial population. Also, the search space is deduced from the last lower surface and footprint topography. In this way, the genetic algorithm performs explorations and exploitations to optimize a Bezier surface model, which generates the adjusted last lower surface, where the surface is recovered via laser line scanning. Thus, the metaheuristic algorithm enhances the last lower-surface adjustment to improve the custom last manufacture. This contribution is elucidated by a discussion based on the proposed metaheuristic algorithm for surface model adjustment and the optimization methods implemented in recent years.

RESUMO

Perception systems for assisted driving and autonomy enable the identification and classification of objects through a concentration of sensors installed in vehicles, including Radio Detection and Ranging (RADAR), camera, Light Detection and Ranging (LIDAR), ultrasound, and HD maps. These sensors ensure a reliable and robust navigation system. Radar, in particular, operates with electromagnetic waves and remains effective under a variety of weather conditions. It uses point cloud technology to map the objects in front of you, making it easy to group these points to associate them with real-world objects. Numerous clustering algorithms have been developed and can be integrated into radar systems to identify, investigate, and track objects. In this study, we evaluate several clustering algorithms to determine their suitability for application in automotive radar systems. Our analysis covered a variety of current methods, the mathematical process of these methods, and presented a comparison table between these algorithms, including Hierarchical Clustering, Affinity Propagation Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH), Density-Based Spatial Clustering of Applications with Noise (DBSCAN), Mini-Batch K-Means, K-Means Mean Shift, OPTICS, Spectral Clustering, and Gaussian Mixture. We have found that K-Means, Mean Shift, and DBSCAN are particularly suitable for these applications, based on performance indicators that assess suitability and efficiency. However, DBSCAN shows better performance compared to others. Furthermore, our findings highlight that the choice of radar significantly impacts the effectiveness of these object recognition methods.

RESUMO

Background: The global need for rapid diagnostic methods for pathogen identification and antimicrobial susceptibility testing (AST) is underscored by the increasing bacterial resistance and limited therapeutic options, especially critical in sepsis management. Summary: This review examines the aspects of the eHealth and mHealth in Antimicrobial Stewardship Programs (ASPs) to improve the treatment of infections and rational use of antimicrobials. Key Messages: The evolution from traditional phenotype-based methods to rapid molecular and mass spectrometry techniques has significantly decreased result turnaround times, improving patient outcomes. Despite advancements, the complex decision-making in antimicrobial therapy often exceeds the capacity of many clinicians, highlighting the importance of ASPs. These programs, integrating mHealth and eHealth, leverage technology to enhance healthcare services and patient outcomes, particularly in remote or resource-limited settings. However, the application of such technologies in antimicrobial management remains underexplored in hospitals. The development of platforms combining antimicrobial prescription data with pharmacotherapeutic algorithms and laboratory integration can significantly reduce costs and improve hospitalization times and mortality rates.

RESUMO

BACKGROUND: Fabry disease (FD) is a rare X-linked lysosomal storage disorder marked by alpha-galactosidase-A (α-Gal A) deficiency, caused by pathogenic mutations in the GLA gene, resulting in the accumulation of glycosphingolipids within lysosomes. The current screening test relies on measuring α-Gal A activity. However, this approach is limited to males. Infrared (IR) spectroscopy is a technique that can generate fingerprint spectra of a biofluid's molecular composition and has been successfully applied to screen numerous diseases. Herein, we investigate the discriminating vibration profile of plasma chemical bonds in patients with FD using attenuated total reflection Fourier-transform IR (ATR-FTIR) spectroscopy. RESULTS: The Fabry disease group (n = 47) and the healthy control group (n = 52) recruited were age-matched (39.2 ± 16.9 and 36.7 ± 10.9 years, respectively), and females were predominant in both groups (59.6% and 65.4%, respectively). All patients had the classic phenotype (100%), and no late-onset phenotype was detected. A generated partial least squares discriminant analysis (PLS-DA) classification model, independent of gender, allowed differentiation of samples from FD vs. control groups, reaching 100% sensitivity, specificity and accuracy. CONCLUSION: ATR-FTIR spectroscopy harnessed to pattern recognition algorithms can distinguish between FD patients and healthy control participants, offering the potential of a fast and inexpensive screening test.

RESUMO

Machine learning classification approaches were used to discriminate a fishy off-flavour identified in beef with health-enhanced fatty acid profiles. The random forest approach outperformed (P < 0.001; receiver operating characteristic curve: 99.8 %, sensitivity: 99.9 % and specificity: 93.7 %) the logistic regression, partial least-squares discrimination analysis and the support vector machine (linear and radial) approaches, correctly classifying 100 % and 82 % of the fishy and non-fishy meat samples, respectively. The random forest algorithm identified 20 volatile compounds responsible for the discrimination of fishy from non-fishy meat samples. Among those, seven volatile compounds (pentadecane, octadecane, γ-dodecalactone, dodecanal, (E,E)-2,4-heptadienal, 2-heptanone, and ethylbenzene) were selected as significant contributors to the fishy off-flavour fingerprint, all being related to lipid oxidation. This fishy off-flavour fingerprint could facilitate the rapid monitoring of beef with enhanced healthy fatty acids to avoid consumer dissatisfaction due to fishy off-flavour.

RESUMO

The current detection method for Chikungunya Virus (CHIKV) involves an invasive and costly molecular biology procedure as the gold standard diagnostic method. Consequently, the search for a non-invasive, more cost-effective, reagent-free, and sustainable method for the detection of CHIKV infection is imperative for public health. The portable Fourier-transform infrared coupled with Attenuated Total Reflection (ATR-FTIR) platform was applied to discriminate systemic diseases using saliva, however, the salivary diagnostic application in viral diseases is less explored. The study aimed to identify unique vibrational modes of salivary infrared profiles to detect CHIKV infection using chemometrics and artificial intelligence algorithms. Thus, we intradermally challenged interferon-gamma gene knockout C57/BL6 mice with CHIKV (20 µl, 1 X 105 PFU/ml, n = 6) or vehicle (20 µl, n = 7). Saliva and serum samples were collected on day 3 (due to the peak of viremia). CHIKV infection was confirmed by Real-time PCR in the serum of CHIKV-infected mice. The best pattern classification showed a sensitivity of 83%, specificity of 86%, and accuracy of 85% using support vector machine (SVM) algorithms. Our results suggest that the salivary ATR-FTIR platform can discriminate CHIKV infection with the potential to be applied as a non-invasive, sustainable, and cost-effective detection tool for this emerging disease.

RESUMO

To evaluate our two non-machine learning (non-ML)-based algorithmic approaches for detecting early ischemic infarcts on brain CT images of patients with acute ischemic stroke symptoms, tailored to our local population, to be incorporated in our telestroke software. One-hundred and thirteen acute stroke patients, excluding hemorrhagic, subacute, and chronic patients, with accessible brain CT images were divided into calibration and test sets. The gold standard was determined through consensus among three neuroradiologist. Four neuroradiologist independently reported Alberta Stroke Program Early CT Scores (ASPECTSs). ASPECTSs were also obtained using a commercial ML solution (CMLS), and our two methods, namely the Mean Hounsfield Unit (HU) relative difference (RELDIF) and the density distribution equivalence test (DDET), which used statistical analyze the of the HUs of each region and its contralateral side. Automated segmentation was perfect for cortical regions, while minimal adjustment was required for basal ganglia regions. For dichotomized-ASPECTSs (ASPECTS < 6) in the test set, the area under the receiver operating characteristic curve (AUC) was 0.85 for the DDET method, 0.84 for the RELDIF approach, 0.64 for the CMLS, and ranged from 0.71-0.89 for the neuroradiologist. The accuracy was 0.85 for the DDET method, 0.88 for the RELDIF approach, and was ranged from 0.83 - 0.96 for the neuroradiologist. Equivalence at a margin of 5% was documented among the DDET, RELDIF, and gold standard on mean ASPECTSs. Noninferiority tests of the AUC and accuracy of infarct detection revealed similarities between both DDET and RELDIF, and the CMLS, and with at least one neuroradiologist. The alignment of our methods with the evaluations of neuroradiologist and the CMLS indicates the potential of our methods to serve as supportive tools in clinical settings, facilitating prompt and accurate stroke diagnosis, especially in health care settings, such as Colombia, where neuroradiologist are limited.

RESUMO

Recent studies introduced the importance of using machine learning algorithms in research focused on the identification of antibiotic resistance. In this study, we highlight the importance of building solid machine learning foundations to differentiate antimicrobial resistance among microorganisms. Using advanced machine learning algorithms, we established a methodology capable of analyzing the FTIR structural profile of the samples of Streptococcus pyogenes and Streptococcus mutans (Gram-positive), as well as Escherichia coli and Klebsiella pneumoniae (Gram-negative), demonstrating cross-sectional applicability in this focus on different microorganisms. The analysis focuses on specific biomolecules-Carbohydrates, Fatty Acids, and Proteins-in FTIR spectra, providing a multidimensional database that transcends microbial variability. The results highlight the ability of the method to consistently identify resistance patterns, regardless of the Gram classification of the bacteria and the species involved, reinforcing the premise that the structural characteristics identified are universal among the microorganisms tested. By validating this approach in four distinct species, our study proves the versatility and precision of the methodology used, in addition to bringing support to the development of an innovative protocol for the rapid and safe identification of antimicrobial resistance. This advance is crucial for optimizing treatment strategies and avoiding the spread of resistance. This emphasizes the relevance of specialized machine learning bases in effectively differentiating between resistance profiles in Gram-negative and Gram-positive bacteria to be implemented in the identification of antibiotic resistance. The obtained result has a high potential to be applied to clinical procedures.

RESUMO

Introducción: el cólico nefrítico es una de las causas más frecuentes de consulta en puerta de emergencia. Cuando se presenta en una paciente embarazada, genera un desafío diagnóstico y terapéutico que requiere un abordaje multidisciplinario. Materiales y métodos: se realizó una búsqueda bibliográfica en la base de datos MedLine/PubMed considerando revisiones sistemáticas de literatura, reportes de casos clínicos, estudios observacionales retrospectivos publicados en los últimos 10 años, con el objetivo de obtener sustento informativo para crear un algoritmo diagnóstico y terapéutico que plantee el manejo del cólico nefrítico en la embarazada, dirigido a médicos emergencistas, urólogos y ginecólogos. Resultados: se obtuvieron en total 39 artículos, que fueron analizados, trabajando finalmente en base a 17 textos, que son los citados. Discusión: el diagnóstico se basa en la historia clínica, examen físico, pruebas de laboratorio e imagen. Tratamiento de inicio conservador, que incluye hidratación, analgésicos y antieméticos, reservando la utilización de antibióticos para cuadros infecciosos. De no funcionar éste, se optará por tratamiento intervencionista. Conclusiones: la embarazada con cólico nefrítico se estudia en base a paraclínica humoral y de imagen (ecografía, resonancia nuclear magnética y tomografía axial computada de baja dosis). El tratamiento es principalmente conservador, ante la falla del mismo o ante cuadros infecciosos es quirúrgico.

Introduction: renal colic is one of the most frequent causes of emergency room visits. When it occurs in a pregnant patient, it generates a diagnostic and therapeutic challenge that requires a multidisciplinary approach. Materials and methods: a bibliographic search was carried out in the MedLine/PubMed database considering systematic literature reviews, clinical case reports, retrospective observational studies published in the last 10 years with the aim of obtaining information to create a diagnostic and therapeutic algorithm for the management of nephritic colic in pregnant women, aimed at emergency physicians, urologists and gynecologists. Results: a total of 39 articles were obtained and analyzed, finally working on the basis of 17 texts, which are those cited. Discussion: diagnosis is based on clinical history, physical examination, laboratory and imaging tests. Conservative initial treatment, including hydration, analgesics and antiemetics, reserving the use of antibiotics for infectious conditions. If this does not work, interventional treatment will be chosen. Conclusions: pregnant women with renal colic are studied on the basis of humoral and imaging (ultrasound, magnetic resonance imaging and low dose computed axial tomography). The treatment is mainly conservative; in case of failure or infectious conditions, surgery is performed.

Introdução: a cólica renal é uma das causas mais frequentes de consulta no departamento de emergência. Quando ocorre em uma paciente grávida, gera um desafio diagnóstico e terapêutico que exige uma abordagem multidisciplinar. Materiais e métodos: foi realizada uma pesquisa bibliográfica no banco de dados MedLine/PubMed, considerando revisões sistemáticas da literatura, relatos de casos clínicos, estudos observacionais retrospectivos publicados nos últimos 10 anos, com o objetivo de obter informações para a criação de um algoritmo diagnóstico e terapêutico para o manejo da cólica nefrética em gestantes, destinado a médicos de emergência, urologistas e ginecologistas. Resultados: um total de 39 artigos foi obtido e analisado, sendo que, por fim, trabalhamos com base em 17 textos, que são os citados. Discussão: o diagnóstico é baseado na história clínica, exame físico, exames laboratoriais e de imagem. O tratamento inicial é conservador, incluindo hidratação, analgésicos e antieméticos, reservando o uso de antibióticos para quadros infecciosos. Se isso não funcionar, o tratamento intervencionista será escolhido. Conclusões: as gestantes com cólica renal são estudadas com base em exames humorais e de imagem (ultrassom, ressonância magnética e tomografia axial computadorizada de baixa dose). O tratamento é principalmente conservador, com cirurgia em caso de falha ou condições infecciosas.

RESUMO

In this paper, we aim to enhance genetic algorithms (GAs) by integrating a dynamic model based on biological life cycles. This study addresses the challenge of maintaining diversity and adaptability in GAs by incorporating stages of birth, growth, reproduction, and death into the algorithm's framework. We consider an asynchronous execution of life cycle stages to individuals in the population, ensuring a steady-state evolution that preserves high-quality solutions while maintaining diversity. Experimental results demonstrate that the proposed extension outperforms traditional GAs and is as good or better than other well-known and well established algorithms like PSO and EvoSpace in various benchmark problems, particularly regarding convergence speed and solution qu/ality. The study concludes that incorporating biological life-cycle dynamics into GAs enhances their robustness and efficiency, offering a promising direction for future research in evolutionary computation.

RESUMO

The urgent requirement for swift diagnostic methods in pathogen identification and antimicrobial susceptibility testing is emphasized by rising bacterial resistance and limited treatment options, which are particularly critical in sepsis management. The shift from traditional phenotype-based methods to rapid molecular and mass spectrometry techniques has significantly reduced result turnaround times, enhancing patient outcomes. In this systematic review with meta-analysis, the aspects of correct empirical antimicrobial therapy are evaluated to determine their impact on mortality. We performed a systematic review and meta-analysis on EMBASE, the Cochrane Library, Web of Science, and MEDLINE. Studies evaluating mortality associated with empirical adequate and inadequate therapy in different sites of infection were included. Outcomes included clinical cures in microbiologically evaluable patients. Among the sites of infection, the most studied were bloodstream infections (n = 9), followed by respiratory tract infections (n = 5), intra-abdominal infections (n = 5), and urinary tract infections (evaluated by 3 studies). Inadequate therapy was associated with an increase in mortality between 11 and 68%. Technologies to speed up pathogen identification are extremely necessary to reduce mortality.

RESUMO

This paper explores the application of complex network models and genetic algorithms in epidemiological modeling. By considering the small-world and Barabási-Albert network models, we aim to replicate the dynamics of disease spread in urban environments. This study emphasizes the importance of accurately mapping individual contacts and social networks to forecast disease progression. Using a genetic algorithm, we estimate the input parameters for network construction, thereby simulating disease transmission within these networks. Our results demonstrate the networks' resemblance to real social interactions, highlighting their potential in predicting disease spread. This study underscores the significance of complex network models and genetic algorithms in understanding and managing public health crises.

RESUMO

INTRODUCTION AND OBJECTIVES: The increasing incidence of hepatocellular carcinoma (HCC) in China is an urgent issue, necessitating early diagnosis and treatment. This study aimed to develop personalized predictive models by combining machine learning (ML) technology with a demographic, medical history, and noninvasive biomarker data. These models can enhance the decision-making capabilities of physicians for HCC in hepatitis B virus (HBV)-related cirrhosis patients with low serum alpha-fetoprotein (AFP) levels. PATIENTS AND METHODS: A total of 6,980 patients treated between January 2012 and December 2018 were included. Pre-treatment laboratory tests and clinical data were obtained. The significant risk factors for HCC were identified, and the relative risk of each variable affecting its diagnosis was calculated using ML and univariate regression analysis. The data set was then randomly partitioned into validation (20 %) and training sets (80 %) to develop the ML models. RESULTS: Twelve independent risk factors for HCC were identified using Gaussian naïve Bayes, extreme gradient boosting (XGBoost), random forest, and least absolute shrinkage and selection operation regression models. Multivariate analysis revealed that male sex, age >60 years, alkaline phosphate >150 U/L, AFP >25 ng/mL, carcinoembryonic antigen >5 ng/mL, and fibrinogen >4 g/L were the risk factors, whereas hypertension, calcium <2.25 mmol/L, potassium ≤3.5 mmol/L, direct bilirubin >6.8 µmol/L, hemoglobin <110 g/L, and glutamic-pyruvic transaminase >40 U/L were the protective factors in HCC patients. Based on these factors, a nomogram was constructed, showing an area under the curve (AUC) of 0.746 (sensitivity = 0.710, specificity=0.646), which was significantly higher than AFP AUC of 0.658 (sensitivity = 0.462, specificity=0.766). Compared with several ML algorithms, the XGBoost model had an AUC of 0.832 (sensitivity = 0.745, specificity=0.766) and an independent validation AUC of 0.829 (sensitivity = 0.766, specificity = 0.737), making it the top-performing model in both sets. The external validation results have proven the accuracy of the XGBoost model. CONCLUSIONS: The proposed XGBoost demonstrated a promising ability for individualized prediction of HCC in HBV-related cirrhosis patients with low-level AFP.

RESUMO

BACKGROUND: Colorectal cancer has a high incidence and mortality rate due to a low rate of early diagnosis. Therefore, efficient diagnostic methods are urgently needed. PURPOSE: This study assesses the diagnostic effectiveness of Carbohydrate Antigen 19-9 (CA19-9), Carcinoembryonic Antigen (CEA), Alpha-fetoprotein (AFP), and Cancer Antigen 125 (CA125) serum tumor markers for colorectal cancer (CRC) and investigates a machine learning-based diagnostic model incorporating these markers with blood biochemical indices for improved CRC detection. METHOD: Between January 2019 and December 2021, data from 800 CRC patients and 697 controls were collected; 52 patients and 63 controls attending the same hospital in 2022 were collected as an external validation set. Markers' effectiveness was analyzed individually and collectively, using metrics like ROC curve AUC and F1 score. Variables chosen through backward regression, including demographics and blood tests, were tested on six machine learning models using these metrics. RESULT: In the case group, the levels of CEA, CA199, and CA125 were found to be higher than those in the control group. Combining these with a fourth serum marker significantly improved predictive efficacy over using any single marker alone, achieving an Area Under the Curve (AUC) value of 0.801. Using stepwise regression (backward), 17 variables were meticulously selected for evaluation in six machine learning models. Among these models, the Gradient Boosting Machine (GBM) emerged as the top performer in the training set, test set, and external validation set, boasting an AUC value of over 0.9, indicating its superior predictive power. CONCLUSION: Machine learning models integrating tumor markers and blood indices offer superior CRC diagnostic accuracy, potentially enhancing clinical practice.

RESUMO

RESUMEN Esta investigación presenta una metodología para optimizar fuerzas de control en edificaciones, las cuales se encuentran sometidas a cargas sísmicas. Se desarrolló un sistema de control llamado CLF-MR_I, el cuál combina un algoritmo genético de clasificación no dominada NSGA-II y un sistema de control basado en lógica difusa. El controlador fue ensayado numéricamente en una edificación real de 96 m de altura, en la cual se instalaron 6 amortiguadores magnetoreológicos MR. La estructura fue sometida a 8 aceleraciones de sismo con diferentes rangos frecuenciales. Los parámetros de entrada para el sistema de control propuesto fueron los desplazamientos y las velocidades del primer piso de la edificación y como único parámetro de salida, se definió el voltaje de los dispositivos MR. La eficiencia del CLF-MR_1 fue comparada con un segundo controlador llamado CLF-MR_2, el cual funciona mediante un sistema de inferencia basado en parámetros lingüísticos. Los resultados obtenidos indican que el CLF-MR_1 mejora significativamente la respuesta dinámica de la edificación, en comparación con los resultados obtenidos con el CLF-MR_2 y con la condición no controlada de la edificación.

ABSTRACT This research presents a methodology to optimize control forces in buildings, which are subjected to seismic loads. A control system called CLF-MR_1 was developed, which combines a genetic algorithm of non-dominated classification NSGA-II and a control system based on fuzzy logic. The controller was numerically evaluated in a real 96 m high building, in which 6 MR magnetorheological dampers were installed. The structure was subjected to 8 earthquake accelerations with different frequency ranges. The input parameters for the proposed control system were the displacements and velocities of the first floor of the building and the only output parameter was the voltage of the MR devices. The efficiency of CLF-MR_1 was compared with a second controller called CLF-MR_2, which operates using an inference system based on linguistic parameters. Results obtained show that CLF-MR_1 significantly improves the dynamic response of the building, compared to the results obtained with CLF-MR_2 and the uncontrolled condition of the building.

RESUMO Esta pesquisa apresenta uma metodologia para otimizar as forças de controle em edifícios sujeitos a cargas sísmicas. Foi desenvolvido um sistema de controle denominado CLF-MR_1, que combina um algoritmo genético de classificação não dominada NSGA-II e um sistema de controle baseado em lógica difusa. O controlador foi testado numericamente em um edifício real de 96 m de altura, no qual foram instalados 6 amortecedores magnetorheológicos MR. A estrutura foi submetida a 8 acelerações de terremoto com diferentes faixas de frequência. Os parâmetros de entrada para o sistema de controle proposto foram os deslocamentos e as velocidades do primeiro andar do edifício, e a tensão dos dispositivos MR foi definida como o único parâmetro de saída. A eficiência do CLF-MR_1 foi comparada com um segundo controlador chamado CLF-MR_2, que opera usando um sistema de inferência baseado em parâmetros linguísticos. Os resultados obtidos indicam que o CLF-MR_1 melhora significativamente a resposta dinâmica do edifício, em comparação com os resultados obtidos com o CLF-MR_2 e a condição não controlada do edifício.