A hanseníase é um agravo relevante na saúde pública pelo seu potencial incapacitante e sua presença expressiva no Brasil. Este estudo comparou modelos de IA aplicados a dados tabulares do SINAN para prever o GIF final de pacientes. Foram avaliados modelos baseados em árvores (RF, LightGBM e CatBoost) e redes neurais (MLP, ResNet e Transformer). O LightGBM apresentou desempenho superior e maior estabilidade entre classes, com um AUC OvO de 71,10%. As redes neurais mostraram um desempenho competitivo, sobretudo o Transformer, com um AUC OvO de 70,69%. Conclui-se que, conforme a base utilizada, modelos baseados em árvores são mais adequados ao prognóstico do GIF, mas as redes neurais são alternativas para contextos multimodais.

The growing availability of large-scale health data has created new opportunities for applying machine learning in public health systems. However, processing these datasets requires scalable computational infrastructures. This paper presents an experience report on the development of cloud-based architectures for scientific health systems using Amazon Web Services (AWS). We analyze three platforms: VALERIA for arboviral disease diagnosis, ANGELS for gestational monitoring, and IAra for malaria forecasting. These systems rely on large epidemiological datasets and machine learning pipelines executed using AWS services such as EC2, S3, ECR, Lambda, and SageMaker. Results show that cloud infrastructures enable scalable data processing, reproducible experimentation, and operational deployment of intelligent healthcare applications.

Purpose : Leprosy (known as Hansen’s disease in Brazil) continues to pose a major public health challenge in several endemic countries. The Simplified Neurological Assessment (from the Portuguese Avalia¸c˜ao Neurol´ogica Simplificada - ANS) is the standard procedure recommended by the Brazilian Ministry of Health to evaluate nerve function in individuals affected by leprosy, and is essential for preventing irreversible disabilities. However, its manual implementation limits data standardization, follow-up, and integration into digital health systems. This study presents the ANSd (Digital Simplified Neurological Assessment, from the Portuguese Avalia¸c˜ao Neurol´ogica Simplificada digital ), a mobile health application designed to support standardized neurological evaluation, improve documentation, and enhance longitudinal monitoring in leprosy care. Methods : The development of ANSd application was conducted as action research, applying Soft Systems Methodology (SSM) through iterative learning cycles with the Global Partnership for Zero Leprosy (GPZL) digital health priorities. The process included domain and requirement modeling, data model design, interface design, and software implementation. Results : ANSd enables digital recording of sensory and motor testing results, aggregates physical disability grading (from Portuguese, Grau de Incapacidade F´ısica - GIF) and the Eye-Hand-Foot (EHF) score, and stores structured records to support longitudinal patient follow-up. The ANSd application runs entirely offline through an offline-first architecture with encrypted on-device storage, and reproduces the official ANS form as a populated PDF, preserving regulatory compliance while enabling structured data capture. Conclusion : By consolidation standardized clinical data in a single digital platform, ANSd strengthens neurological monitoring and supports early detection of nerve impairment in individuals affected by leprosy. The ANSd application aligns with the GPZL’s strategic vision for digital health and offers a scalable solution for improving disability prevention and data-driven decisionmaking in leprosy control programs. Within the Brazilian public healthcare system (from Portuguese Sistema ´ Unico de Sa´ude - SUS), ANSd has the potential to enhance clinical workflows, facilitate integration with national surveillance platforms, and improve the continuity and quality of care for people affected by leprosy.

During the neonatal period, newborns are more susceptible to developing conditions and diseases due to their fragility of the transition and adaptation to the extrauterine environment. Neonatal sepsis is one of the leading causes of morbidity and mortality in newborns, particularly among preterm and low birth weight infants, requiring early diagnosis to reduce complications and deaths. In this work, we evaluate the performance of artificial intelligence models in predicting neonatal sepsis and also identify the attributes that most contribute impact on models’ learning and their relationship with the disease, using real data from the state of Pernambuco, Brazil. The six machine learning models evaluated were AdaBoost, CatBoost, Gradient Boosting, LightGBM, Random Forest and XGBoost. Performance metrics ranged from 0.7213 to 0.8548, with AdaBoost and LightGBM achieving the best results, reaching a sensitivity above 0.8197 and a specificity of 0.8397 in all three experiments. SHAPley Additive exPlanations (SHAP) analysis revealed strong relationships between sepsis and attributes such as intracranial hemorrhage, prematurity, CPAP use, TTN presence, and epicutaneous access, all of which were highly associated with sepsis cases. We conclude that the artificial intelligence models demonstrated promising results in predicting neonatal sepsis, highlighting critical clinical attributes associated with the disease and identifying the most relevant predictors.

The reduction of fetal, neonatal, and infant mortality rates is crucial in maternal and neonatal care. The UN’s 2030 Agenda aims for a better world by 2030, highlighting goal 3: healthy lives for all ages, with a focus on reducing global maternal and child mortality. In this context, prenatal care plays a vital role in identifying pregnant women at risk and enabling proactive interventions to minimize adverse outcomes, whether mortality or morbidity. This work presents the ANGELS (An iNtelligent GEstational foLlow-up System). This work presents ANGELS, an intelligent gestational follow-up platform designed to integrate and operationalize multiple machine learning models within a unified decision support system for maternal and neonatal care. The platform provides integrated services through a modular, API-based architecture, enabling the incorporation of different predictive models throughout pregnancy, childbirth, and postpartum periods. The system is validated in this study through an exemplar case focused on congenital syphilis risk prediction using real-world Brazilian data, demonstrating howpreviously validated models can be integrated, extended, and deployed within the ANGELS platform. In the exemplar case of congenital syphilis risk prediction, the complementary version of ANGELS achieved substantial improvements, with specificity and precision increasing from approximately 60% in the basic version to values close to 95%, reflecting known trade-offs across evaluation metrics. The approach proved promising for practical incorporation into healthcare systems, aiming to provide more accurate and personalized monitoring tailored to individual needs. It is hoped that the ANGELS system may contribute to reduce preventable stillbirths, improve the quality of maternal and child care, and help achieve the UN’s 2030 Agenda goals. Additionally, the integration of ANGELS services into healthcare systems and/or programs will strengthen efforts to promote the well-being of pregnant women and children.

Leprosy, caused by Mycobacterium leprae, remains a global challenge, requiring strategies to achieve disease elimination by 2030. In Brazil, the Simplified Neurological Assessment (from Portuguese Avaliação Neurológica Simplificada, ANS) is mandatory for suspected cases; however, the form is still manually fulfilled, which limits the use of data. This study evaluates computer vision models (YOLOv8x, YOLO11x, Faster R-CNN) for detecting hand and foot sensitivity regions from ANS forms. All models were evaluated based on precision, recall, mean average precision (mAP) and confusion matrix. YOLO variants achieved over 94% precision and 84% recall across all classes. Automating ANS data extraction can facilitate the creation of structured datasets, enhancing disease monitoring and enabling the train of predictive models.

PURPOSE:This study aims to evaluate the performance of machine learning models using different data imputation techniques in different balancing scenarios, employing sociodemographic attributes and maternal health history, using data of a population from the state of Pernambuco, Brazil, to predict fetal death during pregnancy. METHODS:We used a dataset from a social program in Pernambuco, Brazil, covering the period from 2008 to 2022, that includes sociodemographic, prenatal, maternal and family health history data. We separated two scenarios with two balancing techniques to train the models, Random Undersampling (RU scenario) and Hybrid Undersampling 2x (H2X scenario) and we explored using four tree-based machine learning models, each of which was evaluated based on their performance and feature importance. RESULTS:The models were evaluated under different metrics. The XGBoost model stood out with 81.06% specificity and the Random Forest model stood out with 67.73% sensitivity, in different scenarios. The attributes that most impacted the learning process were first prenatal care, age, education and interpregnancy interval. CONCLUSION:This application is particularly valuable in the context of social projects, such as those in Brazil, where innovative solutions can contribute to achieving the SDGs offering a unique perspective on the intersection of technology, healthcare, and social impact.

Changes in telecommunication services demand the development of a new infrastructure to attend new network applications' requirements. The traditional approach to network functions running over dedicated equipment can no longer handle all the dynamics of these new services. The network function virtualization (NFV) paradigm decouples a function from the underlying dedicated hardware thus making networks more flexible and agile. A set of virtual network functions (VNFs) can be deployed as virtual machines or containers across common servers, and orchestrated to compose a service function chain (SFC).Despite the many benefits of NFV, it raises several challenges. SFC placement is a complex task, since it requires taking into consideration the characteristics of VNFs, the SFC requirements, and the state of network infrastructure. This poses a challenge for large scale networks. Information regarding network resources are stored in a large database, and retrieving such data in order to perform SFC placement according to some strategy can be a problem. Limited memory capacity and the presence of a large number of disk operations can compromise the performance of SFC placement algorithms and make it unfeasible. In addition, the amount of memory available to run the allocation algorithm may not be sufficient to load the large amount of information that describe the network resources (occasionally in the hundreds of gigabytes). We address this problem by using a cluster based solution that stores and retrieves data for large scale infrastructures in order to perform SFC placement. The results demonstrate that the use of clusters in the preprocessing step can drastically reduce the size of the resulting database, as well as the execution time to select candidate nodes for a scalable SFC allocation.

Low birth weight (LBW) is a health condition that affects over 20 million gestational outcomes worldwide. The current literature indicates that machine learning models have the potential to assist healthcare professionals in predicting LBW and giving them the opportunity to intervene earlier in the pregnancy, which might include adjusting medical treatments or suggesting changes in diet. This study proposes the evaluation of machine learning models to predict which pregnant women are at risk of neonatal outcomes with LBW. The methodology involves six phases, including data analysis and attribute selection through different techniques, which generated four distinct scenarios. We used five machine learning models and validated them through cross-validation and hyper-parameter optimization and evaluated their performance considering seven distinct metrics and statistical analysis, focusing on the effectiveness of the models in predicting LBW. The results revealed that the models achieved varying levels of performance across the scenarios, with the removal of duplicate data resulting in improvements in recall (0.83) and f1-score (0.64). Statistical analysis confirmed significant differences (p < 0.05) among most models. The conclusions of this study indicate that the removal of duplicate data and careful attribute selection positively influenced the performance of the machine learning models in predicting low birth weight. Additionally, the analysis of attribute importance highlighted socio-demographic characteristics and gestational history as the most influential in the training of the models.

Leprosy, or Hansen's disease, is a Neglected Tropical Disease (NTD) caused by Mycobacterium leprae that mainly affects the skin and peripheral nerves, causing neuropathy to varying degrees. It can result in physical disabilities and functional loss and is particularly prevalent amongst the most vulnerable populations in tropical and subtropical regions worldwide. The persistent stigma and social exclusion associated with leprosy complicate eradication efforts exacerbate the wider challenges faced by NTDs in sourcing the necessary resources and attention for control and elimination. The introduction of Multidrug Therapy (MDT) significantly lowers the global disease burden. Despite this breakthrough in the treatment of leprosy, over 200,000 new leprosy cases are reported annually across more than 120 countries, emphasizing the need for ongoing detection and management efforts. Artificial Intelligence (AI) has the potential to transform leprosy care by accelerating early detection, improving accurate diagnosis, and enabling predictive modeling to improve the quality for those affected. The potential of AI to provide information to assist healthcare professionals in interventions that reduce the risk of disability, and consequently stigma, particularly in endemic regions, presents a promising path to reducing the incidence of leprosy and improving integration social status of patients. This systematic literature review (SLR) examines the state of the art in research on the use of AI for leprosy care. From an initial 657 works from six scientific databases (ACM Digital Library, IEEE Xplore, PubMed, Scopus, Science Direct and Springer), only 30 relevant works were identified, after analysis of three independent reviewers. We have excluded works due duplication, couldn't be retrieved and quality assessment. Results show that current research is focused primarily on the identification of symptoms using image based classification using three main techniques, neural networks, convolutional neural networks, and support vector machines; a small number of studies focus on other thematic areas of leprosy care. A comprehensive systematic approach to research on the application of AI to leprosy care can make a meaningful contribution to a leprosy-free world and help deliver on the promise of the Sustainable Development Goals (SDG).

Premature birth can be defined as birth before 37 weeks of gestation, which is a significant global health issue, being the main cause for neonatal deaths. In this work, we evaluate machine learning models for predicting premature birth using Brazilian sociodemographic and obstetric data, focusing on the challenge of data imbalance, a common problem that can lead to biased predictions. We evaluate five data balancing techniques: Undersampling, Oversampling, and three Hybridsampling configurations where the minority class was increased by factors 2, 3, and 4. The machine learning models, including Decision Tree, Random Forest, and AdaBoost, are trained and evaluated on a dataset of over 483,000 cases. The use of the Hybridsampling approach resulted in an accuracy of 70%, a recall of 64%, and a precision of 74% in the Decision Tree model. Results show that Hybridsampling techniques significantly improves models' performance compared to Undersampling and Oversampling, highlighting the importance of a proper data balancing in predictive models for preterm birth. The relevance of our work is particularly significant for the Brazilian Unified Health System (SUS). By improving the accuracy of premature birth predictions, our models could assist healthcare providers in identifying at-risk pregnancies earlier, allowing for timely interventions. This integration could enhance maternal and neonatal care, reduce the incidence of preterm births, and potentially decrease neonatal mortality, especially in underserved regions.

Malaria remains a major global health challenge, particularly in Brazil’s Legal Amazon region, where environmental and socioeconomic conditions foster favorable conditions for disease transmission. Traditional control measures have shown limited effectiveness, emphasizing the need for better predictive approaches to support timely and targeted public health interventions. This study evaluates the performance of six computational models—Long Short-Term Memory (LSTM), Gated Recurrent Units (GRU), Support Vector Regression (SVR), Random Forest (RF), eXtreme Gradient Boosting (XGBoost), and Autoregressive Integrated Moving Average (ARIMA)—for forecasting weekly malaria cases across multiple states in the Legal Amazon. The results demonstrate that the RF model consistently outperformed the other models, achieving the lowest Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) values in most cases, such as in cluster 02 of the state of Acre, with RMSE of 0.00203 and MAE of 0.00133. The integration of K-means clustering further improved the model predictive accuracy by accounting for spatial heterogeneity and capturing localized transmission dynamics. This hybrid modeling approach, combining machine learning models with spatial clustering, offers a promising tool for enhancing malaria surveillance and guiding more effective public health strategies, especially for malaria control efforts in high-risk regions.

This paper does not have an abstract.

With the emergence of new applications driven by the popularization of mobile devices, the next generation of mobile networks faces challenges to meet different requirements. Virtual Network Functions (VNFs) have been deployed to minimize operational costs and make network management more flexible. In this sense, strategies for VNF placement can impact different metrics of interest. Invoking and visiting VNFs in a specific execution order may be required for different use cases, resulting in a complete network service called Service Function Chain (SFC). The SFC placement problem is to define a feasible path in the physical infrastructure whose nodes and edges meet the computational and bandwidth requirements for the VNFs and virtual links, respectively. It has already been proved that this process is NP-hard and it is difficult to find an optimal solution to this problem. Therefore, in this paper, we propose the use of meta-heuristics to solve the SFC placement problem in cellular networks. We consider a triathlon competition leading to different mobility patterns. We collected real data about the competitors to simulate their movements through the scenario as well as the measured signal quality of the network. We formulate the SFC placement problem as a multi-objective problem where we try to minimize the placement cost and the total SFC delay. To solve the problem, we propose the use of two algorithms, NSGA-II and GDE3, which compare two different greedy approaches that prioritize the different optimization metrics considered in this work. Our results show that the meta-heuristics provide better results for each of the metrics. For all competition stages, GDE3 presented a slightly lower placement costs than NSGA-II, while NSGA-II had a lower delay in some scenarios.

Low birth weight (LBW) is a significant global health concern, affecting millions of newborns and linked to increased risks of neonatal mortality, delayed development, and chronic diseases in adulthood. This work focuses on a subset of LBW called very low birth weight (VBLW), which exacerbates these risks. Utilizing data from the Sistema de Informações sobre Nascidos Vivos (SINASC), from the state of Pernambuco, Brazil, this work applies data analytics techniques to address issues found in the data set and then train machine learning models to predict VBLW. By integrating feature selection (FS) with expert healthcare insights, data balancing methods, and hyperparameter optimization, the research aims to enhance model performance. Additionally, data slicing is employed to identify and discuss (possible) model biases. In the data slicing analyse, particularly with respect to racial disparities in training data, revealed brown race a predominant group with an f1-score over 68

This study investigates the effectiveness of machine learning models in predicting fetal death and identifying significant predictive factors. The study utilized a dataset from the Programa Mãe Coruja Pernambucana (PMCP) that includes socio-demographic, prenatal, maternal, and family health history data. The data underwent pre-processing and was explored using four tree-based machine learning models, each of which was evaluated based on their performance and feature importance. The attributes that significantly impacted the learning process were the first prenatal week, maternal age, and months between pregnancies. The application of predictive models for fetal deaths in this context can enhance the ability to detect such occurrences thus representing a pivotal support tool for the PMCP to identify mothers with high risk of adverse outcomes and promote targeted interventions of monitoring during pregnancy, and ultimately increase the likelihood of positive outcomes for mothers and babies.

Objective Sexually Transmitted Infections (STIs) present significant challenges to global public health, affecting physical and mental well-being and straining healthcare systems and economies. This study aims to enhance the predictive performance of models for congenital syphilis prediction by incorporating additional information obtained during gestational follow-up. Building upon the work of Teixeira et al. [[1][1]], which utilizes clinical and sociodemographic data, our model was enriched with results from venereal disease research laboratory (VDRL) and rapid tests for congenital syphilis conducted on pregnant women. Method The dataset utilized in this study comprised 47,604 records spanning the period from 2013 to 2022, with 27 attributes collected from pregnant women enrolled in the Mãe Coruja Pernambucana Program in Pernambuco, Brazil. Among these attributes, we included clinical and sociodemographic factors, as well as results from venereal disease research laboratory (VDRL) and rapid tests for congenital syphilis. Results Our proposed model surpassed Teixeira’s models exhibiting higher specificity (94.74%) and a slight increase in sensitivity (70.37%). Conclusions Our study highlights the value of incorporating additional information from VDRL and rapid tests into models for predicting congenital syphilis. The combined approach involving both clinical, sociodemographic, and test result data enhances the accuracy of predictions thereby facilitating better informed healthcare decisions at different stages of pregnancy. This approach also holds significant potential in combating and managing congenital syphilis by providing assistance to health system decision makers and public policymakers. As a result, it can ultimately enhance the overall outcomes of maternal and child health and contribute to disease control. ### Competing Interest Statement The authors have declared no competing interest. ### Funding Statement The author(s) received no specific funding for this work. ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: We declare that the research has been approved by the Brazilian Human Research Ethics Board (Comitê de Ética em Pesquisa [CEP]) under number 12438019.2.0000.5208 and all methods were performed in accordance with the Brazilian regulations that do not require consent for studies using unidentified data from the Brazilian data health systems. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes All data are available from the Mendeley Data database (accession number(s) doi:10.17632/3zkcvybvkz.2). <https://www.doi.org/10.17632/3zkcvybvkz.2> [1]: #ref-1

Syphilis can be transmitted congenitally and may cause serious consequences forthe child if not treated. The Programa Mãe Coruja Pernambucana (PMCP) is a brazilian public health program that helps pregnant women and saves data that can be used for prediction of potential congenital syphilis cases, through machine learning models. Only one work was found that predicts congenital syphilis through machine learning. This research uses a different methodology that evaluates undersampling in prediction. Random Undersampling, UnderSampling Based on Clustering (SBC) and NearMiss were used. The data was preprocessed and undersampling applied, generating different balanced datasets to train and test different machine learning models and different metrics for evaluation. Undersampling discarded data evaluation and analysis of distribution of the best attributes were applied to evaluate undersampling in best models. In models results, NearMiss trained models had high metrics, and very low in the discarded data. SBC models had smaller metrics, and in the discarded data went lower. Random Undersampling models had the lowest metrics, however in the discarded data showed similar results. The distribution of best attributes of NearMiss models were not similar to the original, contrary to Random Undersampling and SBC. NearMiss models had best results in the models, through this work evaluation showed that they cannot generalize the PMCP data and have not a representative distribution of the original data. Random Undersampling models had the lowest metrics but showed consistency through the evaluations, and thus are recommended for the congenital syphilis prediction.

Malaria, a life-threatening disease, is preventable and curable. However, the World Health Organization reported alarming statistics for the year 2021, estimating 247 million cases of malaria worldwide, resulting in 619,000 deaths. In Brazil, cases of malaria are concentrated in the region of the Amazon rainforest, known as the Legal Amazon. In response to calls for more research to adapt malaria mitigation and eradication strategies to local conditions, this article presents a dataset based on data collected from the Malaria Epidemiological Surveillance Information System (SIVEP-Malaria) for the period from January 2003 to December 2022 on patients diagnosed with malaria. The pre-processed data set is composed of 1,251,309 records of confirmed malaria cases and included four attributes (Date, Municipality, Test results, and Notifications), attributes considered essential for the development and application of series modeling techniques temporal. By making this dataset available, it is intended to provide researchers with the necessary tools to develop effective strategies for mitigating and eradicating malaria in the region.

The Zika virus (ZIKV) has emerged as a significant global health concern, particularly for pregnant women, given the potential complications it poses to the fetus. To effectively combat the disease, geospatial analysis of Zika cases has become increasingly important. By examining the incidence and distribution of Zika cases geographically, valuable insights can be gained, and high-risk areas can be identified. These findings are essential for formulating effective disease control measures. This article introduces a comprehensive Zika dataset based on data sourced from the Brazilian Notifiable Diseases Information System, encompassing the years 2016 to 2021. The dataset enables visualization and analysis of epidemiological information including case numbers, geographic distribution, spatial and temporal patterns, as well as common symptoms and complications associated with Zika infection.