Estimation and probabilistic projection of age- and sex-specific mortality rates across Brazilian municipalities between 2010 and 2030.

BACKGROUND: Mortality rate estimation in small areas can be difficult due the low number of events/exposure (i.e. stochastic error). If the death records are not completed, it adds a systematic uncertainty on the mortality estimates. Previous studies in Brazil have combined demographic and statistical methods to partially overcome these issues. We estimated age- and sex-specific mortality rates for all 5,565 Brazilian municipalities in 2010 and forecasted probabilistic mortality rates and life expectancy between 2010 and 2030. METHODS: We used a combination of the Tool for Projecting Age-Specific Rates Using Linear Splines (TOPALS), Bayesian Model, Spatial Smoothing Model and an ad-hoc procedure to estimate age- and sex-specific mortality rates for all Brazilian municipalities for 2010. Then we adapted the Lee-Carter model to forecast mortality rates by age and sex in all municipalities between 2010 and 2030. RESULTS: The adjusted sex- and age-specific mortality rates for all Brazilian municipalities in 2010 reveal a distinct regional pattern, showcasing a decrease in life expectancy in less socioeconomically developed municipalities when compared to estimates without adjustments. The forecasted mortality rates indicate varying regional improvements, leading to a convergence in life expectancy at birth among small areas in Brazil. Consequently, a reduction in the variability of age at death across Brazil's municipalities was observed, with a persistent sex differential. CONCLUSION: Mortality rates at a small-area level were successfully estimated and forecasted, with associated uncertainty estimates also generated for future life tables. Our approach could be applied across countries with data quality issues to improve public policy planning.

A comprehensive dataset integrating household energy consumption and weather conditions in a north-eastern Mexican urban city.

The prediction of domestic electricity consumption is relevant because it helps to plan energy production, among many other benefits. In this work a dataset was collected from one house in an urban city of north-east of Mexico. An ad-hoc acquisition system was implemented to collect the data using a smart meter and the open weather API. The data was collected every minute over a period of 14 months since November 5, 2022, to January 5, 2024. The dataset contains 605,260 samples of 19 variables related with energy consumption and weather data. This dataset is specifically tailored for predicting domestic energy consumption and understanding consumption behaviours, filling a void in the existing literature where such datasets for Mexico are scarce. Moreover, the multivariate nature of the dataset allows researchers to investigate and propose new techniques for forecasting or pattern classification using multivariate data collected in a real scenario.

A New Auto-Regressive Multi-Variable Modified Auto-Encoder for Multivariate Time-Series Prediction: A Case Study with Application to COVID-19 Pandemics.

The SARS-CoV-2 global pandemic prompted governments, institutions, and researchers to investigate its impact, developing strategies based on general indicators to make the most precise predictions possible. Approaches based on epidemiological models were used but the outcomes demonstrated forecasting with uncertainty due to insufficient or missing data. Besides the lack of data, machine-learning models including random forest, support vector regression, LSTM, Auto-encoders, and traditional time-series models such as Prophet and ARIMA were employed in the task, achieving remarkable results with limited effectiveness. Some of these methodologies have precision constraints in dealing with multi-variable inputs, which are important for problems like pandemics that require short and long-term forecasting. Given the under-supply in this scenario, we propose a novel approach for time-series prediction based on stacking auto-encoder structures using three variations of the same model for the training step and weight adjustment to evaluate its forecasting performance. We conducted comparison experiments with previously published data on COVID-19 cases, deaths, temperature, humidity, and air quality index (AQI) in São Paulo City, Brazil. Additionally, we used the percentage of COVID-19 cases from the top ten affected countries worldwide until May 4th, 2020. The results show 80.7% and 10.3% decrease in RMSE to entire and test data over the distribution of 50 trial-trained models, respectively, compared to the first experiment comparison. Also, model type#3 achieved 4th better overall ranking performance, overcoming the NBEATS, Prophet, and Glounts time-series models in the second experiment comparison. This model shows promising forecast capacity and versatility across different input dataset lengths, making it a prominent forecasting model for time-series tasks.

Application of the JULES-crop model and agrometeorological indicators for forecasting off-season maize yield in Brazil.

Zea mays L is a crucial crop for Brazil, ranking second in terms of production and sixth in terms of exports. In Brazil, the second season, or off-season, accounts for 80 % of the overall maize output, which primarily occurs after the soybean main season. A maize yield forecast model for the off-season was developed and implemented throughout Brazilian territory due to its importance to the country's economy and food security. The model was built using multiple linear regressions that connected outputs simulated from a land surface model used in large-scale analysis for agriculture (JULES-crop), to agrometeorological indicators. The application of the developed model occurred every 10 days from the sowing until the maturation. A comparison of the forecasting model was verified with the official off-season maize yields for the years 2003-2016. Agrometeorological indicators during the reproductive phase accounted for 60 % of the interannual variability in maize production. When outputs simulated by JULES-crop were included, the forecasting model achieved Nash-Sutcliffe modeling efficiency (EF) of 0.77 in the maturation and EF = 0.72 in the filling-grain stage, suggesting that this approach can generate useful predictions for final maize yield beginning on the 80th day of the cycle. Outputs of JULES crop enhanced modeling performance during the vegetative stage, reducing the standard deviation error in prediction from 0.59 to 0.49 Mg ha-1.

Diseño de una escala predictiva de mortalidad en pacientes con enfermedad renal crónica

Introducción: La predicción de mortalidad en pacientes con enfermedad renal crónica, mediante escalas o índices pronósticos presenta limitaciones reales. Objetivo: Diseñar una escala predictiva de mortalidad en pacientes con enfermedad renal crónica. Métodos: Se realizó un estudio observacional, analítico, longitudinal prospectivo en 169 pacientes con enfermedad renal crónica desde el 1 de enero de 2022 al 31 de diciembre de 2022. La investigación se desarrolló en 2 etapas: durante los primeros 6 meses del año se analizaron las variables de estudio para el diseño de la escala predictiva. En los próximos 6 meses, los pacientes fueron seguidos para identificar la ocurrencia o no de la variable dependiente mortalidad. Se determinó la capacidad discriminatoria de la escala predictiva y se evaluaron curvas de supervivencia. Resultados: Las variables que conformaron la escala predictiva fueron edad > 65 años, enfermedad cardiovascular, albúmina 390 mmol/L. El poder discriminatorio para predecir mortalidad fue bueno, índice C: 0,856 (IC 95 %: 0,783-0,929; p< 0,001). Los pacientes con valores menores a 4 puntos presentaron media de supervivencia de 149,438 ± 7,296 días. En cambio, los que tenían valores superiores presentaron media de supervivencia de 93,128 ± 8,545 días. Conclusiones: La escala predictiva contribuyó a la estratificación del riesgo de mortalidad de los pacientes. Las variables incluidas son de fácil determinación e interpretación por lo que es un modelo útil en la toma de decisiones médicas en el ámbito clínico actual.

Introduction: The prediction of mortality in patients with chronic kidney disease using scales or prognostic indices has real limitations. Objective: Design a mortality predictive scale in patients with chronic kidney disease. Methods: A prospective observational, analytical, longitudinal study was carried out in 169 patients with chronic kidney disease from January 1, 2022 to December 31, 2022. The research was developed in 2 stages: during the first 6 months of the year, the variables were analyzed for the design of the predictive scale. In the next 6 months, patients were followed to identify the occurrence or not of the dependent variable mortality. The discriminatory capacity of the predictive scale was determined and survival curves were evaluated. Results: The variables that made up the predictive tool were age > 65 years, cardiovascular disease, albumin 390 mmol/L. The discriminatory power to predict mortality was good, C index: 0.856 (95% CI: 0.783-0.929; p< 0.001). Patients with values less than 4 points had a mean survival of 149.438 ± 7.296 days. In contrast, those with higher values presented a mean survival of 93.128 ± 8.545 days. Conclusions: The scale contributed to the stratification of the mortality risk of the patients. The variables included are easy to determine and interpret, making it a useful model for medical decision making in the current clinical setting.

Stochastic transmission in epidemiological models.

Recent empirical evidence suggests that the transmission coefficient in susceptible-exposed-infected-removed-like (SEIR-like) models evolves with time, presenting random patterns, and some stylized facts, such as mean-reversion and jumps. To address such observations we propose the use of jump-diffusion stochastic processes to parameterize the transmission coefficient in an SEIR-like model that accounts for death and time-dependent parameters. We provide a detailed theoretical analysis of the proposed model proving the existence and uniqueness of solutions as well as studying its asymptotic behavior. We also compare the proposed model with some variations possibly including jumps. The forecast performance of the considered models, using reported COVID-19 infections from New York City, is then tested in different scenarios. Despite the simplicity of the epidemiological model, by considering stochastic transmission, the forecasted scenarios were fairly accurate.

Análisis de los factores pronósticos de mortalidad en los pacientes con enfermedad renal crónica en hemodiálisis / Analysis of Mortality Prognostic Factors in Patients with chronic Kidney Disease on Hemodialysis

Introducción: La enfermedad renal crónica es una de las principales causas de mortalidad en todo el mundo. La estratificación del riesgo a través del análisis de factores pronósticos podría generar un cambio de paradigma. Objetivo: Analizar los factores pronósticos de mortalidad en los pacientes con enfermedad renal crónica en hemodiálisis. Métodos: Se realizó un estudio no experimental, longitudinal de cohorte retrospectivo en los pacientes con enfermedad renal crónica en hemodiálisis en el Hospital General Docente: Dr. Ernesto Guevara de la Serna durante el período del 1 de enero de 2017 al 31 de diciembre de 2021. En general, se analizaron los factores pronósticos de mortalidad mediante el análisis multivariado de regresión logística binaria y se determinó el porcentaje correcto de clasificación del modelo de regresión. Resultados: Se analizaron como variables pronosticas de mortalidad la enfermedad cardiovascular [B = 3,831; p = 0,000; Exp (B) = 46,118], Albúmina 17 mmol/L [B = 1,326; p = 0,027; Exp (B) = 3,767], glucemia < 4 mmol/L [B = 1,600; p = 0,015; Exp (B) = 4,955] y ganancia de peso interdialítica excesiva [B = 2,243; p = 0,001; Exp (B) = 9,420]. El porcentaje global de clasificación del modelo de regresión logística binaria fue de 89,5 por ciento. Conclusiones: Se analizó el modelo predictivo de regresión logística que presentó una buena precisión con los factores de pronósticos asociados a la mortalidad en los pacientes en hemodiálisis(AU)

Introduction: Chronic kidney disease is one of the main causes of mortality worldwide. Risk stratification through the analysis of prognostic factors could generate a paradigm shift. Objective: To analyze the prognostic factors of mortality in patients with chronic kidney disease on hemodialysis. Methods: A non-experimental, longitudinal retrospective cohort study was carried out on patients with chronic kidney disease on hemodialysis at Dr. Ernesto Guevara de la Serna General Teaching Hospital from January 2017 to December 31, 2021. The prognostic factors of mortality were analyzed using multivariate binary logistic regression analysis and the correct percentage of classification of the regression model was determined. Results: Prognostic variables of mortality were analyzed, such as cardiovascular disease [B = 3.831; p = 0.000; Exp (B) = 46.118], albumin 17 mmol/L [B = 1.326; p = 0.027; Exp (B) = 3.767], blood glucose < 4 mmol/L [B = 1.600; p = 0.015; Exp (B) = 4.955] and excessive interdialytic weight gain [B = 2.243; p = 0.001; Exp(B) = 9.420]. The overall classification percentage of the binary logistic regression model was 89.5percent. Conclusions: The logistic regression predictive model was analyzed, which showed good precision with the prognostic factors associated with mortality in hemodialysis patients(AU)

COVID-19 Epidemic Forecast in Brazil.

This study advocates a novel spatio-temporal method for accurate prediction of COVID-19 epidemic occurrence probability at any time in any Brazil state of interest, and raw clinical observational data have been used. This article describes a novel bio-system reliability approach, particularly suitable for multi-regional environmental and health systems, observed over a sufficient time period, resulting in robust long-term forecast of the virus outbreak probability. COVID-19 daily numbers of recorded patients in all affected Brazil states were taken into account. This work aimed to benchmark novel state-of-the-art methods, making it possible to analyse dynamically observed patient numbers while taking into account relevant regional mapping. Advocated approach may help to monitor and predict possible future epidemic outbreaks within a large variety of multi-regional biological systems. Suggested methodology may be used in various modern public health applications, efficiently using their clinical survey data.

Participatory Surveillance for COVID-19 Trend Detection in Brazil: Cross-sectional Study.

BACKGROUND: The ongoing COVID-19 pandemic has emphasized the necessity of a well-functioning surveillance system to detect and mitigate disease outbreaks. Traditional surveillance (TS) usually relies on health care providers and generally suffers from reporting lags that prevent immediate response plans. Participatory surveillance (PS), an innovative digital approach whereby individuals voluntarily monitor and report on their own health status via web-based surveys, has emerged in the past decade to complement traditional data collection approaches. OBJECTIVE: This study compared novel PS data on COVID-19 infection rates across 9 Brazilian cities with official TS data to examine the opportunities and challenges of using PS data, and the potential advantages of combining the 2 approaches. METHODS: The TS data for Brazil are publicly accessible on GitHub. The PS data were collected through the Brazil Sem Corona platform, a Colab platform. To gather information on an individual's health status, each participant was asked to fill out a daily questionnaire on symptoms and exposure in the Colab app. RESULTS: We found that high participation rates are key for PS data to adequately mirror TS infection rates. Where participation was high, we documented a significant trend correlation between lagged PS data and TS infection rates, suggesting that PS data could be used for early detection. In our data, forecasting models integrating both approaches increased accuracy up to 3% relative to a 14-day forecast model based exclusively on TS data. Furthermore, we showed that PS data captured a population that significantly differed from a traditional observation. CONCLUSIONS: In the traditional system, the new recorded COVID-19 cases per day are aggregated based on positive laboratory-confirmed tests. In contrast, PS data show a significant share of reports categorized as potential COVID-19 cases that are not laboratory confirmed. Quantifying the economic value of PS system implementation remains difficult. However, scarce public funds and persisting constraints to the TS system provide motivation for a PS system, making it an important avenue for future research. The decision to set up a PS system requires careful evaluation of its expected benefits, relative to the costs of setting up platforms and incentivizing engagement to increase both coverage and consistent reporting over time. The ability to compute such economic tradeoffs might be key to have PS become a more integral part of policy toolkits moving forward. These results corroborate previous studies when it comes to the benefits of an integrated and comprehensive surveillance system, and shed light on its limitations and on the need for additional research to improve future implementations of PS platforms.

COVID-19 Patterns in Araraquara, Brazil: A Multimodal Analysis.

The epidemiology of COVID-19 presented major shifts during the pandemic period. Factors such as the most common symptoms and severity of infection, the circulation of different variants, the preparedness of health services, and control efforts based on pharmaceutical and non-pharmaceutical interventions played important roles in the disease incidence. The constant evolution and changes require the continuous mapping and assessing of epidemiological features based on time-series forecasting. Nonetheless, it is necessary to identify the events, patterns, and actions that were potential factors that affected daily COVID-19 cases. In this work, we analyzed several databases, including information on social mobility, epidemiological reports, and mass population testing, to identify patterns of reported cases and events that may indicate changes in COVID-19 behavior in the city of Araraquara, Brazil. In our analysis, we used a mathematical approach with the fast Fourier transform (FFT) to map possible events and machine learning model approaches such as Seasonal Auto-regressive Integrated Moving Average (ARIMA) and neural networks (NNs) for data interpretation and temporal prospecting. Our results showed a root-mean-square error (RMSE) of about 5 (more precisely, a 4.55 error over 71 cases for 20 March 2021 and a 5.57 error over 106 cases for 3 June 2021). These results demonstrated that FFT is a useful tool for supporting the development of the best prevention and control measures for COVID-19.

How Does Neural Network Model Capacity Affect Photovoltaic Power Prediction? A Study Case.

The use of models capable of forecasting the production of photovoltaic (PV) energy is essential to guarantee the best possible integration of this energy source into traditional distribution grids. Long Short-Term Memory networks (LSTMs) are commonly used for this purpose, but their use may not be the better option due to their great computational complexity and slower inference and training time. Thus, in this work, we seek to evaluate the use of neural networks MLPs (Multilayer Perceptron), Recurrent Neural Networks (RNNs), and LSTMs, for the forecast of 5 min of photovoltaic energy production. Each iteration of the predictions uses the last 120 min of data collected from the PV system (power, irradiation, and PV cell temperature), measured from 2019 to mid-2022 in Maceió (Brazil). In addition, Bayesian hyperparameters optimization was used to obtain the best of each model and compare them on an equal footing. Results showed that the MLP performs satisfactorily, requiring much less time to train and forecast, indicating that they can be a better option when dealing with a very short-term forecast in specific contexts, for example, in systems with little computational resources.

Application of AI for Short-Term PV Generation Forecast.

The efficient use of the photovoltaic power requires a good estimation of the PV generation. That is why the use of good techniques for forecast is necessary. In this research paper, Long Short-Term Memory, Bidirectional Long Short-Term Memory and the Temporal convolutional network are studied in depth to forecast the photovoltaic power, voltage and efficiency of a 1320 Wp amorphous plant installed in the Technology Support Centre in the University Rey Juan Carlos, Madrid (Spain). The accuracy of these techniques are compared using experimental data along one year, applying 1 timestep or 15 min and 96 step times or 24 h, showing that TCN exhibits outstanding performance, compared with the two other techniques. For instance, it presents better results in all forecast variables and both forecast horizons, achieving an overall Mean Squared Error (MSE) of 0.0024 for 15 min forecasts and 0.0058 for 24 h forecasts. In addition, the sensitivity analyses for the TCN technique is performed and shows that the accuracy is reduced as the forecast horizon increases and that the 6 months of dataset is sufficient to obtain an adequate result with an MSE value of 0.0080 and a coefficient of determination of 0.90 in the worst scenarios (24 h of forecast).

Frequency of SARS-CoV-2 variants identified by real-time PCR in the AUNA healthcare network, Peru.

Introduction: In Peru, on 11 February 2023, the Ministry of Health registered 4 million patients infected with COVID-19 and around 219,260 deaths. In 2020, the SARS-CoV-2 virus was acquiring mutations that impacted the properties of transmissibility, infectivity, and immune evasion, leading to new lineages. In the present study, the frequency of COVID-19 variants was determined during 2021 and 2022 in patients treated in the AUNA healthcare network. Methods: The methodology used to detect mutations and identify variants was the Allplex™ SARS-CoV-2 Variants Assay I, II, and VII kit RT-PCR. The frequency of variants was presented by epidemiological weeks. Results: In total, 544 positive samples were evaluated, where the Delta, Omicron, and Gamma variants were identified. The Delta variant was found in 242 (44.5%) patients between epidemiological weeks 39 and 52 in 2021. In the case of Gamma, it was observed in 8 (1.5%) patients at weeks 39, 41, 43, 45, and 46 of 2021. The Omicron variant was the most frequent with 289 (53.1%) patients during weeks 49 to 52 of 2021 and 1 to 22 of 2022. During weeks 1 through 22 of 2022, it was possible to discriminate between BA. 1 (n = 32) and BA.2 (n = 82). Conclusion: The rapid identification of COVID-19 variants through the RT-PCR methodology contributes to timely epidemiological surveillance, as well as appropriate patient management.

Price trends of Agave Mezcalero in Mexico using multiple linear regression models / Estimativa de tendências de preços de Agave Mezcalero no México usando modelos de regressão linear múltipla

ABSTRACT: This study developed a multiple linear regression model to estimate the Average rural prices (ARP) in Mexico with information taken from the period 1999-2018. The variables used to generate this model were the supply and demand as represented by planted area, yield, exports and the ARP of Agave Tequilero and Mezcalero. The analysis was carried out through the multiple linear regression model (MLRM) with the least squares method and using the statistical package R. The following variables were identified as having a significant influence on the determination of the ARP: the yield of Agave Mezcalero (YAM), the ARP of Agave Tequilero and the new planted area of Agave Tequilero (NPAATt-6) with an adjustment of 6 periods. Overall, three models were generated: model 2 was considered the most appropriate because it allows carrying out future forecasts with the new planted area with Agave Tequilero with 2 independent variables. YAM and NPAATt-6 were useful in predicting 65.5% of the annual variations in the ARP and helped recognize the negative trend of the Agave price from 2020 to 2024. Therefore, the use of the MLRM to estimate the Agave ARP can be a useful tool in predicting the performance of this crop.

RESUMO: O objetivo deste estudo é desenvolver um modelo de regressão linear múltipla para estimar o Preços médios rurais (PRM) no México com informações retiradas do período 1999-2018. As variáveis ​​utilizadas para gerar este modelo foram a oferta e a demanda representadas pela área plantada, produtividade, exportações e o PRM da Agave Tequilero e Mezcalero. A análise foi realizada através do modelo de regressão linear múltipla (MRLM) com o método dos mínimos quadrados e utilizando o pacote estatístico R. As seguintes variáveis ​​foram identificadas como tendo influência significativa na determinação do PRM: o rendimento da Agave Mezcalero (RAM), o PMR da Agave Tequilero e a nova área plantada da Agave Tequilero (NPAATt-6) com um ajuste de 6 períodos. Ao todo, foram gerados três modelos: o modelo 2 foi considerado o mais adequado porque permite fazer previsões futuras com a nova área plantada com Agave Tequilero com dois variáveis ​​independentes. RAM e NPAATt-6 foram úteis na previsão de 65,5% das variações anuais no ARP e ajudaram a reconhecer a tendência negativa do preço da Agave de 2020 a 2024. Portanto, o uso do MRLM para estimar o PMR da Agave pode ser uma ferramenta útil na previsão do desempenho desta cultura.

Price trends of Agave Mezcalero in Mexico using multiple linear regression models / Estimativa de tendências de preços de Agave Mezcalero no México usando modelos de regressão linear múltipla

This study developed a multiple linear regression model to estimate the Average rural prices (ARP) in Mexico with information taken from the period 1999-2018. The variables used to generate this model were the supply and demand as represented by planted area, yield, exports and the ARP of Agave Tequilero and Mezcalero. The analysis was carried out through the multiple linear regression model (MLRM) with the least squares method and using the statistical package R. The following variables were identified as having a significant influence on the determination of the ARP: the yield of Agave Mezcalero (YAM), the ARP of Agave Tequilero and the new planted area of Agave Tequilero (NPAATt-6) with an adjustment of 6 periods. Overall, three models were generated: model 2 was considered the most appropriate because it allows carrying out future forecasts with the new planted area with Agave Tequilero with 2 independent variables. YAM and NPAATt-6 were useful in predicting 65.5% of the annual variations in the ARP and helped recognize the negative trend of the Agave price from 2020 to 2024. Therefore, the use of the MLRM to estimate the Agave ARP can be a useful tool in predicting the performance of this crop.

O objetivo deste estudo é desenvolver um modelo de regressão linear múltipla para estimar o Preços médios rurais (PRM) no México com informações retiradas do período 1999-2018. As variáveis ​​utilizadas para gerar este modelo foram a oferta e a demanda representadas pela área plantada, produtividade, exportações e o PRM da Agave Tequilero e Mezcalero. A análise foi realizada através do modelo de regressão linear múltipla (MRLM) com o método dos mínimos quadrados e utilizando o pacote estatístico R. As seguintes variáveis ​​foram identificadas como tendo influência significativa na determinação do PRM: o rendimento da Agave Mezcalero (RAM), o PMR da Agave Tequilero e a nova área plantada da Agave Tequilero (NPAATt-6) com um ajuste de 6 períodos. Ao todo, foram gerados três modelos: o modelo 2 foi considerado o mais adequado porque permite fazer previsões futuras com a nova área plantada com Agave Tequilero com dois variáveis ​​independentes. RAM e NPAATt-6 foram úteis na previsão de 65,5% das variações anuais no ARP e ajudaram a reconhecer a tendência negativa do preço da Agave de 2020 a 2024. Portanto, o uso do MRLM para estimar o PMR da Agave pode ser uma ferramenta útil na previsão do desempenho desta cultura.

Temporal and Spatiotemporal Arboviruses Forecasting by Machine Learning: A Systematic Review.

Arboviruses are a group of diseases that are transmitted by an arthropod vector. Since they are part of the Neglected Tropical Diseases that pose several public health challenges for countries around the world. The arboviruses' dynamics are governed by a combination of climatic, environmental, and human mobility factors. Arboviruses prediction models can be a support tool for decision-making by public health agents. In this study, we propose a systematic literature review to identify arboviruses prediction models, as well as models for their transmitter vector dynamics. To carry out this review, we searched reputable scientific bases such as IEE Xplore, PubMed, Science Direct, Springer Link, and Scopus. We search for studies published between the years 2015 and 2020, using a search string. A total of 429 articles were returned, however, after filtering by exclusion and inclusion criteria, 139 were included. Through this systematic review, it was possible to identify the challenges present in the construction of arboviruses prediction models, as well as the existing gap in the construction of spatiotemporal models.

Predicting Brazilian Court Decisions.

Predicting case outcomes is useful for legal professionals to understand case law, file a lawsuit, raise a defense, or lodge appeals, for instance. However, it is very hard to predict legal decisions since this requires extracting valuable information from myriads of cases and other documents. Moreover, legal system complexity along with a huge volume of litigation make this problem even harder. This paper introduces an approach to predicting Brazilian court decisions, including whether they will be unanimous. Our methodology uses various machine learning algorithms, including classifiers and state-of-the-art Deep Learning models. We developed a working prototype whose F1-score performance is ~80.2% by using 4,043 cases from a Brazilian court. To our knowledge, this is the first study to present methods for predicting Brazilian court decision outcomes.

Future health spending forecast in leading emerging BRICS markets in 2030: health policy implications.

BACKGROUND: The leading emerging markets of Brazil, Russia, India, China and South Africa (BRICS) are increasingly shaping the landscape of the global health sector demand and supply for medical goods and services. BRICS' share of global health spending and future projections will play a prominent role during the 2020s. The purpose of the current research was to examine the decades-long underlying historical trends in BRICS countries' health spending and explore these data as the grounds for reliable forecasting of their health expenditures up to 2030. METHODS: BRICS' health spending data spanning 1995-2017 were extracted from the Institute for Health Metrics and Evaluation (IHME) Financing Global Health 2019 database. Total health expenditure, government, prepaid private and out-of-pocket spending per capita and gross domestic product (GDP) share of total health spending were forecasted for 2018-2030. Autoregressive integrated moving average (ARIMA) models were used to obtain future projections based on time series analysis. RESULTS: Per capita health spending in 2030 is projected to be as follows: Brazil, $1767 (95% prediction interval [PI] 1615, 1977); Russia, $1933 (95% PI 1549, 2317); India, $468 (95% PI 400.4, 535); China, $1707 (95% PI 1079, 2334); South Africa, $1379 (95% PI 755, 2004). Health spending as a percentage of GDP in 2030 is projected as follows: Brazil, 8.4% (95% PI 7.5, 9.4); Russia, 5.2% (95% PI 4.5, 5.9); India, 3.5% (95% PI 2.9%, 4.1%); China, 5.9% (95% PI 4.9, 7.0); South Africa, 10.4% (95% PI 5.5, 15.3). CONCLUSIONS: All BRICS countries show a long-term trend towards increasing their per capita spending in terms of purchasing power parity (PPP). India and Russia are highly likely to maintain stable total health spending as a percentage of GDP until 2030. China, as a major driver of global economic growth, will be able to significantly expand its investment in the health sector across an array of indicators. Brazil is the only large nation whose health expenditure as a percentage of GDP is about to contract substantially during the third decade of the twenty-first century. The steepest curve of increased per capita spending until 2030 seems to be attributable to India, while Russia should achieve the highest values in absolute terms. Health policy implications of long-term trends in health spending indicate the need for health technology assessment dissemination among the BRICS ministries of health and national health insurance funds. Matters of cost-effective allocation of limited resources will remain a core challenge in 2030 as well.

Cervical cancer mortality among young women in Latin America and the Caribbean: trend analysis from 1997 to 2030.

BACKGROUND: Cervical cancer continues to show a high burden among young women worldwide, particularly in low- and middle-income countries. Limited data is available describing cervical cancer mortality among young women in Latin America and the Caribbean (LAC). The purpose of this study was to examine the mortality trends of cervical cancer among young women in LAC and predict mortality rates to 2030. METHODS: Deaths from cervical cancer were obtained from the World Health Organization mortality database. Age-standardized mortality rates per 100,000 women-years were estimated in women aged 20-44 years using the world standard population for 16 countries (and territories) in LAC from 1997 to 2017. We estimated the average mortality rates for the last 4 years (2014-2017). Joinpoint regression models were used to identify significant changes in mortality trends. Nordpred method was used for the prediction of the mortality rates to 2030. RESULTS: Between 2014 and 2017, Paraguay and Venezuela had the highest mortality rates of cervical cancer, whereas Puerto Rico had the lowest rates. Overall, most of the LAC countries showed downward trends of cervical cancer mortality over the entire period. Significant decreases were observed in Chile (Average annual percent change [AAPC]: - 2.4%), Colombia (AAPC: - 2.0%), Cuba (AAPC: - 3.6%), El Salvador (AAPC: - 3.1%), Mexico (AAPC: - 3.9%), Nicaragua (AAPC: - 1.7%), Panama (AAPC: - 1.7%), and Peru (AAPC: - 2.2%). In contrast, Brazil (AAPC: + 0.8%) and Paraguay (AAPC: + 3.7%) showed significant upward trends. By 2030, mortality rates are not predicted to further decrease in some LAC countries, including Argentina, Paraguay, and Venezuela. CONCLUSIONS: Mortality trends of cervical cancer among young women have large variability in LAC countries. Cervical cancer screening programs have a high priority for the region. Primary and secondary prevention in the community are necessary to accelerate a reduction of cervical cancer mortality by 2030.

Multivariate data driven prediction of COVID-19 dynamics: Towards new results with temperature, humidity and air quality data.

Since the start of the COVID-19 pandemic many studies investigated the correlation between climate variables such as air quality, humidity and temperature and the lethality of COVID-19 around the world. In this work we investigate the use of climate variables, as additional features to train a data-driven multivariate forecast model to predict the short-term expected number of COVID-19 deaths in Brazilian states and major cities. The main idea is that by adding these climate features as inputs to the training of data-driven models, the predictive performance improves when compared to equivalent single input models. We use a Stacked LSTM as the network architecture for both the multivariate and univariate model. We compare both approaches by training forecast models for the COVID-19 deaths time series of the city of São Paulo. In addition, we present a previous analysis based on grouping K-means on AQI curves. The results produced will allow achieving the application of transfer learning, once a locality is eventually added to the task, regressing out using a model based on the cluster of similarities in the AQI curve. The experiments show that the best multivariate model is more skilled than the best standard data-driven univariate model that we could find, using as evaluation metrics the average fitting error, average forecast error, and the profile of the accumulated deaths for the forecast. These results show that by adding more useful features as input to a multivariate approach could further improve the quality of the prediction models.