Leandro Brito Santos
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Leandro Brito Santos has a total of 46 co-authors. Here are the top co-authors by number of shared publications:
- Alexandre Do Nascimento Silva
Universidade do Estado da Bahia, Alagoinhas, Bahia, Brasil
3 shared publications - José Roberto de Araújo Fontoura
Universidade do Estado da Bahia. Alagoinhas, BA, Brasil
3 shared publications - Marcos Batista Figueredo
Universidade do Estado da Bahia, Alagoinhas, Bahia, Brasil
3 shared publications - Roberto Luiz Souza Monteiro
Universidade SENAI CIMATEC, Salvador, Bahia, Brazil
3 shared publications - Thiago Barros Murari
Modelagem Computacional, SENAI CIMATEC, Salvador, Brazil
3 shared publications - Eber da Silva de Santana
Salvador University -UNIFACS Salvador Bahia -PPGCOMP
2 shared publications
Latest Publications
- Open Access
- 13 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. PP, 1
Wavelet-based hybrid model for short-term wind speed forecasting in coastal Brazil
Short-term wind speed forecasting remains challenging due to the inherent nonstationarity and multiscale variability of atmospheric processes. This study proposes a hybrid multiscale forecasting framework combining four-level discrete wavelet transform decomposition with scale-specialized predictive models and an adaptive Self-Error Stabilization Mechanism. Unlike conventional wavelet-machine learning hybrids, which do not explicitly address reconstruction-stage error propagation, the proposed framework introduces a causal, scale-aware attenuation operator that suppresses inconsistent high-frequency predictions before inverse reconstruction. Low-frequency dynamics are modeled using a Long Short-Term Memory network, while medium- and high-frequency components are predicted by dedicated machine learning regressors. Evaluated on a 22-year daily wind speed dataset from the Brazilian coast (INMET, 2001–2023), the LSTM–MLP configuration achieved an RMSE of 0.275 m/s, MAE of 0.188 m/s, sMAPE of 1.99%, R2 of 0.991, and C30 of 99.97% at the one-step-ahead horizon. Structural ablation analysis confirms the importance of explicit reconstruction control, as removing SESM increased RMSE from 0.275 to 1.350 m/s and reduced R2 from 0.991 to 0.784. These results indicate that reconstruction-aware multiscale modeling improves forecasting stability and physical consistency while preserving controlled structural complexity, providing a computationally efficient framework for wind farm operation and grid management. When benchmarked against comparable studies reported in the literature in comparable daily t+1 wind speed forecasting studies, the proposed framework can yield up to a 47% reduction in RMSE.
- 13 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. PP, 1
Open AccessWavelet-based hybrid model for short-term wind speed forecasting in coastal Brazil
Short-term wind speed forecasting remains challenging due to the inherent nonstationarity and multiscale variability of atmospheric processes. This study proposes a hybrid multiscale forecasting framework combining four-level discrete wavelet transform decomposition with scale-specialized predictive models and an adaptive Self-Error Stabilization Mechanism. Unlike conventional wavelet-machine learning hybrids, which do not explicitly address reconstruction-stage error propagation, the proposed framework introduces a causal, scale-aware attenuation operator that suppresses inconsistent high-frequency predictions before inverse reconstruction. Low-frequency dynamics are modeled using a Long Short-Term Memory network, while medium- and high-frequency components are predicted by dedicated machine learning regressors. Evaluated on a 22-year daily wind speed dataset from the Brazilian coast (INMET, 2001–2023), the LSTM–MLP configuration achieved an RMSE of 0.275 m/s, MAE of 0.188 m/s, sMAPE of 1.99%, R2 of 0.991, and C30 of 99.97% at the one-step-ahead horizon. Structural ablation analysis confirms the importance of explicit reconstruction control, as removing SESM increased RMSE from 0.275 to 1.350 m/s and reduced R2 from 0.991 to 0.784. These results indicate that reconstruction-aware multiscale modeling improves forecasting stability and physical consistency while preserving controlled structural complexity, providing a computationally efficient framework for wind farm operation and grid management. When benchmarked against comparable studies reported in the literature in comparable daily t+1 wind speed forecasting studies, the proposed framework can yield up to a 47% reduction in RMSE.
- Open Access
- 27 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 214761-214771
An Approach for Parkinson’s Disease Detection Based on Artificial Intelligence Techniques
Parkinson’s disease (PD) is a progressive neurodegenerative disorder caused by the depletion of dopamine-producing neurons in the substantia nigra, leading to motor dysfunction and alterations in speech patterns. Temporal variations in speech signals have been identified as potential biomarkers of PD, emphasizing the importance of early detection for slowing disease progression and allowing timely therapeutic interventions. This study presents an optimized Convolutional Neural Network (CNN) classifier designed for accurate and efficient PD detection. The model leverages sustained vowel phonations (“ah”) from the Parkinson’s Disease Voice Recording Data Set, which includes samples from 147 patients with PD and 48 healthy controls. Sustained vowels are widely used in clinical protocols and research because they minimize articulatory variability and highlight relevant vocal biomarkers such as jitter, shimmer, and the harmonic-to-noise ratio (HNR). The preprocessing pipeline involved denoising, feature normalization, and Bayesian hyperparameter tuning to optimize performance. The CNN achieved a detection accuracy of 96.85%, along with high precision, sensitivity, and a well-balanced F1 score, demonstrating robust predictive capabilities. These findings validate the model’s reliability in detecting PD-specific voice patterns, even in noisy conditions, underscoring its potential for integration into clinical diagnostic workflows.
- 27 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 214761-214771
Open AccessAn Approach for Parkinson’s Disease Detection Based on Artificial Intelligence Techniques
Parkinson’s disease (PD) is a progressive neurodegenerative disorder caused by the depletion of dopamine-producing neurons in the substantia nigra, leading to motor dysfunction and alterations in speech patterns. Temporal variations in speech signals have been identified as potential biomarkers of PD, emphasizing the importance of early detection for slowing disease progression and allowing timely therapeutic interventions. This study presents an optimized Convolutional Neural Network (CNN) classifier designed for accurate and efficient PD detection. The model leverages sustained vowel phonations (“ah”) from the Parkinson’s Disease Voice Recording Data Set, which includes samples from 147 patients with PD and 48 healthy controls. Sustained vowels are widely used in clinical protocols and research because they minimize articulatory variability and highlight relevant vocal biomarkers such as jitter, shimmer, and the harmonic-to-noise ratio (HNR). The preprocessing pipeline involved denoising, feature normalization, and Bayesian hyperparameter tuning to optimize performance. The CNN achieved a detection accuracy of 96.85%, along with high precision, sensitivity, and a well-balanced F1 score, demonstrating robust predictive capabilities. These findings validate the model’s reliability in detecting PD-specific voice patterns, even in noisy conditions, underscoring its potential for integration into clinical diagnostic workflows.
- Open Access
- 1 January 2025
- Published by Wiley in Applied Computational Intelligence and Soft Computing
- Vol. 2025 (1)
Predictive Modeling of River Water Temperatures in Catu River: A Neural Network‐Based Approach
Predicting water temperature ( T w ) in tropical environments is crucial for ecosystem monitoring and the sustainable management of water resources. Highly accurate and reliable T w forecasts are essential for the ecological management of rivers. This study evaluates the performance of machine learning‐based predictive models in forecasting T w in the Catu River. The models were trained using climatic and hydrological data collected from 2009 to 2016 and validated with real data from 2023. The evaluated models include backpropagation neural network (BPNN), Random Forest, Bidirectional LSTM (BiLSTM), Air2Stream, and NARX, employing nine input variables such as atmospheric pressure, air temperature, and water vapor concentration. The results show that the BiLSTM model achieved the best performance, with a root mean square error (RMSE) of 0.12°C and R 2 = 0.98, followed by BPNN with an RMSE of 0.18°C and R 2 = 0.91, and the Random Forest model, which obtained an NSE of 0.95. These models demonstrated a strong ability to predict T w under both normal and extreme conditions, capturing the thermal dynamics of the Catu River with high precision during events involving minor thermal variations. Conversely, the NARX and Air2Stream models exhibited lower performance, proving more prone to errors under conditions of extreme variability. The findings of this study provide valuable scientific insights for river T w prediction and the protection of aquatic ecosystems, with practical applications in water resource management in tropical regions.
- 1 January 2025
- Published by Wiley in Applied Computational Intelligence and Soft Computing
- Vol. 2025 (1)
Open AccessPredictive Modeling of River Water Temperatures in Catu River: A Neural Network‐Based Approach
Predicting water temperature ( T w ) in tropical environments is crucial for ecosystem monitoring and the sustainable management of water resources. Highly accurate and reliable T w forecasts are essential for the ecological management of rivers. This study evaluates the performance of machine learning‐based predictive models in forecasting T w in the Catu River. The models were trained using climatic and hydrological data collected from 2009 to 2016 and validated with real data from 2023. The evaluated models include backpropagation neural network (BPNN), Random Forest, Bidirectional LSTM (BiLSTM), Air2Stream, and NARX, employing nine input variables such as atmospheric pressure, air temperature, and water vapor concentration. The results show that the BiLSTM model achieved the best performance, with a root mean square error (RMSE) of 0.12°C and R 2 = 0.98, followed by BPNN with an RMSE of 0.18°C and R 2 = 0.91, and the Random Forest model, which obtained an NSE of 0.95. These models demonstrated a strong ability to predict T w under both normal and extreme conditions, capturing the thermal dynamics of the Catu River with high precision during events involving minor thermal variations. Conversely, the NARX and Air2Stream models exhibited lower performance, proving more prone to errors under conditions of extreme variability. The findings of this study provide valuable scientific insights for river T w prediction and the protection of aquatic ecosystems, with practical applications in water resource management in tropical regions.
- Open Access
- 20 November 2023
- Published by Uniao Atlantica de Pesquisadores in Concilium
- Vol. 23 (21) , 347-363
Wind speed prediction model based on DWT and Randon Forest
Wind energy is one of the fastest power generation technologies in the power generation industry and one of the most cost-effective methods of generating electrical power. For system reliability, improving highly appropriate wind speed forecasting methods is desirable. The wavelet transform is a powerful mathematical technique that converts an analyzed signal into a time-frequency representation. This technique helps forecast non-stationary time series. The objective is to evaluate the performance of the "DWT-Random Forest" model in predicting wind speed through a comparative analysis of performance metrics (MSE, RMSE, and MAPE) with similar studies. Our motivation is rooted in the pressing need to improve wind forecasting methods to optimize renewable energy generation. The method involved implementing the model, which presented performance metrics: MSE of 0.0099, RMSE of 0.0996, and MAPE of 0.0779. However, comparative analysis with previous studies reveals that our model demonstrates competitive performance. The main result of this study is the finding that the "DWT-Random Forest" model exhibits a respectable performance in predicting wind speed, although there is room for improvement.
- 20 November 2023
- Published by Uniao Atlantica de Pesquisadores in Concilium
- Vol. 23 (21) , 347-363
Open AccessWind speed prediction model based on DWT and Randon Forest
Wind energy is one of the fastest power generation technologies in the power generation industry and one of the most cost-effective methods of generating electrical power. For system reliability, improving highly appropriate wind speed forecasting methods is desirable. The wavelet transform is a powerful mathematical technique that converts an analyzed signal into a time-frequency representation. This technique helps forecast non-stationary time series. The objective is to evaluate the performance of the "DWT-Random Forest" model in predicting wind speed through a comparative analysis of performance metrics (MSE, RMSE, and MAPE) with similar studies. Our motivation is rooted in the pressing need to improve wind forecasting methods to optimize renewable energy generation. The method involved implementing the model, which presented performance metrics: MSE of 0.0099, RMSE of 0.0996, and MAPE of 0.0779. However, comparative analysis with previous studies reveals that our model demonstrates competitive performance. The main result of this study is the finding that the "DWT-Random Forest" model exhibits a respectable performance in predicting wind speed, although there is room for improvement.
- Open Access
- 15 March 2023
- Published by Universidade Federal da Bahia in Cadernos de Prospecção
- Vol. 16 (2) , 516-529
Portal de Acesso às Informações das Ações das Universidades Federais em Resposta à Pandemia de Covid-19: uma análise do período pandêmico até a transição para uma pós-pandemia
Durante a pandemia, foi necessária uma nova organização das atividades acadêmicas das Universidades Federais para empreender ações internas e externas de enfrentamento à Covid-19. Para compartilhar as ações realizadas, o Ministério da Educação (MEC) desenvolveu dois portais de informação sobre os números da Covid-19 e sobrea as ações desenvolvidas para a educação básica e superior. Um desses portais é o “Painel Coronavírus – Monitoramento da Rede Federal”, que reúne um conjunto de informações referente às ações realizadas pelas instituições de ensino desde o início da pandemia até o momento atual de transição para uma pós-pandemia. Utilizando um método de análise de dados quantitativos e qualitativos, buscou-se discorrer sobre os dados de casos de Covid-19 e seus picos como as ações envolvendo modalidade de ensino, prevenção e apoio à comunidade acadêmica. Como resultado, são apresentados de forma correlacionada alguns painéis do portal e a análise dos dados das ações, permitindo compartilhar as estratégias e a compreensão histórica de quais ações foram utilizadas no enfrentamento da pandemia e na transição de uma pós-pandemia.
- 15 March 2023
- Published by Universidade Federal da Bahia in Cadernos de Prospecção
- Vol. 16 (2) , 516-529
Open AccessPortal de Acesso às Informações das Ações das Universidades Federais em Resposta à Pandemia de Covid-19: uma análise do período pandêmico até a transição para uma pós-pandemia
Durante a pandemia, foi necessária uma nova organização das atividades acadêmicas das Universidades Federais para empreender ações internas e externas de enfrentamento à Covid-19. Para compartilhar as ações realizadas, o Ministério da Educação (MEC) desenvolveu dois portais de informação sobre os números da Covid-19 e sobrea as ações desenvolvidas para a educação básica e superior. Um desses portais é o “Painel Coronavírus – Monitoramento da Rede Federal”, que reúne um conjunto de informações referente às ações realizadas pelas instituições de ensino desde o início da pandemia até o momento atual de transição para uma pós-pandemia. Utilizando um método de análise de dados quantitativos e qualitativos, buscou-se discorrer sobre os dados de casos de Covid-19 e seus picos como as ações envolvendo modalidade de ensino, prevenção e apoio à comunidade acadêmica. Como resultado, são apresentados de forma correlacionada alguns painéis do portal e a análise dos dados das ações, permitindo compartilhar as estratégias e a compreensão histórica de quais ações foram utilizadas no enfrentamento da pandemia e na transição de uma pós-pandemia.
- 30 June 2022
- Published by Instituto de Comunicacao e Informacao Cientifica e Tecnologica em Saude in RECIIS
- Vol. 16 (2)
Uso de Python para detecção de fake news sobre a covid-19: desafios e possibilidades
Este trabalho tem como objetivo relatar estratégias para coleta de um conjunto de dados em português para treinamento de modelos de Inteligência Artificial com vistas a identificar de forma automática fake news sobre covid-19 disseminadas durante a pandemia, a partir de código Python. Analisamos um método de detecção de fake news baseado em uma Rede Neural Recorrente e de aprendizagem supervisionada. Selecionamos um corpus com 7,2 mil textos coletados em websites e agências de notícias por Monteiro et al. (2018) com cada um previamente catalogado como verdadeiro ou falso como conjunto de dados de treino e validação. O modelo foi usado para detecção de fake news sobre covid-19 em um conjunto de notícias coletadas e classificadas pelos autores deste trabalho. O índice de acerto foi de 70%, ou seja, essa foi a taxa de sucesso da detecção dos itens catalogados.
- 30 June 2022
- Published by Instituto de Comunicacao e Informacao Cientifica e Tecnologica em Saude in RECIIS
- Vol. 16 (2)
Uso de Python para detecção de fake news sobre a covid-19: desafios e possibilidades
Este trabalho tem como objetivo relatar estratégias para coleta de um conjunto de dados em português para treinamento de modelos de Inteligência Artificial com vistas a identificar de forma automática fake news sobre covid-19 disseminadas durante a pandemia, a partir de código Python. Analisamos um método de detecção de fake news baseado em uma Rede Neural Recorrente e de aprendizagem supervisionada. Selecionamos um corpus com 7,2 mil textos coletados em websites e agências de notícias por Monteiro et al. (2018) com cada um previamente catalogado como verdadeiro ou falso como conjunto de dados de treino e validação. O modelo foi usado para detecção de fake news sobre covid-19 em um conjunto de notícias coletadas e classificadas pelos autores deste trabalho. O índice de acerto foi de 70%, ou seja, essa foi a taxa de sucesso da detecção dos itens catalogados.
- 1 January 2021
- Published by Even3
Modelagem da Dinâmica Populacional SIRM Aplicada à COVID-19 no Estado da Bahia Pelo Método de Runge-Kutta de 4ª Ordem
Encontro Nacional de Modelagem Computacional e Encontro de Ciência e Tecnologia de Materiais (978-65-86084-34-4) - Modelagem Da Dinâmica Populacional Sirm Aplicada À Covid-19 No Estado Da Bahia Pelo Método De Runge-Kutta De 4ª Ordem
- 1 January 2021
- Published by Even3
Modelagem da Dinâmica Populacional SIRM Aplicada à COVID-19 no Estado da Bahia Pelo Método de Runge-Kutta de 4ª Ordem
Encontro Nacional de Modelagem Computacional e Encontro de Ciência e Tecnologia de Materiais (978-65-86084-34-4) - Modelagem Da Dinâmica Populacional Sirm Aplicada À Covid-19 No Estado Da Bahia Pelo Método De Runge-Kutta De 4ª Ordem
- 1 January 2021
- Published by Even3
- p. 1-9
COVID-19: APLICAÇÃO DA INTEGRAÇÃO NUMÉRICA NA ANÁLISE DO ISOLAMENTO SOCIAL EM BOM JESUS DA LAPA
Encontro Nacional de Modelagem Computacional e Encontro de Ciência e Tecnologia de Materiais (978-65-86084-34-4) - Covid-19: Aplicação Da Integração Numérica Na Análise Do Isolamento Social Em Bom Jesus Da Lapa
- 1 January 2021
- Published by Even3
- p. 1-9
COVID-19: APLICAÇÃO DA INTEGRAÇÃO NUMÉRICA NA ANÁLISE DO ISOLAMENTO SOCIAL EM BOM JESUS DA LAPA
Encontro Nacional de Modelagem Computacional e Encontro de Ciência e Tecnologia de Materiais (978-65-86084-34-4) - Covid-19: Aplicação Da Integração Numérica Na Análise Do Isolamento Social Em Bom Jesus Da Lapa
- Open Access
- 5 May 2020
- Published by International Association of Online Engineering (IAOE) in International Journal of Engineering Pedagogy (iJEP)
- Vol. 10 (3) , 7-20
Teaching of Ordinary Differential Equations Using the Assumptions of the PBL Method
Mathematics is of fundamental importance to any natural sciences program (be-cause it provides analytical and approximate results that can be simulated and modeled) and perhaps to other areas of human knowledge. Ordinary differential equations are especially of fundamental importance to engineering programs be-cause the modeling of all phenomena of interest for these programs involves Or-dinary differential equations solutions; at the same time, students experience diffi-culties when learning Ordinary differential equations and about their applications to real physical scenarios. In this study, the problem-based learning method was used to study a group of mechanical engineering students at the SENAI CIMATEC University Center. The study analyzes the effectiveness of Ordinary differential equations instruction given through this program using the specified methodology. To consolidate the study, Conceptual Field Theory is used together with an evaluation of the results to observe student behaviors and attitudes in re-lation to actions arising from the method. At the end of the study, a questionnaire was completed by the students, with which they evaluated the effectiveness of the teaching methodology employed.
- 5 May 2020
- Published by International Association of Online Engineering (IAOE) in International Journal of Engineering Pedagogy (iJEP)
- Vol. 10 (3) , 7-20
Open AccessTeaching of Ordinary Differential Equations Using the Assumptions of the PBL Method
Mathematics is of fundamental importance to any natural sciences program (be-cause it provides analytical and approximate results that can be simulated and modeled) and perhaps to other areas of human knowledge. Ordinary differential equations are especially of fundamental importance to engineering programs be-cause the modeling of all phenomena of interest for these programs involves Or-dinary differential equations solutions; at the same time, students experience diffi-culties when learning Ordinary differential equations and about their applications to real physical scenarios. In this study, the problem-based learning method was used to study a group of mechanical engineering students at the SENAI CIMATEC University Center. The study analyzes the effectiveness of Ordinary differential equations instruction given through this program using the specified methodology. To consolidate the study, Conceptual Field Theory is used together with an evaluation of the results to observe student behaviors and attitudes in re-lation to actions arising from the method. At the end of the study, a questionnaire was completed by the students, with which they evaluated the effectiveness of the teaching methodology employed.
- Open Access
- 16 April 2020
- Published by Universidade Federal da Bahia in Cadernos de Prospecção
- Vol. 13 (2) , 513
Face Shield for Life 3D: produção colaborativa, usando a comunidade de makers, dos protetores faciais padrão RC3 para os profissionais de saúde em Salvador
A pandemia de COVID-19 gerou uma alta demanda por máscaras de proteção para profissionais da área de saúde. Nesse contexto, surgiu o Projeto Face Shield for Life 3D, cujo objetivo é apresentar o modelo de produção colaborativa de desenvolvimento e de produção de Face Shields. O artigo se inicia apresentando um breve histórico sobre o novo patógeno e as suas formas de contágio, a Resolução RDC n. 356, de 23 de março de 2020, e algumas orientações da ANVISA para os serviços de saúde (hospitais, postos, profissionais, suporte e assistência) em relação à importância das medidas de prevenção e de controle. Na sequência, são expostos os resultados obtidos no Projeto Face Shield for Life 3D com a criação de um hub com mais de 60 makers, somando mais de 90 impressoras 3D conectadas com o propósito de obter uma resposta rápida na produção dos Face Shields aos profissionais de saúde gratuitamente.
- 16 April 2020
- Published by Universidade Federal da Bahia in Cadernos de Prospecção
- Vol. 13 (2) , 513
Open AccessFace Shield for Life 3D: produção colaborativa, usando a comunidade de makers, dos protetores faciais padrão RC3 para os profissionais de saúde em Salvador
A pandemia de COVID-19 gerou uma alta demanda por máscaras de proteção para profissionais da área de saúde. Nesse contexto, surgiu o Projeto Face Shield for Life 3D, cujo objetivo é apresentar o modelo de produção colaborativa de desenvolvimento e de produção de Face Shields. O artigo se inicia apresentando um breve histórico sobre o novo patógeno e as suas formas de contágio, a Resolução RDC n. 356, de 23 de março de 2020, e algumas orientações da ANVISA para os serviços de saúde (hospitais, postos, profissionais, suporte e assistência) em relação à importância das medidas de prevenção e de controle. Na sequência, são expostos os resultados obtidos no Projeto Face Shield for Life 3D com a criação de um hub com mais de 60 makers, somando mais de 90 impressoras 3D conectadas com o propósito de obter uma resposta rápida na produção dos Face Shields aos profissionais de saúde gratuitamente.