Use este identificador para citar ou linkar para este item: http://repositorio.ufc.br/handle/riufc/73453
Tipo: Artigo de Periódico
Título: Active balancing mechanism for imbalanced medical data in deep learning–based classification models
Autor(es): Zhang, Hongyi
Zhang, Haoke
Pirbhulal, Sandeep
Wu, Wanqing
Albuquerque, Victor Hugo Costa de
Palavras-chave: Applied computing;Computing methodologies;Life and medical sciences;Machine learning;Computação aplicada;Metodologias de computação;Ciências da vida e medicina;Aprendizado de máquina
Data do documento: 2020
Instituição/Editor/Publicador: ACM Transactions on Multimedia Computing Communications and Applications
Citação: ZHANG, Hongyi; ZHANG, Haoke; PIRBHULAI, Sandeep;WU, Wanqing; ALBUQUERQUE, Victor Hugo Costa de. Active balancing mechanism for imbalanced medical data in deep learning–based classification models. ACM Transactions on Multimedia Computing Communications and Applications, [s.l.], v. 16, n. 1s, p. 1-15, 2020.
Abstract: Imbalanced data always has a serious impact on a predictive model, and most under-sampling techniques consume more time and suffer from loss of samples containing critical information during imbalanced data processing, especially in the biomedical field. To solve these problems, we developed an active balancing mechanism (ABM) based on valuable information contained in the biomedical data. ABM adopts the Gaussian naïve Bayes method to estimate the object samples and entropy as a query function to evaluate sample infor- mation and only retains valuable samples of the majority class to achieve under-sampling. The Physikalisch Technische Bundesanstalt diagnostic electrocardiogram (ECG) database, including 5,173 normal ECG samples and 26,654 myocardial infarction ECG samples, is applied to verify the validity of ABM. At imbalance rates of 13 and 5, experimental results reveal that ABM takes 7.7 seconds and 13.2 seconds, respectively. Both results are significantly faster than five conventional under-sampling methods. In addition, at the imbalance rate of 13, ABM-based data obtained the highest accuracy of 92.23% and 97.52% using support vector machines and modified convolutional neural networks (MCNNs) with eight layers, respectively. At the imbalance rate of 5, the processed data by ABM also achieved the best accuracy of 92.31% and 98.46% based on support vector machines and MCNNs, respectively. Furthermore, ABM has better performance than two compared methods in F1-measure, G-means, and area under the curve. Consequently, ABM could be a useful and effective approach to deal with imbalanced data in general, particularly biomedical myocardial infarction ECG datasets, and the MCNN can also achieve higher performance compared to the state of the art
URI: http://www.repositorio.ufc.br/handle/riufc/73453
ISSN: 1551-6865
Tipo de Acesso: Acesso Aberto
Aparece nas coleções:DEEL - Artigos publicados em revista científica

Arquivos associados a este item:
Arquivo Descrição TamanhoFormato 
2020_art_hzhang.pdf2,38 MBAdobe PDFVisualizar/Abrir


Os itens no repositório estão protegidos por copyright, com todos os direitos reservados, salvo quando é indicado o contrário.