Introducing lifelong machine learning in the active tuberculosis classification through chest radiographs

Página do item simplificado
dc.contributor.authorAlves, Regina Reis da Costa
dc.contributor.authorTavares, Frederico Caetano Jandre de Assis
dc.contributor.authorSeixas, José Manoel de
dc.contributor.authorTrajman, Anete
dc.date.accessioned2024-12-09T17:15:45Z
dc.date.available2024-12-09T17:15:45Z
dc.date.issued2021
dc.description.abstractTuberculosis (TB) is a contagious disease which is among the top 10 causes of death in the world. In order to eliminate the disease by 2050, the treatment of TB infection (TBI) is essential, which requires radiological reports to exclude active tuberculosis. The automatic X-ray classifiers used today are based on models that do not guarantee the retention of knowledge if they need to learn new tasks over time. This work proposes the introduction of the lifelong machine learning (LML) paradigm in automatic X-ray classifiers aimed at helping to diagnose active TB (ATB). Two LML algorithms, Efficient Lifelong Learning Algorithm (ELLA) and Learning without Forgetting (LwF), are applied to the TB and pneumonia classification tasks. The results show that it is possible to keep the performance in both tasks with the LML paradigm.
dc.identifier.citationAlves RRC, Tavares FCJA, Seixas JM, Trajman A. Introducing lifelong machine learning in the active tuberculosis classification through chest radiographs [Apresentação no XV Congresso Brasileiro de Inteligência Computacional: Artigos Apresentados nas Seções Técnicas CBIC; 2021, Joinville, SC, Brasil]. doi: 10.21528/CBIC2021-119.
dc.identifier.otherDOI: 10.21528/CBIC2021-119
dc.identifier.urihttps://dspace.inc.saude.gov.br/handle/123456789/684
dc.language.isoen
dc.publisherSociedade Brasileira de Inteligência Computacional
dc.subjectlifelong learningen
dc.subjectmachine learningen
dc.subjecttuberculosisen
dc.subjecttuberculosis infectionen
dc.subjectlatent tuberculosisen
dc.subjectx-ray classificationen
dc.titleIntroducing lifelong machine learning in the active tuberculosis classification through chest radiographs
dc.typePresentation
Arquivos
Original bundle
Agora exibindo 1 - 1 de 1
thumbnail.default.alt
Nome:
Alves RRC et al_SBIC.pdf
Tamanho:
84.74 KB
Formato:
Adobe Portable Document Format
Descrição:
Baixar
License bundle
Agora exibindo 1 - 1 de 1
thumbnail.default.placeholder
Nome:
license.txt
Tamanho:
1.71 KB
Formato:
Item-specific license agreed to upon submission
Descrição:
Baixar
Coleções
Produção Docente