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

thumbnail.default.alt
Data
2021
Autores
Alves, Regina Reis da Costa
Tavares, Frederico Caetano Jandre de Assis
Seixas, José Manoel de
Trajman, Anete
Journal Title
Journal ISSN
Volume Title
Publisher
Sociedade Brasileira de Inteligência Computacional
Resumo
Tuberculosis (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.
Description
Palavras-chave
lifelong learning, machine learning, tuberculosis, tuberculosis infection, latent tuberculosis, x-ray classification
Citação
Alves 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.