APPLYING THE LIFELONG MACHINE LEARNING PARADIGM IN TUBERCULOSIS TRIAGE
APPLYING THE LIFELONG MACHINE LEARNING PARADIGM IN TUBERCULOSIS TRIAGE
Data
2022
Autores
Alves, Regina Reis da Costa
Tavares, Frederico Caetano Jandre de Assis
Trajman, Anete
Seixas, Jose Manoel de
Journal Title
Journal ISSN
Volume Title
Publisher
Sociedade Brasileira de Inteligência Computacional
Resumo
Tuberculosis (TB) and pneumonia, including pneumonia from SARS-CoV-2 infection, are among the main causes
of lower respiratory infections, which are the fourth cause of death worldwide. Recently, the World Health Organization rec-
ommended the use of computer-aided diagnosis (CAD) software as a tool to analyze chest radiographs (CXR) for TB screening
and triage. Most CAD developed to date aim to screen exclusively for TB. This work applies the lifelong machine learning
paradigm to detect both pneumonia and TB through CXRs and evaluate the models’ ability to retain and acquire knowledge.
Two well-known lifelong learning models, the Efficient Lifelong Learning Algorithm (ELLA) and Learning without Forgetting
(LwF), were applied to two public CXR datasets containing TB and pneumonia samples together with healthy CXR samples.
Pneumonia detection was learned first and TB detection was learned as second task. The SP index, a function of sensitivity and
specificity, was used to evaluate the models. We concluded that both algorithms were able to retain knowledge about pneumonia
detection and were also able to learn TB detection.
Description
Palavras-chave
Lifelong machine learning, continuous learning, tuberculosis, chest radiographs, pneumonia.
Citação
Alves RRC, Tavares FCJA, Trajman A, Seixas JM. APPLYING THE LIFELONG MACHINE LEARNING PARADIGM IN [Artigo publicado no Learning and Nonlinear Models - Journal of the Brazilian Society on Computational Intelligence (SBIC), Vol. 20, Iss. 2, pp. 63-73, 2022] Sociedade Brasileira de Inteligência Computacional. doi: 10.21528/lnlm-vol20-no2-art5.