Please use this identifier to cite or link to this item:
https://www.arca.fiocruz.br/handle/icict/66894
SCREENING FOR ACTIVE PULMONARY TUBERCULOSIS: DEVELOPMENT AND APPLICABILITY OF ARTIFICIAL NEURAL NETWORK MODELS
Author
Affilliation
Electrical Engineering Program, Department of Electronics and Computer Engineering, COPPE/POLI, Federal University of Rio de Janeiro, Brazil
Department of Public Health, School of Health Sciences, University of Brasilia, Brazil / Centre for Operational Research, The International Union Against Tuberculosis and Lung Disease, Paris, France
Electrical Engineering Program, Department of Electronics and Computer Engineering, COPPE/POLI, Federal University of Rio de Janeiro, Brazil / Signal Processing Lab, Electrical Engineering Program, Alberto Coimbra Institute, Polytechnic School, Federal University of Rio de Janeiro, Brazil
Signal Processing Lab, Electrical Engineering Program, Alberto Coimbra Institute, Polytechnic School, Federal University of Rio de Janeiro, Brazil
Tuberculosis Academic Program, Medical School, Federal University of Rio de Janeiro, Brazil
Tuberculosis Academic Program, Medical School, Federal University of Rio de Janeiro, Brazil
Tuberculosis Academic Program, Medical School, Federal University of Rio de Janeiro, Brazil
Tuberculosis Academic Program, Medical School, Federal University of Rio de Janeiro, Brazil
Centre for Operational Research, The International Union Against Tuberculosis and Lung Disease, Paris, France
Tuberculosis Academic Program, Medical School, Federal University of Rio de Janeiro, Brazil
Fundação Oswaldo Cruz. Centro de Desenvolvimento Tecnológico em Saúde. Instituto Nacional de Ciência e Tecnologia de Inovação em Doenças de Populações Negligenciadas. Rio de Janeiro, RJ, Brasil.
Department of Public Health, School of Health Sciences, University of Brasilia, Brazil / Centre for Operational Research, The International Union Against Tuberculosis and Lung Disease, Paris, France
Electrical Engineering Program, Department of Electronics and Computer Engineering, COPPE/POLI, Federal University of Rio de Janeiro, Brazil / Signal Processing Lab, Electrical Engineering Program, Alberto Coimbra Institute, Polytechnic School, Federal University of Rio de Janeiro, Brazil
Signal Processing Lab, Electrical Engineering Program, Alberto Coimbra Institute, Polytechnic School, Federal University of Rio de Janeiro, Brazil
Tuberculosis Academic Program, Medical School, Federal University of Rio de Janeiro, Brazil
Tuberculosis Academic Program, Medical School, Federal University of Rio de Janeiro, Brazil
Tuberculosis Academic Program, Medical School, Federal University of Rio de Janeiro, Brazil
Tuberculosis Academic Program, Medical School, Federal University of Rio de Janeiro, Brazil
Centre for Operational Research, The International Union Against Tuberculosis and Lung Disease, Paris, France
Tuberculosis Academic Program, Medical School, Federal University of Rio de Janeiro, Brazil
Fundação Oswaldo Cruz. Centro de Desenvolvimento Tecnológico em Saúde. Instituto Nacional de Ciência e Tecnologia de Inovação em Doenças de Populações Negligenciadas. Rio de Janeiro, RJ, Brasil.
Abstract in Portuguese
A tuberculose (TB) continua sendo um desafio significativo para a saúde pública, motivado pela diversidade de cenários epidemiológicos de assistência médica, como outros fatores. A triagem custo-efetiva tem importância substancial para o controle da TB, exigindo novas ferramentas de diagnóstico. Este artigo propõe uma ferramenta de suporte à decisão (DST) para triagem de pacientes com TB pulmonar (PTB) em uma clínica secundária. A DST é composta por um modelo de ressonância adaptativa (iART) para identificação de grupos de risco (baixo, médio e alto) e uma rede neural perceptron multicamadas (MLP) para classificar pacientes como PTB ativo ou inativo. Nossa ferramenta atinge uma sensibilidade geral (SE) e especificidade (SP) de 92% (IC de 95%; 79-97) e 58% (IC de 95%; 47-68), respectivamente. Os valores de SE para pacientes com esfregaço positivo e negativo são 96% (IC de 95%; 80-99) e 82% (IC de 95%; 52-95), bem como maiores que 83% (IC de 95%; 43-97) em casos de baixo e alto risco. Mesmo em cenários com prevalência de até 20%, valores preditivos negativos superiores a 95% são obtidos. O DST proposto fornece um pré-teste rápido e de baixo custo para pacientes com PTB presuntivo, o que é útil para orientar testes confirmatórios e gerenciamento de pacientes, especialmente em cenários com recursos limitados em países de baixa e média renda.
Abstract
Tuberculosis (TB) remains a significant public health challenge, motivated by the diversity of healthcare epidemiological settings, as other factors. Cost-effective screening has substantial importance for TB control, demanding new diagnostic tools. This paper proposes a decision support tool (DST) for screening pulmonary TB (PTB) patients at a secondary clinic. The DST is composed of an adaptive resonance model (iART) for risk group identification (low, medium and high) and a multilayer perceptron (MLP) neural network for classifying patients as active or inactive PTB. Our tool attains an overall sensitivity (SE) and specificity (SP) of 92% (95% CI; 79-97) and 58% (95% CI; 47-68), respectively. SE values for smear-positive and smear-negative patients are 96% (95% CI; 80-99) and 82% (95% CI; 52-95), as well as higher than 83% (95% CI; 43-97) in low and high-risk cases. Even in scenarios with prevalence up to 20%, negative predictive values superior to 95% are obtained. The proposed DST provides a quick and low-cost pretest for presumptive PTB patients, which is useful to guide confirmatory testing and patient management, especially in settings with limited resources in low and middle-incoming countries.
Share