Detection of respiratory changes in cystic fibrosis with the use of machine learning algorithms

• Noemi da Paixão Pinto UERJ
• Jorge Luís do Amaral UERJ
• Pedro Lopes de Melo UERJ

Abstract

Advances in the treatment of cystic fibrosis have allowed patients to reach adulthood. As an alternative, the Forced Oscillations Technique (FOT) is being used in the respiratory system analysis and must prove its efficiency. Thus, this work proposes the use of machine learning algorithms to aid the investigation and diagnosis of respiratory changes in cystic fibrosis through the data provided by FOT. During the experiments, the used models presented an AUC value varying from 0.87 to 0.89, showing that the use of machine learning algorithms increased accuracy in diagnosis of respiratory changes in patients who suffer from cystic fibrosis.

Keywords: Fibrose cística, Diagnóstico de alterações respiratórias, Aprendizado de máquina, Técnica de Oscilações Forçadas

References

Amaral, J., Lopes, A., Faria, A. e Melo, P. (2015) “Machine learning algorithms and forced oscillation measurements to categorise the airway obstruction severity in chronic obstructive pulmonary disease”, Em: Computer Methods and Programs in Biomedicine, Elsevier, Volume 118, páginas 186-197.

Amaral, J., Lopes, A., Jansen, J., Faria, A. e Melo, P. (2013) “An improved method of early diagnosis of smoking-induced respiratory changes using machine learning algorithms”, Em: Computer Methods and Programs in Biomedicine, Elsevier, Volume 112, páginas 441-454.

Amaral, J., Lopes, A., Veiga, J., Faria, A. e Melo, P. (2017) “High-accuracy detection of airway obstruction in asthma using machine learning algorithms and forced oscillation measurements”, Em: Computer Methods and Programs in Biomedicine, Elsevier, Volume 144, páginas 113–125.

Andersen, D. (1938) “Cystic fibrosis of the pancreas and its relation to celiac disease clinical and pathologic study”, Em: American Journal of Diseases of Children, Volume 56 (2), páginas 344-399.

Dalcin, P. e Silva, F. (2008) “Fibrose cística no adulto: aspectos diagnósticos e terapêuticos”, Em: Jornal Brasileiro de Pneumologia, páginas 107-117.

Dubois, A., Brody, A., Lewis D. e Burgess, B. (1956) “Oscillation mechanics of lungs and chest in man”, Em: Journal of applied physiology, Volume 8, páginas 587-594.

Faceli, K., Lorena, A., Gama, J. e Carvalho, A. (2011) “Inteligência Artificial: uma Abordagem de Aprendizado de Máquina”, Editora LTC.

Hastie, T., Tibshirani, R. e Friedman, J. (2008) “The Elements of Statistical Learning: Data Mining, Inference and Prediction”, Em: Springer.

Huang, J. e Ling C. (2005) “Using AUC and Accuracy in Evaluating Learning Algorithms”, Em: IEEE Transaction Knowledge and Data Engineering, Volume 17, Número 3, páginas 299–310.

Liaw, A. e Wiener, M. (2002) “Classification and Regression by Random Forest”, Em: R. News, Volume 2/3, páginas 18–22.

Lima, A., Faria, A. e Lopes, A. (2010) “Técnica de oscilações forcadas na avaliação funcional de pacientes com fibrose cística com idade superior a 18 anos”, Em: Pulmão RJ.

Margineantu, D. e Dietterich, T. (1997) “Pruning Adaptive Boosting, Machine Learning”, Em: Proceedings of the Fourteenth International Conference, páginas 211-218, 1997.

Metz, C. (1978) “Basic Principles of ROC Analysis”, Em: Seminars in Nuclear Medicine, Volume 8, Número 4.