Document Type : Original Paper
Authors
1 1 Department of Statistics, Faculty of Mathematical Sciences, Tarbiat Modares University, Tehran, Islamic Republic of Iran
2 2 Department of Mathematics, Faculty of Mathematics and Computer Science, Kharazmi University, Tehran, Islamic Republic of Iran
Abstract
The objective of this study is to investigate the factors influencing asthma attacks in children under six years old using machine learning (ML) methods. There are many statistical methods for data classification that can be used to classify medical data. But using the data itself as well as a set of different methods in machine learning can provide vast and more comparable results. Hence, this study applied ML approaches to predict asthma and second anoxic tonic seizures due to asthma (ATSA) based on variables such as first ATSA, age, region of residence, parent smoking status, and parents' asthma history. The results revealed that children's age and place of residence significantly affected the duration of asthma attacks, with children living in certain areas of Tehran experiencing shorter intervals between attacks due to high air pollution. Machine learning techniques proved useful in predicting ATSA based on age, gender, living region, parents' smoking status, and asthma history, with the AdaBoost method highlighting the importance of the child's age and living area in predicting ATSA.
Keywords
Main Subjects
3. Yunginger JW, Reed CE, O’Connell EJ, Melton 3rd LJ, O’Fallon WM, Silverstein MD. A community-based study of the epidemiology of asthma. Incidence rates, 1964–1983. Am Rev Respir Dis. 1992;146(4):888–94