Title: Using Machine Learning Techniques to Predict Type 2 Asthma Disease
DOI:
https://doi.org/10.65405/cjos.2025.810Keywords:
artificial intelligence (AI), asthma, , chronic disease, machine learning, asthma, XGBoostAbstract
Asthma, a chronic inflammatory disease of the airways, affects millions globally. Among its
phenotypes, Type 2 asthma is characterized by eosinophilic inflammation and responds differently
to treatment compared to non-Type 2 variants. Accurate early diagnosis of this subtype is critical to
ensuring appropriate therapy and reducing long-term complications. This research investigates the
application of machine learning techniques to predict the presence of Type 2 asthma using a
structured and anonymized clinical dataset obtained from a trusted health registry. The study
leverages a real dataset from the University of Washington, which includes 2,392 patient records
with a wide range of features, including demographic information, lifestyle habits, environmental
exposures, clinical symptoms, and spirometry results. Various supervised machine learning
algorithms, including Logistic Regression, Support Vector Machines, Decision Trees, Random
Forest, and XGBoost, were trained and evaluated. Results show that ensemble methods outperform
baseline models, highlighting the promise of machine learning in improving diagnostic precision in
asthma care. The findings demonstrate that the employed methods can predict and provide a
preliminary diagnosis of asthma based on disease-related symptoms, thereby assisting physicians in
delivering better and faster care to patients before their condition deteriorates
Downloads
References
[1] Global Initiative for Asthma. Global Strategy for Asthma Management and
Prevention, 2023.
[2] Wenzel, S. E. (2012). Asthma phenotypes: the evolution from clinical to
molecular approaches. Nature Medicine, 18(5), 716–725.
[3] Berry, M., et al. (2007). The use of exhaled nitric oxide concentration to identify
eosinophilic airway inflammation: an observational study. Thorax, 62(12), 1053-
1057.
[4] Taylor, D. R., et al. (2006). A systematic review of the diagnostic accuracy of
exhaled nitric oxide in the management of asthma. Health Technology Assessment,
10(8), 1-158.
[5] Wagener, A. H., et al. (2015). External validation of blood eosinophils, FE(NO)
and serum
[6] Zhang, Z., Deng, L., & Wang, Y. (2018). Asthma Severity Classification Using
Machine Learning. Journal of Medical Systems, 42(5), 87.
[7] Sun, X., Wang, J., & Li, Q. (2019). A Support Vector Machine Model for
Detecting Asthma in Children Using Environmental and Clinical Data. IEEE Journal
of Biomedical and Health Informatics, 23(4), 1625-1632.
[8] Wenzel, S. E. (2012). Asthma phenotypes: the evolution from clinical to
molecular approaches. Nature Medicine, 18(5), 716–725.
[9] Kachroo, P., Stewart, I. D., Kelly, R. S., et al. (2021). Unsupervised phenotyping
of severe asthma reveals distinct clusters with high healthcare utilization. Journal of
Allergy and Clinical Immunology: In Practice, 9(7), 2765-2774.
[10] Lakhani, P., & Sundaram, B. (2017). Deep Learning at Chest Radiography:
Automated Classification of Pulmonary Tuberculosis by Using Convolutional Neural
Networks. Radiology, 284(2), 574-582.
[11] Abdel-Rahman, M., El-Hag, N. A., & Seddik, A. F. (2020). Feature Selection
using LASSO Regression for Asthma Exacerbation Prediction. Computers in Biology
and Medicine, 125, 103996.
[12] Dharmage, S. C., Perret, J. L., & Custovic, A. (2019). Epidemiology of Asthma
in Children and Adults. Frontiers in Pediatrics, 7, 246.
[13] Deo, R. C. (2015). Machine Learning in Medicine. Circulation, 132(20), 1920-
1930.
[14] Nguyen, B. P., Pham, H. N., Tran, H., et al. (2020). Predicting COPD
Progression Using Ensemble Machine Learning. Scientific Reports, 10(1), 22067.
[15] Chen, T., & Guestrin, C. (2016). XGBoost: A Scalable Tree Boosting System.
In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge
Discovery and Data Mining (pp. 785–794).
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Comprehensive Journal of Science
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
