An Explainable Machine Learning Model for Early Detection of Asthma Using Clinical and Environmental Data.
Article Sidebar
Main Article Content
Asthma is a lung disease that is a chronic respiratory disease in millions of people worldwide and, in most cases, leads to a lower quality of life and high healthcare expenditure. As early as possible, it is essential to ensure the management and prevention of serious exacerbations. The objective of the study is to come up with an explainable machine learning (ML) model, which exploits clinical and environmental data to forecast the risk of asthma in a person. The dataset combines patient-related clinical characteristics, such as age, symptoms, medical history, and results of spirometry, with the environmental variables of air pollution, humidity, and temperature. The approach will include training and testing various trained learning algorithms, such as Logistic Regression, Randome Forest, and XGBoost. SHAP and LIME are explainable AI methods that are used to achieve transparency, measure feature importance, and describe the explanation of individual predictions. The standard measurements of model performance such as accuracy, precision, recall, F1-score and ROC-AUC are used to evaluate model performance, ensuring predictive reliability and clinical relevance. Among the main results, it is possible to note that XGBoost gives the best predictive results in all measures, and the analysis of feature importance shows that the level of PM 2.5, humidity, wheezing, shortness of breath and the results of spirometry can be considered the most significant. Explainability analysis states that the predictions of the model are interpretable, which contributes to a better understanding of the model and clinical trust. Finally, the paper shows that a combination of clinical and environmental data with elucidatable machine learning offers a strong and clear framework to detect asthma at its initial stages. The method improves predictive power, enables informed medical decision-making, and provides a base of applied practice in healthcare systems, which ultimately increases patient outcomes and the adoption of explainable AI in respiratory medicine.
Downloads
References
Abbas, A., Okpapi, J. U., Njoku, C. H., Abba, A. A., Isezuo, S. A., & Danasabe, I. M. (2021). Seasonal changes and asthma exacerbations in a Sudan savanna region. Annals of African Medicine, 20(4), 302–306. https://doi.org/10.4103/aam.aam_66_20
AbdulRaheem, Y. (2023). Integrating prevention levels in healthcare: Significance and challenges. Journal of Primary Care & Community Health, 14, 1–5. https://doi.org/10.1177/21501319231186500
Alhumaidi, N. H., Dermawan, D., Kamaruzaman, H. F., & Alotaiq, N. (2025). Machine learning for disease prediction using real-world data: A systematic review. JMIR Medical Informatics, 13, e68898. https://doi.org/10.2196/68898
Alkhanani, M. F. (2025). Air quality, socioeconomic factors, and respiratory disease. Tropical Medicine and Infectious Disease, 10(2), 56. https://doi.org/10.3390/tropicalmed10020056
Alkhanbouli, R., Almadhaani, H. M. A., Alhosani, F., & Simsekler, M. C. E. (2025). Explainable AI in disease prediction: A systematic review. BMC Medical Informatics and Decision Making, 25(1), 110. https://doi.org/10.1186/s12911-025-02944-6
Chen, M.-H., Lee, G., & Hung, L.-P. (2025). AI-driven data analysis for asthma risk prediction. Healthcare, 13(7), 774. https://doi.org/10.3390/healthcare13070774
D'Amato, G. et al. (2015). Meteorological conditions, climate change, and allergic diseases. World Allergy Organization Journal, 8(1), 25. https://doi.org/10.1186/s40413-015-0073-0
Fahim, Y. A., Hasani, I. W., Kabba, S., & Ragab, W. M. (2025). Artificial intelligence in healthcare: Clinical applications and future directions. European Journal of Medical Research, 30(1), 848. https://doi.org/10.1186/s40001-025-03196-w
Goldin, J., & Cataletto, M. E. (2026). Asthma. In StatPearls. StatPearls Publishing. https://www.ncbi.nlm.nih.gov/books/NBK430901/
Gupta, S. (2022). Diagnosing asthma and COPD: The role of pulmonary function testing. Canadian Family Physician, 68(6), 441–444. https://doi.org/10.46747/cfp.6806441
Häder, A., Köse-Vogel, N., Schulz, L., Mlynska, L., Hornung, F., Hagel, S., Teichgräber, U., Lang, S. M., Pletz, M. W., Saux, C. J. L., Löffler, B., & Deinhardt-Emmer, S. (2023). Respiratory infections in the aging lung: Diagnosis, therapy, and prevention. Aging and Disease, 14(4), 1091–1104. https://doi.org/10.14336/AD.2023.0329
Kostakou, E., Kaniaris, E., Filiou, E., Vasileiadis, I., Katsaounou, P., Tzortzaki, E., Koulouris, N., Koutsoukou, A., & Rovina, N. (2019). Acute severe asthma in adolescents and adults: Assessment and management. Journal of Clinical Medicine, 8(9), 1283. https://doi.org/10.3390/jcm8091283
Johannssen, A., & Chukhrova, N. (2025). The role of explainable AI in healthcare management. Health Care Management Science, 28, 565–570. https://doi.org/10.1007/s10729-025-09720-y
Molfino, N. A., Turcatel, G., & Riskin, D. (2024). Machine learning for predicting asthma exacerbations: A narrative review. Advances in Therapy, 41(2), 534–552. https://doi.org/10.1007/s12325-023-02743-3
Monteiro, G. O. d. A. et al. (2025). Interpreting machine learning models with SHAP: Crude protein prediction in grass pastures. Agronomy, 15(12), 2780. https://doi.org/10.3390/agronomy15122780
Nouis, S. C., Uren, V., & Jariwala, S. (2025). Accountability and bias in AI-assisted healthcare decisions: Perspectives from UK professionals. BMC Medical Ethics, 26(1), 89. https://doi.org/10.1186/s12910-025-01243-z
Sundas, A., Contreras, I., Mujahid, O., Beneyto, A., & Vehi, J. (2024). Environmental factors and human health: A scoping review. Healthcare, 12(21), 2123. https://doi.org/10.3390/healthcare12212123
Tiotiu, A. I., Novakova, P., Nedeva, D., Chong-Neto, H. J., Novakova, S., Steiropoulos, P., & Kowal, K. (2020). Air pollution and asthma outcomes. International Journal of Environmental Research and Public Health, 17(17), 6212. https://doi.org/10.3390/ijerph17176212
Wiratsin, I., & Ragkhitwetsagul, C. (2025). Effectiveness of XAI in building human trust: A systematic review. IEEE Access, 13, 1–1. https://doi.org/10.1109/ACCESS.2025.3575022
Zhang, C., & Liu, L. (2025). Interpretable machine learning for medical environment comfort. Scientific Reports, 15(1), 39269. https://doi.org/10.1038/s41598-025-22972-6=
This work is licensed under a Creative Commons Attribution 4.0 International License.
All articles published in our journal are licensed under CC-BY 4.0, which permits authors to retain copyright of their work. This license allows for unrestricted use, sharing, and reproduction of the articles, provided that proper credit is given to the original authors and the source.