Advanced Ensemble Learning Approach for Asthma Prediction: Optimization and Evaluation
Vatsal, Sunil Kumar, Riya Riya, Sanat Rampal, Mrigank Gaur, Mrinal Gaur
Abstract
Chronic respiratory conditions like asthma have a major negative influence on patients’ standards of life and place a heavy strain on healthcare systems. Accurately predicting asthma exacerbations is essential for prompt intervention and better disease control. This study focused on creating and assessing an ensemble model that incorporates decision tree, random forest, and gradient-boost algorithms to propose artificial intelligence (ML) based models for forecasting asthma exacerbations. 29,396 asthma patients’ data, gathered from national registers and electronic medical records that encompass clinical and epidemiological characteristics (such as comorbidities and healthcare contacts), are included in the study. The ensemble model’s 90% accuracy rate indicates that it can accurately forecast exacerbations of asthma. The study also addresses the obstacles and suggestions for further research regarding the application of machine learning (ML)–based asthma forecasting models in clinical settings. The results demonstrate the potential of ML algorithms to increase asthma prediction accuracy, which could lead to better-individualized asthma treatment and lessened asthma exacerbations’ impact on patients and the health care system,