A Machine Learning Model for the Early Detection of Adults' Serious Asthma

Asthma is a significant global health concern, and early detection in both children and adults is crucial for effective intervention. However, developing a reliable predictive model for asthma has proven challenging due to low accuracy and limited sample sizes in existing studies. This research addresses these limitations by leveraging telemonitoring data to establish a robust predictive model for asthma symptoms. The primary objective is to assess the efficacy of machine learning (ML) algorithms in detecting asthma severity levels, categorized as “Severity-None,” “Severity-Mild,” and “Severity-Moderate.” This work presents a comprehensive perspective on asthma disease prediction (ADP) through various clustering and classification techniques aimed at enhancing accuracy, particularly with ample training data. We employ ML clustering and classifiers, including K-Means, Multi-Layer Perceptron (MLP), Decision Tree (DT), Stochastic Gradient Descent (SGD), and Naive Bayes (NB), to classify asthma symptoms as either positive or negative indicators.