A healthcare monitoring system needs the support of recent technologies such as artificial intelligence (AI), machine learning (ML), and big data, especially during the COVID-19 pandemic. This global pandemic has already taken millions of lives. Both infected and uninfected people have generated big data where AI and ML can use to combat and detect COVID-19 at an early stage. Motivated by this, an improved ML framework for the early detection of this disease is proposed in this paper. The state-of-the-art Harris hawks optimization (HHO) algorithm with an improved objective function is proposed and applied to optimize the hyperparameters of the ML algorithms, namely HHO-based eXtreme gradient boosting (HHOXGB), light gradient boosting (HHOLGB), categorical boosting (HHOCAT), random forest (HHORF) and support vector classifier (HHOSVC). An ensemble technique was applied to these optimized ML models to improve the prediction performance. Our proposed method was applied to publicly available big COVID-19 data and yielded a prediction accuracy of 92.38% using the ensemble model. In contrast, HHOXGB provided the highest accuracy of 92.23% as a single optimized model. The performance of the proposed method was compared with the traditional algorithms and other ML-based methods. In both cases, our proposed method performed better. Furthermore, not only the classification improvement, but also the features are analyzed in terms of feature importance calculated by SHapely adaptive exPlanations (SHAP) values. A graphical user interface is also discussed as a potential tool for nonspecialist users such as clinical staff and nurses. The processed data, trained model, and codes related to this study are available at GitHub.

Asthma is a chronic and airway-induced disease, causing the incidence of bronchus inflammation, breathlessness, wheezing, is drastically becoming life-threatening. Even in the worst cases, it may destroy the quality to lead. Therefore, early detection of asthma is urgently needed, and machine learning can help identify asthma accurately. In this paper, a novel machine learning framework, namely BOMLA ( B ayesian O ptimisation-based M achine L earning framework for A sthma) detector has been proposed to detect asthma. Ten classifiers have been utilized in the BOMLA detector, where Support Vector Classifier (SVC), Random Forest (RF), Gradient Boosting Classifier (GBC), eXtreme Gradient Boosting (XGB), and Artificial Neural Network (ANN) are state-of-the-art classifiers. In contrast, Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QLDA), Naive Bayes (NB), Decision Tree (DT), and K-Nearest Neighbor (KNN) are conventional popular classifiers. ADASYN algorithm has also been employed in the BOMLA detector to eradicate the issues created due to the imbalanced dataset. It has even been attempted to delineate how the ADASYN algorithm affects the classification performance. The highest accuracy (ACC) and Matthews’s correlation coefficient (MCC) for an Asthma dataset provide 94.35% and 88.97%, respectively, using BOMLA detector when SVC is adapted, and it has been increased to 96.52% and 93.04%, respectively, when ensemble technique is adapted. The one-way analysis of variance (ANOVA) has also been performed in the 10-fold cross-validation to measure the statistical significance. A decision support system has been built as a potential application of the proposed system to visualize the probable outcome of the patient. Finally, it is expected that the BOMLA detector will help patients in their early diagnosis of asthma.