<p>This study presents a novel approach for detecting and classifying chronic diseases (CDs), focusing on prevalent conditions such as Asthma, cancer, and chronic obstructive pulmonary disease (COPD). Early identification of these diseases is challenging due to overlapping symptoms, limited clinical expertise, and variations in device sensitivity. To address this, this study proposes FLOSE, a multi-optimization-based stacked ensemble model integrating the firework and flower pollination algorithms. It operates in three stages: feature selection, optimal model generation, and stacked ensemble model creation aiming to detect CDs and distinguish symptomatically similar conditions like COPD and Asthma. The model’s effectiveness was evaluated using two datasets, Exasens and Breast Cancer. The stacked ensemble framework consists of four base classifiers—Random Forest, Support Vector Machine, Multilayer Perceptron (MLP), and ExtraTrees (ET)—with their predictions aggregated by the meta-learner, Extreme Learning Machine. Experimental results show that FLOSE outperforms individual base classifiers, achieving 96.25% accuracy, 96.24% precision, 96.25% recall, and a 96% F1-score for Exasens, and 95.6% accuracy, 95% precision, 95.1% recall, and a 96% F1-score for Breast Cancer. These findings highlight FLOSE as a promising approach for accurate chronic disease detection and classification.</p>