Detecting and classifying COPD early among other chronic lung conditions poses challenges for several reasons. These include asymptomatic disease states, symptom similarity with conditions like difficulty breathing and coughing, under diagnosis, and the complexity of diagnostic tests. Addressing  these challenges  are crucial to reduce the mortality and morbidity caused by COPD. Advanced machine learning (ML) techniques outshine classical ML tools by leveraging complex algorithms to analyze huge datasets with higher accuracy and effectiveness. These methods, such as hybrid methods can discover convoluted patterns in data, leading to further precise predictions and enhanced overall performance in various real life applications including healthcare. Thus the objective of this study was to implement hybrid machine learning methods for the detection and classification of COPD and healthy individuals. In this study, the proposed Meta Stratified Optimal Hybrid Model (MSOHM) has achieved 99.5% accuracy in identifying and classifying COPD and healthy subjects. The proposed model further compared with other researchers proposed existing models validated on the same dataset that is used in this study. The suggested robust MSOHM classification model offers significant advantages to medical professionals by enabling early prediction of COPD with high accuracy. This eventually supports in preventing exacerbations of the COPD disease and ultimately saving lives.