<p>Respiratory Sound Analysis (ReSA) has advanced significantly with the integration of artificial intelligence, attracting increased research interest due to its potential to aid in diagnosing lung pathologies. The automated detection and classification of abnormal lung sounds can help diagnose conditions such as asthma, lung cancer, acute respiratory infections, tuberculosis, and chronic obstructive pulmonary disease. However, despite progress in recognizing adventitious lung sounds, their clinical application remains limited. Computational respiratory sound analysis has been extensively studied for over seven decades, leading to the development of various techniques for classifying, detecting, and identifying respiratory diseases. This review highlights the impact of machine learning and deep learning in computer-based respiratory sound analysis, emphasizing their ability to distinguish abnormal lung sounds linked to specific disorders. A systematic review of 100 studies from 2015 to February 2025 ensures a robust analysis. This paper explores respiratory sound databases, provides an overview of lung sounds and their features, and summarizes pre-processing techniques, feature extraction methods, classification approaches, and performance evaluation metrics for ReSA models. Comparative analyses of state-of-the-art methodologies and datasets enable researchers to assess effective techniques for ReSA. Based on these findings, this study offers precise recommendations to address critical gaps in identifying and categorizing respiratory sounds, fostering innovative approaches and advancing medical knowledge in this domain.</p>