Leveraging Machine Learning for Pattern-Based Detection of Respiratory Illnesses

Lung sound analysis presents a promising, noninvasive, and potentially cost-effective approach for detecting respiratory diseases. This research aims to assess the feasibility of using lung sounds for pattern recognition and detection of various respiratory disorders, specifically by conducting an analysis of cough sounds. We review a range of recent studies that explore the use of deep learning models, machine learning algorithms, and feature extraction techniques in this context. These studies are categorized based on their methodologies, and their performance in accurately classifying lung sounds is discussed. Research findings highlight the significant diagnostic potential of lung sound analysis, demonstrating improved classification accuracies across various disease types. However, challenges such as data inconsistency, noise interference, and limited model generalizability remain. It captures the current state of lung sound analysis research in respiratory disease detection, offering insights that could drive future advancements in this critical field.