Advances in the development of COPD screening models using respiratory oscillometry: An evolution from traditional regression to machine learning.

AIM OF THE STUDY: Chronic obstructive pulmonary disease (COPD) is a leading cause of morbidity and mortality worldwide, and early screening is critical for improving patient outcomes. This review aims to provide a comprehensive theoretical foundation and practical guidance for developing more accurate and efficient COPD screening tools by summarizing the research progress in constructing COPD screening models based on respiratory oscillometry (RO). MATERIALS AND METHODS: We conducted a narrative literature review of PubMed, Scopus, and Embase through May 2026 to identify studies on respiratory oscillometry-based COPD screening models, with particular attention to comparisons between traditional regression/discriminant approaches and machine-learning algorithms. Respiratory oscillometry (RO), a noninvasive and convenient method for assessing lung function, was used as the technical basis. The review systematically traces the evolution of COPD screening models from traditional statistical regression models to advanced machine learning approaches, analyzing technical characteristics, performance metrics, clinical applicability, and challenges associated with different modeling techniques. RESULTS: Key findings indicate machine learning algorithms outperform traditional logistic regression in COPD screening model accuracy, with RO technology demonstrating unique advantages in noninvasive lung function assessment through cross-comparison of modeling techniques. CONCLUSIONS: This review confirms RO-based machine learning models as the optimal approach for COPD screening, providing critical guidance for developing next-generation tools to enhance early detection accuracy and clinical applicability.