AI-Driven Smartphone Screening for Acute COPD Exacerbations: A Non-Self-Report Approach to Improve Health Equity in Developing Regions

Acute exacerbations of chronic obstructive pulmonary disease (AECOPD) are critical clinical events that necessitate prompt intervention. However, their detection remains challenging in primary care settings, especially in resource-limited regions. A lack of disease awareness among both patients and healthcare providers often leads to delayed diagnoses and suboptimal management. To address this issue, we developed an AI-based AECOPD detection system that leverages standard mobile phone microphones for auscultation, specifically designed for novice users and eliminating the need for subjective symptom assessments or patientreported scales. Our system demonstrated robust performance in automatically detecting exacerbations, achieving an area under the curve (AUC) of 0.955 (95% CI: 0.929-0.976). This research highlights the potential of AI-driven solutions to enhance COPD management in underserved populations with limited access to specialist medical resources, thereby promoting health equity. The digital health system shows promise for improved long-term management of COPD and is projected to save 41.81 billion CNY (Median: 24.25 billion, 95% CI: -7.75-198.33 billion), particularly benefiting primary care settings where access to pulmonary specialists is constrained.