Enabling Continuous Breathing-Phase Contextualization via Wearable-Based Impedance Pneumography and Lung Sounds: A...

Area of Science:

• Respiratory Medicine
• Biomedical Engineering
• Signal Processing

Background:

• Chronic respiratory diseases are a major global health burden.
• Pulmonary auscultation is key for respiratory health assessment.
• Current technology limits computerized lung sound analysis by not capturing breathing context.

Purpose of the Study:

• To develop and evaluate a wearable system for simultaneous lung sound and impedance pneumography measurements.
• To extract breathing-phase context from impedance pneumography signals.
• To assess the value of phase-contextualized acoustic features for analyzing lung sounds.

Main Methods:

• A wearable system (WIRS) captured lung sounds and impedance pneumography simultaneously from 10 healthy volunteers.
• Participants performed normal and forced-expiratory breathing maneuvers.
• Machine learning analyzed phase-by-phase and breath-by-breath acoustic features.

Main Results:

• Phase-contextualized features accurately distinguished between different breathing types, achieving an F1 Score of 84.1 ±11.4%.
• Phase-by-phase features were the most significant contributors to performance.
• Contextualized models consistently outperformed traditional breath-by-breath models.

Conclusions:

• Incorporating breathing-phase context significantly enhances computerized lung sound analysis.
• Multimodal sensing systems like WIRS can improve the clinical utility of lung sound analysis.
• This technology has the potential to advance patient care for respiratory conditions.