Study of Voice and Capnometry to Detect Emphysema in COPD Patients Using Machine Learning

What this Trial Is About

Researchers are investigating whether voice recordings or capnometry measurements, alone or combined with other non-invasive biomarkers, can detect emphysema on chest CT scans in people with chronic obstructive pulmonary disease COPD. The study aims to develop a machine-learning algorithm to classify the extent of emphysema using features from speech and capnometry data. This research is observational and focuses on patients diagnosed with COPD with related respiratory symptoms and CT imaging results. Participants will perform various voice tasks such as paced reading, sustained vowels, coughing, and quiet breathing. Capnometry measurements will be taken before and after a light exercise task, specifically the 5-sit-to-stand test. These procedures will be done in a single clinic visit. Clinical data including lung function tests and CT scans are already collected as part of routine care. The study will analyze voice and capnometry data using machine learning models to classify emphysema severity. During the visit, participants will complete voice-related tasks and capnometry tests, perform a short exercise, and have blood drawn. Researchers will measure outcomes including emphysema extent on CT scans, various voice features, and capnography parameters. Additional lung function measures and serum biomarkers will also be reviewed. The study monitors data quality and uses statistical models to explore how well these non-invasive measures predict emphysema. Participation involves a single visit with no intervention treatment.

Exploring Novel Biomarkers for Emphysema Detection