Assessing the Clinical and Functional Status of COPD Patients Using Speech Analysis During and After Exacerbation

Area of Science:

• Pulmonary Medicine
• Biomedical Engineering
• Data Science

Background:

• Chronic obstructive pulmonary disease (COPD) significantly impacts respiratory function, speech, and coughing.
• Accurate classification of COPD severity is crucial for effective patient management and treatment.

Purpose of the Study:

• To evaluate the efficacy of machine learning-based speech analysis in classifying COPD disease severity.
• To compare the performance of speech analysis against established clinical scores (BORG, CAT) for COPD assessment.

Main Methods:

• Prospective recruitment of 50 COPD patients (GOLD II-IV, Group E) in a single-center study.
• Extraction of spectral, prosodic, and temporal speech features.
• Utilized Support Vector Machine (SVM) models for classification, comparing speech features against BORG and CAT scores.

Main Results:

• Speech analysis achieved 84% accuracy in distinguishing COPD exacerbation status, outperforming BORG and CAT scores.
• CAT scores correlated with reading rhythm; BORG scales correlated with articulation stability.
• Pulmonary function testing (PFT) showed correlations with speech pause rate and rhythm variability.

Conclusions:

• Speech analysis demonstrates potential as a viable tool for classifying COPD status.
• This technology opens avenues for remote monitoring of COPD patients.
• Machine learning applied to speech offers a novel approach to COPD assessment.