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

Wolfgang Mayr,1,* Andreas Triantafyllopoulos,2,3,* Anton Batliner,2,3 Björn W Schuller,2– 5 Thomas M Berghaus1,6 1Department of Cardiology, Respiratory Medicine and Intensive Care, University Hospital Augsburg, Augsburg, Germany; 2Chair of Health Informatics (CHI), Department of Clinical Medicine, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany; 3Munich Center for Machine Learning (MCML), Munich, Germany; 4Group on Language Audio, & Music (GLAM), Imperial College, London, UK; 5Munich Data Science Institute (MDSI), Munich, Germany; 6Medical Faculty, Ludwig Maximilians University of Munich, Munich, Germany*These authors contributed equally to this workCorrespondence: Wolfgang Mayr, Department of Cardiology, Respiratory Medicine and Intensive Care, University Hospital Augsburg, Stenglinstrasse 2, Augsburg, D-86156, Germany, Email [email protected] Andreas Triantafyllopoulos, Chair of Health Informatics, Department of Clinical Medicine, Klinikum rechts der Isar, Technical University of Munich, Ismaninger Straße 22, Munich, 81675, Germany, Email [email protected]: Chronic obstructive pulmonary disease (COPD) affects breathing, speech production, and coughing. We evaluated a machine learning analysis of speech for classifying the disease severity of COPD.Methods: In this single centre study, non-consecutive COPD patients were prospectively recruited for comparing their speech characteristics during and after an acute COPD exacerbation. We extracted a set of spectral, prosodic, and temporal variability features, which were used as input to a support vector machine (SVM). Our baseline for predicting patient state was an SVM model using self-reported BORG and COPD Assessment Test (CAT) scores.Results: In 50 COPD patients (52% males, 22% GOLD II, 44% GOLD III, 32% GOLD IV, all patients group E), speech analysis was superior in distinguishing during and after exacerbation status compared to BORG and CAT scores alone by achieving 84% accuracy in prediction. CAT scores correlated with reading rhythm, and BORG scales with stability in articulation. Pulmonary function testing (PFT) correlated with speech pause rate and speech rhythm variability.Conclusion: Speech analysis may be a viable technology for classifying COPD status, opening up new opportunities for remote disease monitoring.Keywords: COPD, pathological speech, feature interpretation, personalization, digital health