Detection of severe obstructive sleep apnea through voice analysis

• The present study focuses on voice alone as a primary discriminating source of information between healthy subjects and severe OSA.
• Statistical analysis as well as performance of several classifiers indicate that voice has a clear potential to detect severe OSA.
• Analysing the features we conclude that vowel and phrase features and both uttering positions are useful.
• CCR, Sensitivity and Specificity (all above 80%) point out the potential of voice as a discriminating factor between healthy subjects and severe OSA
• Future work will focus on clinical validation results for a comprehensive body of new subjects, with an already trained classifier using this model.

This paper deals with the potential and limitations of using voice and speech processing to detect Obstructive Sleep Apnea (OSA). An extensive body of voice features has been extracted from patients who present various degrees of OSA as well as healthy controls. We analyse the utility of a reduced set of features for detecting OSA. We apply various feature selection and reduction schemes (statistical ranking, Genetic Algorithms, PCA, LDA) and compare various classifiers (Bayesian Classifiers, kNN, Support Vector Machines, neural networks, Adaboost). S-fold crossvalidation performed on 248 subjects shows that in the extreme cases (that is, 127 controls and 121 patients with severe OSA) voice alone is able to discriminate quite well between the presence and absence of OSA. However, this is not the case with mild OSA and healthy snoring patients where voice seems to play a secondary role. We found that the best classification schemes are achieved using a Genetic Algorithm for feature selection/reduction.

Figure optionsDownload as PowerPoint slide