Novel sleep screening based on tracheal body sound and actigraphy

Abstract

The gold standard for assessment of most sleep disorders is the in-laboratory polysomnography (PSG). This approach produces high costs and inconveniences for patients due to its extensive setup, whereas alternative ambulatory systems are limited through reduced diagnostic abilities. The work presented here, therefore, aims to develop and validate a new, reliable, and simplified ambulant sleep monitor, utilizing tracheal body sound and movement data to automatically diagnose obstructive sleep apnea (OSA), one of the most common sleep disorders. To further improve the diagnostic ability of this monitor, automated sleep staging should be performed by utilizing body sound to extract cardiorespiratory features and actigraphy to extract movement features.

The main criteria to indicate the severity of OSA is the apnea-hypopnea index (AHI). Therefore, a new algorithm for the automated calculation of AHI was developed. For validation, the data of 60 subjects was recorded at the University Hospital Ulm. Subjects underwent a full-night screening using PSG and the new monitoring system concurrently. The AHI was scored blindly by a medical technician using PSG (AHIPSG) and by the automated algorithm (AHIest). AHIest strongly correlates with AHIPSG (r2=0.9871). A mean ±1.96 SD difference between AHIest and AHIPSG of 1.2 ± 5.14 is achieved. In terms of classifying subjects into groups of mild, moderate and severe sleep apnea, the evaluated new sleep monitor shows a strong correlation with the results obtained by PSG (Cohen’s Kappa > 0.81). These results clearly outperform similar approaches which were previously used. Additionally, a linear discriminant classifier was used to perform automated sleep staging using the new sleep monitor. The classifier achieved 86.9% accuracy with a Kappa of 0.69 for sleep/wake classification, 76.3% accuracy with a Kappa of 0.42 for wake/REM/NREM classification and 56.5% accuracy with a Kappa of 0.36 for wake/REM/light sleep/deep sleep classification. For the calculation of sleep efficiency (SE) a coefficient of determination r2 of 0.78 was reached. Here, subjects were also classified into groups of SEs (SE ≥ 40%, SE ≥ 60% and SE ≥ 80%). A Cohen’s Kappa > 0.61 was achieved for all groups.

The proposed sleep monitor accurately estimates AHI and diagnoses sleep apnea and its severity reliably. Furthermore, the monitor provides good performance in sleep/wake and wake/REM/NREM sleep staging while maintaining a simple setup and offering high comfort.