Automated sleep stage classification based on tracheal body sound and actigraphy

Abstract

The current gold standard for assessment of most sleep disorders is the in-laboratory polysomnography (PSG). This approach produces high costs and inconveniences for the patients. An accessible and simple preliminary screening method to diagnose the most common sleep disorders and to decide whether a PSG is necessary or not is therefore desirable. A minimalistic type-4 monitoring system which utilized tracheal body sound and actigraphy to accurately diagnose the obstructive sleep apnea syndrome was previously developed. To further improve the diagnostic ability of said system, this study aims to examine if it is possible to perform automated sleep staging utilizing body sound to extract cardiorespiratory features and actigraphy to extract movement features. A linear discriminant classifier based on those features was used for automated sleep staging using the type-4 sleep monitor. For validation 53 subjects underwent a full-night screening at Ulm University Hospital using the developed sleep monitor in addition to polysomnography. To assess sleep stages from PSG, a trained technician manually evaluated EEG, EOG, and EMG recordings. The classifier reached 86.9% accuracy and a Kappa of 0.69 for sleep/wake classification, 76.3% accuracy and a Kappa of 0.42 for Wake/REM/NREM classification, and 56.5% accuracy and 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 is reached. Additionally, subjects were classified into groups of SEs (SE≥40%, SE≥60% and SE≥80%). A Cohen's Kappa >0.61 was reached for all groups, which is considered as substantial agreement. The presented method provides satisfactory performance in sleep/wake and wake/REM/NREM sleep staging while maintaining a simple setup and offering high comfort. This minimalistic approach may address the need for a simple yet reliable preliminary sleep screening in an ambulatory setting.

Keywords: monitoring; movement analysis; respiratory sounds; sleep staging.

Table 1. Anthropometric information of the subjects

Table 2. List of all features used for automated sleep staging

Table 3. Evaluation of the 2-, 3-, 4-stage classifier performance for sleep staging

Table 4. System performance based on the evaluation of the subject classification into different groups of sleep efficiency (SE) including all 53 subjects

Table 5. Overview of comparable sleep staging results found in literature

Figure 1. Abstract illustration of the setup of the new sleep monitor system.

The microphone is attached on the subject’s neck in close vicinity to the trachea. The remaining hardware is attached to the existing thoracic belt of the respiratory inductance plethysmograph of the polysomnography.

Figure 2. Illustration of the signal processing steps of the tracheal body sound for respiratory feature extraction;

(A) raw audio signal; (B) audio signal after FIR-filtering and spectral subtraction; (C) estimation of airflow, values below the horizontal line are considered no breathing.

Figure 3. Illustration of the signal processing steps using the tracheal body sound for cardiac feature extraction;

(A) raw audio signal, (B) audio signal after filtering in frequency domain; stars mark detected peaks, circles mark detected heart beat consisting of two peaks.

Figure 4. Relationship between sleep efficiency of the new sleep monitor (SE_est) and the sleep efficiency of the polysomnography (SE_PSG);

r2: coefficient of determination; n: number of data points

Figure 5. Relationship between wake after onset (WASO), total wake time (TWT) and total sleep time (TST) of the new sleep monitor and the polysomnography including all 53 subjects;

r²: coefficient of determination

Figure 6. Receiver operating characteristic (ROC) of subject classification into groups of sleep efficiency (SE>40%, >60%, >80%); AUC: area under curve