Sleep and Wake Classification With Actigraphy and Respiratory Effort Using Dynamic Warping

Abstract

This paper proposes the use of dynamic warping (DW) methods for improving automatic sleep and wake classification using actigraphy and respiratory effort. DW is an algorithm that finds an optimal nonlinear alignment between two series allowing scaling and shifting. It is widely used to quantify (dis)similarity between two series. To compare the respiratory effort between sleep and wake states by means of (dis)similarity, we constructed two novel features based on DW. For a given epoch of a respiratory effort recording, the features search for the optimally aligned epoch within the same recording in time and frequency domain. This is expected to yield a high (or low) similarity score when this epoch is sleep (or wake). Since the comparison occurs throughout the entire-night recording of a subject, it may reduce the effects of within- and between-subject variations of the respiratory effort, and thus help discriminate between sleep and wake states. The DW-based features were evaluated using a linear discriminant classifier on a dataset of 15 healthy subjects. Results show that the DW-based features can provide a Cohen's Kappa coefficient of agreement κ = 0.59 which is significantly higher than the existing respiratory-based features and is comparable to actigraphy. After combining the actigraphy and the DW-based features, the classifier achieved a κ of 0.66 and an overall accuracy of 95.7%, outperforming an earlier actigraphy- and respiratory-based feature set ( κ = 0.62). The results are also comparable with those obtained using an actigraphy- and cardiorespiratory-based feature set but have the important advantage that they do not require an ECG signal to be recorded.

Actigraphy-based sleep/wake detection for insomniacs

“…Although the addition of the proposed actigraphic feature DHAL improved the performance in sleep/wake detection for insomnia subjects, it is still inferior to that reported for healthy subjects (kappa value of 0.58 in [13]). As shown in Figure 2, the main challenge of actigraphy-based sleep/wake detection is to separate wakefulness with low activity and sleep in which a similar level of activity was found.…”

“…As shown in Figure 2, the main challenge of actigraphy-based sleep/wake detection is to separate wakefulness with low activity and sleep in which a similar level of activity was found. In fact, in order to improve the detection performance, many previous studies have shown the effectiveness of incorporating cardiac and respiratory features in addition to actigraphy for healthy subjects [11], [13], [16], [18]. These features exploit the relation between sleep stages and autonomic cardiorespiratory activity [19], [20].…”

“…It was computed by counting frequency of each epoch being scored as each state (sleep or wake). The classifier used in this work was the same as that used in our previous work [13], [16].…”

“…In the past decade, a relatively high performance of sleep/wake detection using actigraphy has been achieved for healthy subjects without sleep disturbances [11], [12], [13], [14]. However, this is not yet the case for subjects with insomnia who have difficulty of falling or maintaining sleep [6], causing increased wake periods and more 'active' sleep epochs.…”

“…We use a linear discriminant-based classifier, which has been extensively used for sleep/wake detection in previous studies [11], [13], [16]. …”