How many sleep stages do we need for an efficient automatic insomnia diagnosis?

Abstract

Tools used by clinicians to diagnose and treat insomnia typically include sleep diaries and questionnaires. Overnight polysomnography (PSG) recordings are used when the initial diagnosis is uncertain due to the presence of other sleep disorders or when the treatment, either behavioral or pharmacologic, is unsuccessful. However, the analysis and the scoring of PSG data are time-consuming. To simplify the diagnosis process, in this paper we have proposed an efficient insomnia detection algorithm based on a central single electroencephalographic (EEG) channel (C3) using only deep sleep. We also analyzed several spectral and statistical EEG features of good sleeper controls and subjects suffering from insomnia in different sleep stages to identify the features that offered the best discrimination between the two groups. Our proposed algorithm was evaluated using EEG recordings from 19 patients diagnosed with primary insomnia (11 females, 8 males) and 16 matched control subjects (11 females, 5 males). The sensitivity of our algorithm is 92%, the specificity is 89.9%, the Cohen's kappa is 0.81 and the agreement is 91%, indicating the effectiveness of our proposed method.

Automatic Identification of Insomnia Based on Single-Channel EEG Labelled With Sleep Stage Annotations

“…Previous studies demonstrated that relative power and Hjorth parameters of EEG signal are important features for insomnia identification [29]- [31]. According to the method in [13], EEG was filtered at 6 frequency bands: delta (0.5-4Hz), theta (4-8Hz), alpha (8-12Hz), sigma (12-16Hz), beta (16-30Hz) and gamma . A total of 22 features including power and Hjorth parameters were extracted.…”

“…where x is the EEG epoch. According to the method in literature [13], we leveraged the epochs of SWS sleep stages in C4-A1 channel EEG for implementing this method. Then we applied PCA for dimensionality reduction and leveraged the first principal component as the final one-dimensional feature.…”

“…S. Aydin et al [12] extracted the 10-dimensional singular spectrum features of the sleep EEG signal, and then fed the features into a single hidden layer artificial neural network (ANN) for insomnia identification. HAMIDA et al [13] extracted spectral and Hjorth's parameters features, and applied principal component analysis (PCA) for dimension reduction. The first principal component was then classified according to the set threshold.…”

“…Several studies have investigated the insomnia identification performance based on different sleep stages of EEG [13], [14]. HAMIDA et al [13] evaluated the performance of wake, S1, S2, SWS and REM stages, finding SWS epochs have the best insomnia identification performance.…”

“…Several studies have investigated the insomnia identification performance based on different sleep stages of EEG [13], [14]. HAMIDA et al [13] evaluated the performance of wake, S1, S2, SWS and REM stages, finding SWS epochs have the best insomnia identification performance. M. Shahin et al [14] compared the performance of all stages, NREM, S2+S3, NREM+REM stages, finding NREM+REM epochs have the best performance, whereas they did not evaluate the performance only leveraging REM epochs.…”