The diagnostic value of power spectra analysis of the sleep electroencephalography in narcoleptic patients

A Review on Automated Sleep Study

In recent years, research on automated sleep analysis has witnessed significant growth, reflecting advancements in understanding sleep patterns and their impact on overall health. This review synthesizes findings from an exhaustive analysis of 87 papers, systematically retrieved from prominent databases such as Google Scholar, PubMed, IEEE Xplore, and ScienceDirect. The selection criteria prioritized studies focusing on methods employed, signal modalities utilized, and machine learning algorithms applied in automated sleep analysis. The overarching goal was to critically evaluate the strengths and weaknesses of the proposed methods, shedding light on the current landscape and future directions in sleep research. An in-depth exploration of the reviewed literature revealed a diverse range of methodologies and machine learning approaches employed in automated sleep studies. Notably, K-Nearest Neighbors (KNN), Ensemble Learning Methods, and Support Vector Machine (SVM) emerged as versatile and potent classifiers, exhibiting high accuracies in various applications. However, challenges such as performance variability and computational demands were observed, necessitating judicious classifier selection based on dataset intricacies. In addition, the integration of traditional feature extraction methods with deep structures and the combination of different deep neural networks were identified as promising strategies to enhance diagnostic accuracy in sleep-related studies. The reviewed literature emphasized the need for adaptive classifiers, cross-modality integration, and collaborative efforts to drive the field toward more accurate, robust, and accessible sleep-related diagnostic solutions. This comprehensive review serves as a solid foundation for researchers and practitioners, providing an organized synthesis of the current state of knowledge in automated sleep analysis. By highlighting the strengths and challenges of various methodologies, this review aims to guide future research toward more effective and nuanced approaches to sleep diagnostics.

Excessive daytime sleepiness is associated with relative delta frequency power among patients with mild OSA

Background

Excessive daytime sleepiness (EDS) is a cause of low quality of life among obstructive sleep apnoea (OSA) patients. Current methods of assessing and predicting EDS are limited due to time constraints or differences in subjective experience and scoring. Electroencephalogram (EEG) power spectral densities (PSDs) have shown differences between OSA and non-OSA patients, and fatigued and non-fatigued patients. Therefore, polysomnographic EEG PSDs may be useful to assess the extent of EDS among patients with OSA.

Methods

Patients presenting to Israel Loewenstein hospital reporting daytime sleepiness who recorded mild OSA on polysomnography and undertook a multiple sleep latency test. Alpha, beta, and delta relative powers were assessed between patients categorized as non-sleepy (mean sleep latency (MSL) ≥10 min) and sleepy (MSL <10 min).

Results

139 patients (74% male) were included for analysis. 73 (53%) were categorized as sleepy (median MSL 6.5 min). There were no significant differences in demographics or polysomnographic parameters between sleepy and non-sleepy groups. In multivariate analysis, increasing relative delta frequency power was associated with increased odds of sleepiness (OR 1.025 (95% CI 1.024-1.026)), while relative alpha and beta powers were associated with decreased odds. The effect size of delta PSD on sleepiness was significantly greater than that of either alpha or beta frequencies.

Conclusion

Delta PSD during polysomnography is significantly associated with a greater degree of objective daytime sleepiness among patients with mild OSA. Further research is needed to corroborate our findings and identify the direction of potential causal correlation between delta PSD and EDS.

A review of automated sleep disorder detection

Automated sleep disorder detection is challenging because physiological symptoms can vary widely. These variations make it difficult to create effective sleep disorder detection models which support hu-man experts during diagnosis and treatment monitoring. From 2010 to 2021, authors of 95 scientific papers have taken up the challenge of automating sleep disorder detection. This paper provides an expert review of this work. We investigated whether digital technology and Artificial Intelligence (AI) can provide automated diagnosis support for sleep disorders. We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines during the content discovery phase. We compared the performance of proposed sleep disorder detection methods, involving differ-ent datasets or signals. During the review, we found eight sleep disorders, of which sleep apnea and insomnia were the most studied. These disorders can be diagnosed using several kinds of biomedical signals, such as Electrocardiogram (ECG), Polysomnography (PSG), Electroencephalogram (EEG), Electromyogram (EMG), and snore sound. Subsequently, we established areas of commonality and distinctiveness. Common to all reviewed papers was that AI models were trained and tested with labelled physiological signals. Looking deeper, we discovered that 24 distinct algorithms were used for the detection task. The nature of these algorithms evolved, before 2017 only traditional Machine Learning (ML) was used. From 2018 onward, both ML and Deep Learning (DL) methods were used for sleep disorder detection. The strong emergence of DL algorithms has considerable implications for future detection systems because these algorithms demand significantly more data for training and testing when compared with ML. Based on our review results, we suggest that both type and amount of labelled data is crucial for the design of future sleep disorder detection systems because this will steer the choice of AI algorithm which establishes the desired decision support. As a guiding principle, more labelled data will help to represent the variations in symptoms. DL algorithms can extract information from these larger data quantities more effectively, therefore; we predict that the role of these algorithms will continue to expand.

Disrupted nighttime sleep and sleep instability in narcolepsy

STUDY OBJECTIVES

This review aimed to summarize current knowledge about disrupted nighttime sleep (DNS) and sleep instability in narcolepsy, including self-reported and objective assessments, potential causes of sleep instability, health consequences and functional burden, and management.

METHODS

One hundred two peer-reviewed publications from a PubMed search were included.

RESULTS

DNS is a key symptom of narcolepsy but has received less attention than excessive daytime sleepiness and cataplexy. There has been a lack of clarity regarding the definition of DNS, as many sleep-related symptoms and conditions disrupt sleep quality in narcolepsy (eg, hallucinations, sleep paralysis, rapid eye movement sleep behavior disorder, nightmares, restless legs syndrome/periodic leg movements, nocturnal eating, sleep apnea, depression, anxiety). In addition, the intrinsic sleep instability of narcolepsy results in frequent spontaneous wakings and sleep stage transitions, contributing to DNS. Sleep instability likely emerges in the setting of orexin insufficiency/deficiency, but its exact pathophysiology remains unknown. DNS impairs quality of life among people with narcolepsy, and more research is needed to determine its contributions to cardiovascular risk. Multimodal treatment is appropriate for DNS management, including behavioral therapies, counseling on sleep hygiene, and/or medication. There is strong evidence showing improvement in self-reported sleep quality and objective sleep stability measures with sodium oxybate, but rigorous clinical trials with other pharmacotherapies are needed. Treatment may be complicated by comorbidities, concomitant medications, and mood disorders.

CONCLUSIONS

DNS is a common symptom of narcolepsy deserving consideration in clinical care and future research.

CITATION

Maski K, Mignot E, Plazzi G, Dauvilliers Y. Disrupted nighttime sleep and sleep instability in narcolepsy. J Clin Sleep Med. 2022;18(1):289-304.

Artificial intelligence and sleep: Advancing sleep medicine

Artificial intelligence (AI) allows analysis of "big data" combining clinical, environmental and laboratory based objective measures to allow a deeper understanding of sleep and sleep disorders. This development has the potential to transform sleep medicine in coming years to the betterment of patient care and our collective understanding of human sleep. This review addresses the current state of the field starting with a broad definition of the various components and analytic methods deployed in AI. We review examples of AI use in screening, endotyping, diagnosing, and treating sleep disorders and place this in the context of precision/personalized sleep medicine. We explore the opportunities for AI to both facilitate and extend providers' clinical impact and present ethical considerations regarding AI derived prognostic information. We cover early adopting specialties of AI in the clinical realm, such as radiology and pathology, to provide a road map for the challenges sleep medicine is likely to face when deploying this technology. Finally, we discuss pitfalls to ensure clinical AI implementation proceeds in the safest and most effective manner possible.

Cortical arousal frequency is increased in narcolepsy type 1

STUDY OBJECTIVES

Hypocretin deficient narcolepsy (type 1, NT1) presents with multiple sleep abnormalities including sleep-onset rapid eye movement (REM) periods (SOREMPs) and sleep fragmentation. We hypothesized that cortical arousals, as scored by an automatic detector, are elevated in NT1 and narcolepsy type 2 (NT2) patients as compared to control subjects.

METHODS

We analyzed nocturnal polysomnography (PSG) recordings from 25 NT1 patients, 20 NT2 patients, 18 clinical control subjects (CC, suspected central hypersomnia but with normal cerebrospinal (CSF) fluid hypocretin-1 (hcrt-1) levels and normal results on the multiple sleep latency test), and 37 healthy control (HC) subjects. Arousals were automatically scored using Multimodal Arousal Detector (MAD), a previously validated automatic wakefulness and arousal detector. Multiple linear regressions were used to compare arousal index (ArI) distributions across groups. Comparisons were corrected for age, sex, body-mass index, medication, apnea-hypopnea index, periodic leg movement index, and comorbid rapid eye movement sleep behavior disorder.

RESULTS

NT1 was associated with an average increase in ArI of 4.02 events/h (p = 0.0246) compared to HC and CC, while no difference was found between NT2 and control groups. Additionally, a low CSF hcrt-1 level was predictive of increased ArI in all the CC, NT2, and NT1 groups.

CONCLUSIONS

The results further support the hypothesis that a loss of hypocretin neurons causes fragmented sleep, which can be measured as an increased ArI as scored by the MAD.

Sleep State Classification Using Power Spectral Density and Residual Neural Network with Multichannel EEG Signals

This paper proposes a classification framework for automatic sleep stage detection in both male and female human subjects by analyzing the electroencephalogram (EEG) data of polysomnography (PSG) recorded for three regions of the human brain, i.e., the pre-frontal, central, and occipital lobes. Without considering any artifact removal approach, the residual neural network (ResNet) architecture is used to automatically learn the distinctive features of different sleep stages from the power spectral density (PSD) of the raw EEG data. The residual block of the ResNet learns the intrinsic features of different sleep stages from the EEG data while avoiding the vanishing gradient problem. The proposed approach is validated using the sleep dataset of the Dreams database, which comprises of EEG signals for 20 healthy human subjects, 16 female and 4 male. Our experimental results demonstrate the effectiveness of the ResNet based approach in identifying different sleep stages in both female and male subjects compared to state-of-the-art methods with classification accuracies of 87.8% and 83.7%, respectively.

Increased EEG Theta Spectral Power in Sleep in Myotonic Dystrophy Type 1

STUDY OBJECTIVES

Myotonic dystrophy type 1 (DM1) is a multisystemic disorder that involves the central nervous system (CNS). Individuals with DM1 commonly present with sleep dysregulation, including excessive daytime sleepiness and sleep-disordered breathing. We aim to characterize electroencephalogram (EEG) power spectra from nocturnal polysomnography (PSG) in patients with DM1 compared to matched controls to better understand the potential CNS sleep dysfunction in DM1.

METHODS

A retrospective, case-control (1:2) chart review of patients with DM1 (n = 18) and matched controls (n = 36) referred for clinical PSG at the Stanford Sleep Center was performed. Controls were matched based on age, sex, apnea-hypopnea index (AHI), body mass index (BMI), and Epworth Sleepiness Scale (ESS). Sleep stage and respiratory metrics for the two groups were compared. Power spectral analysis of the EEG C3-M2 signal was performed using the fast Fourier transformation.

RESULTS

Patients with DM1 had significantly increased theta percent power in stage N2 sleep compared to matched controls. Theta/beta and theta/alpha percent power spectral ratios were found to be significantly increased in stage N2, N3, all sleep stages combined, and all wake periods combined in patients with DM1 compared to controls. A significantly lower nadir O2 saturation was also found in patients with DM1 versus controls.

CONCLUSIONS

Compared to matched controls, patients with DM1 had increased EEG theta spectral power. Increased theta/beta and theta/alpha power spectral ratios in nocturnal PSG may reflect DM1 pathology in the CNS.

Role of N-Arachidonoyl-Serotonin (AA-5-HT) in Sleep-Wake Cycle Architecture, Sleep Homeostasis, and Neurotransmitters Regulation

The endocannabinoid system comprises several molecular entities such as endogenous ligands [anandamide (AEA) and 2-arachidonoylglycerol (2-AG)], receptors (CB₁ and CB₂), enzymes such as [fatty acid amide hydrolase (FAHH) and monoacylglycerol lipase (MAGL)], as well as the anandamide membrane transporter. Although the role of this complex neurobiological system in the sleep–wake cycle modulation has been studied, the contribution of the blocker of FAAH/transient receptor potential cation channel subfamily V member 1 (TRPV1), N-arachidonoyl-serotonin (AA-5-HT) in sleep has not been investigated. Thus, in the present study, varying doses of AA-5-HT (5, 10, or 20 mg/Kg, i.p.) injected at the beginning of the lights-on period of rats, caused no statistical changes in sleep patterns. However, similar pharmacological treatment given to animals at the beginning of the dark period decreased wakefulness (W) and increased slow wave sleep (SWS) as well as rapid eye movement sleep (REMS). Power spectra analysis of states of vigilance showed that injection of AA-5-HT during the lights-off period diminished alpha spectrum across alertness in a dose-dependent fashion. In opposition, delta power spectra was enhanced as well as theta spectrum, during SWS and REMS, respectively. Moreover, the highest dose of AA-5-HT decreased wake-related contents of neurotransmitters such as dopamine (DA), norepinephrine (NE), epinephrine (EP), serotonin (5-HT) whereas the levels of adenosine (AD) were enhanced. In addition, the sleep-inducing properties of AA-5-HT were confirmed since this compound blocked the increase in W caused by stimulants such as cannabidiol (CBD) or modafinil (MOD) during the lights-on period. Additionally, administration of AA-5-HT also prevented the enhancement in contents of DA, NE, EP, 5-HT and AD after CBD of MOD injection. Lastly, the role of AA-5-HT in sleep homeostasis was tested in animals that received either CBD or MOD after total sleep deprivation (TSD). The injection of CBD or MOD increased alertness during sleep rebound period after TSD. However, AA-5-HT blocked this effect by allowing animals to display an enhancement in sleep across sleep rebound period. Overall, our findings provide evidence that AA-5-HT is an important modulator of sleep, sleep homeostasis and neurotransmitter contents.

Diagnostic value of sleep stage dissociation as visualized on a 2-dimensional sleep state space in human narcolepsy

BACKGROUND

Type 1 narcolepsy (NT1) is characterized by symptoms believed to represent Rapid Eye Movement (REM) sleep stage dissociations, occurrences where features of wake and REM sleep are intermingled, resulting in a mixed state. We hypothesized that sleep stage dissociations can be objectively detected through the analysis of nocturnal Polysomnography (PSG) data, and that those affecting REM sleep can be used as a diagnostic feature for narcolepsy.

NEW METHOD

A Linear Discriminant Analysis (LDA) model using 38 features extracted from EOG, EMG and EEG was used in control subjects to select features differentiating wake, stage N1, N2, N3 and REM sleep. Sleep stage differentiation was next represented in a 2D projection. Features characteristic of sleep stage differences were estimated from the residual sleep stage probability in the 2D space. Using this model we evaluated PSG data from NT1 and non-narcoleptic subjects. An LDA classifier was used to determine the best separation plane.

COMPARISON WITH EXISTING METHODS

This method replicates the specificity/sensitivity from the training set to the validation set better than many other methods.

RESULTS

Eight prominent features could differentiate narcolepsy and controls in the validation dataset. Using a composite measure and a specificity cut off 95% in the training dataset, sensitivity was 43%. Specificity/sensitivity was 94%/38% in the validation set. Using hypersomnia subjects, specificity/sensitivity was 84%/15%. Analyzing treated narcoleptics the specificity/sensitivity was 94%/10%.

CONCLUSION

Sleep stage dissociation can be used for the diagnosis of narcolepsy. However the use of some medications and presence of undiagnosed hypersomnolence patients impacts the result.