A-phase classification using convolutional neural networks.

A series of short events, called A-phases, can be observed in the human electroencephalogram (EEG) during Non-Rapid Eye Movement (NREM) sleep. These events can be classified in three groups (A1, A2, and A3) according to their spectral contents, and are thought to play a role in the transitions between the different sleep stages. A-phase detection and classification is usually performed manually by a trained expert, but it is a tedious and time-consuming task. In the past two decades, various researchers have designed algorithms to automatically detect and classify the A-phases with varying degrees of success, but the problem remains open. In this paper, a different approach is proposed: instead of attempting to design a general classifier for all subjects, we propose to train ad-hoc classifiers for each subject using as little data as possible, in order to drastically reduce the amount of time required from the expert. The proposed classifiers are based on deep convolutional neural networks using the log-spectrogram of the EEG signal as input data. Results are encouraging, achieving average accuracies of 80.31% when discriminating between A-phases and non A-phases, and 71.87% when classifying among A-phase sub-types, with only 25% of the total A-phases used for training. When additional expert-validated data is considered, the sub-type classification accuracy increases to 78.92%. These results show that a semi-automatic annotation system with assistance from an expert could provide a better alternative to fully automatic classifiers. Graphical abstract A/N Deep Learning Classifier.

A hemodynamic network involving the insula, the cingulate, and the basal forebrain correlates with EEG synchronization phases of sleep instability.

The cyclic alternating pattern (CAP) encompasses the pseudoperiodic appearance of synchronized brain waves and rhythms and is considered a regulator of the nonrapid eye movement (NREM) sleep vigilance level, reflecting sleep instability. To determine the brain regions responsible for this phenomenon, we scored and analyzed sleep functional magnetic resonance imaging data acquired with simultaneous electroencephalography (EEG-fMRI). Group analysis revealed a set of brain areas showing statistically significant blood oxygen-level dependent signal correlated positively with the synchronization phase of the CAP, most prominent being the insula, the middle cingulate gyrus, and the basal forebrain. These areas may form a network acting as a synchronization pacemaker, controlling the level of NREM sleep vigilance and the sleeper's arousability.

Visual and automatic classification of the cyclic alternating pattern in electroencephalography during sleep

Cyclic alternating pattern (CAP) is a neurophysiological pattern that can be visually scored by international criteria. The aim of this study was to verify the feasibility of visual CAP scoring using only one channel of sleep electroencephalogram (EEG) to evaluate the inter-scorer agreement in a variety of recordings, and to compare agreement between visual scoring and automatic scoring systems. Sixteen hours of single-channel European data format recordings from four different sleep laboratories with either C4-A1 or C3-A2 channels and with different sampling frequencies were used in this study. Seven independent scorers applied visual scoring according to international criteria. Two automatic blind scorings were also evaluated. Event-based inter-scorer agreement analysis was performed. The pairwise inter-scorer agreement (PWISA) was between 55.5 and 84.3%. The average PWISA was above 60% for all scorers and the global average was 69.9%. Automatic scoring systems showed similar results to those of visual scoring. The study showed that CAP could be scored using only one EEG channel. Therefore, CAP scoring might also be integrated in sleep scoring features and automatic scoring systems having similar performances to visual sleep scoring systems.

Non-REM Sleep Instability in Children With Primary Monosymptomatic Sleep Enuresis.

STUDY OBJECTIVES: Sleep enuresis is one of the most common sleep disturbances in childhood. Parental perception of deeper sleep in children with sleep enuresis is not confirmed by objective studies. However, evidence of disturbed sleep has been demonstrated by questionnaire, actigraphy, and polysomnographic studies, but no neurophysiological correlation with low arousability has been found. The goal of this study was to analyze the sleep microstructure of children with sleep enuresis using cyclic alternating pattern (CAP) analysis. METHODS: Forty-nine children were recruited, 27 with enuresis (19 males and 8 females, mean age 9.78 years, 2.52 standard deviation) and 22 normal control patients (11 males and 11 females, mean age 10.7 years, 3.43 standard deviation); all subjects underwent clinical evaluation followed by a full-night polysomnographic recording. Psychiatric, neurological, respiratory, and renal diseases were excluded. RESULTS: No differences in sex, age, and apnea-hypopnea index were noted in the patients with enuresis and the control patients. Sleep stage architecture in children with sleep enuresis showed a decrease in percentage of stage N3 sleep. CAP analysis showed an increase in CAP rate in stage N3 sleep and in phase A1 index during stage N3 sleep in the sleep enuresis group, but also a significant reduction of A2% and A3% and of phases A2 and A3 indexes, supporting the concept of decreased arousability in patients with sleep enuresis. The decrease of phase A2 and A3 indexes in our patients might reflect the impaired arousal threshold of children with sleep enuresis. Sleep fragmentation might result in a compensatory increase of slow wave activity (indicated by the increase of CAP rate in stage N3 sleep) and may explain the higher arousal threshold (indicated by a decrease of phase A2 and A3 indexes) linked to an increased sleep pressure. CONCLUSIONS: The findings of this study indicate the presence of a significant disruption of sleep microstructure (CAP) in children with sleep enuresis, supporting the hypothesis of a higher arousal threshold.

Sleep cyclic alternating pattern in otherwise healthy overweight school-age children.

STUDY OBJECTIVES: To compare sleep microstructure (cyclic alternating pattern, CAP) characteristics in otherwise healthy overweight (OW) and normal weight (NW) children. DESIGN: Polysomnographic cross-sectional study. SETTING: Sleep laboratory. PARTICIPANTS: Fifty-eight (26 NW and 32 OW) 10-year-old children. INTERVENTIONS: N/A. MEASUREMENTS AND RESULTS: Participants were part of a longitudinal study beginning in infancy and free of sleep disorders. Groups were based on body-mass index (BMI) z-score. From polysomnographic overnight recordings, sleep-waking states were scored according to international criteria. CAP analysis was performed visually during NREM sleep. Conventional sleep parameters were similar between groups. BMI was positively related to CAP rate and CAP sequences but inversely related to CAP B phase duration. Differences between groups were confined to slow-wave sleep (SWS), with OW children showing higher CAP rate, CAP cycles, and CAP A1 number and index and shorter CAP cycles and B phase duration. They also showed more CAP class intervals shorter than 30 s, and a suggestive trend for fewer intervals longer than 30 s. CONCLUSIONS: Cyclic alternating pattern characteristics in children related to nutritional status and were altered in overweight subjects during slow-wave sleep. We suggest that the more frequent oscillatory pattern of electroencephalographic slow activity in overweight subjects might reflect less stable slow-wave sleep episodes.

Avaliação da arquitetura do sono em crianças com epilepsia refratária / Evaluation of sleep architecture in children with refractory epilepsy

INTRODUÇÃO: Há um interesse crescente nas relações entre sono e epilepsia incentivado pela compreensão de que existem interações potencialmente relevantes nas duas direções. Embora o papel do sono na hipersincronização e a crescente preocupação na geração de crises sejam bem conhecidos, o grau no qual o sono pode facilitar ou induzir a um fenômeno epileptogênico, nas epilepsias lesionais, permanece indefinido. As epilepsias lesionais parecem apresentar um mecanismo particular de epileptogenicidade e o esclarecimento do papel da macro e microarquitetura do sono pode auxiliar na antecipação e monitorização de fenômenos epilépticos relacionados ao sono, conforme a etiologia da epilepsia. OBJETIVO: revisar e discutir as relações entre sono e epilepsia na infância e adolescência relacionando as alterações estruturais do sono à etiologia da epilepsia. MÉTODOS: revi-são bibliográfica utilizando o banco de dados Medline, abrangendo os estudos publicados nos últimos quinze anos, com as palavras-chave (unitermos) sono e epilepsia. CONCLUSÕES: epilepsia refratária durante a infância parece influenciar a organização do sono principalmente naqueles pacientes com etiologia lesional. A definição do tipo de epilepsia pode ser importante na antecipação dos distúrbios de sono nesta população.

ABSTRACT INTRODUCTION: There has been a growing interest in the relations between sleep and epilepsy, kindled by the realization that there are many potentially relevant two-way interactions. Even though the hyper-synchronizing role of sleep and its attending increase in the probability of seizure generation are well known, the degree to which sleep may facilitate or induce epileptogenic phenomena in lesional epilepsies remains unclear. The lesional epilepsies seems to have intrinsic epileptogenic properties and the knowledge about sleep macro and microarchiteture could help clinician to anticipate and monitor sleep-related epileptic phenomena according to the etiology of the epilepsy OBJECTIVE: Discuss the relationship between sleep and epilepsy in childhood and adolescence. METHODS: Literature review in journals indexed through Medline, from the last 15 years. CONCLUSION: Refractory epilepsy during childhood influences sleep organization mainly in patients with lesional etiology. The definition of the type of epilepsy is important to anticipate sleep disorders in this population.

Cyclic alternating pattern in normal children aged 12 to 24 months / Padrão alternante cíclico em crianças normais de 12 a 24 meses

OBJECTIVE: The aim of this study is to complement existing data on the expression and characteristics of cyclic alternating pattern (CAP) in children, specifically in the 12 to 24 month age bracket. METHOD: Descriptive study. Settings: a university pediatric sleep laboratory. Participants: Twelve normal and healthy subjects (6 girls and 6 boys, mean age 18.9±4.72 months; range 12-24 months underwent a standard polysomnography night recording in our pediatric sleep laboratory. Sleep stages and CAP were analyzed according to standard international rules. RESULTS: CAP rate found in children of 12 to 24 months of age was (25.78±10.18 percent) and it is characterized by a linear increase of 2 percent per month, from 12 percent at 12 months reaching 35 percent at 24 months. With coefficient of determination R² of 0.91. The duration of A phases was 6.93±1.06 seconds, and B phases was found to last 21.44±2.31 seconds. The number of CAP cycles was 173.25±73.85 with an average index per hour of 33.55±14.61. The number of CAP sequences reached 25.25±9.55 per recording. CAP rate for the different type of A phases are (21.83±9.68 percent), for A1, (2.43±2.30 percent) for A3 and (1.67±1.11 percent) for A2. CONCLUSION: Our study provides normative data on CAP in a group of young children (12 to 24 months). The most salient result of this study is the strong correlation of CAP rate of 2 percent per month in this age group.

OBJETIVO: O objetivo deste estudo é complementar os dados normativos sobre a expressão e as características do padrão alternante cíclico (CAP) em crianças, especificamente na faixa etária de 12 para 24 meses. MÉTODO: Estudo descritivo. Participantes: Doze indivíduos normais e saudáveis (6 meninas e 6 meninos, entre 12-24 meses de idade (média de 18,9±4,72 meses), foram submetidos a uma noite de gravação de polissonografia em nosso laboratório de sono pediátrico. As fases do sono e CAP foram analisados de acordo com padrão e normas internacionais. RESULTADOS: A taxa máxima encontrada em crianças de 12 a 24 meses de idade foi de (25,78 por cento±10,18) e é caracterizada por um aumento linear de 2 por cento ao mês, a partir de 12 por cento em 12 meses atingindo 35 por cento em 24 meses de idade. A duração da fase A foi de 6,93±1,06 segundos, e nas fases B foi encontrado a média de 21,44±2,31 segundos. O número de ciclos CAP foi 173,25±73,85, com um índice médio por hora de 33,55±14,61. O número de sequências CAP chegou a 25,25±9,55 por registro.Os índices dos subtipos de fases A foram (21,83 por cento±9,68), para a A1, (2,43 por cento±2,30) para a A3 e (1,67 por cento±1,11) para A2. CONCLUSÃO: Nosso estudo fornece dados normativos sobre CAP em um grupo de crianças (12 a 24 meses de idade). O resultado mais saliente deste estudo é a forte correlação entre taxa do CAP e idade, mostrando que a cada mês ocorreu o aumento de 2 por cento.

Delta sleep instability in children with chronic arthritis

The objective of the present study was to evaluate the expression of a cyclic alternating pattern (CAP) in slow wave sleep (SWS) in children with the well-defined chronic syndrome juvenile idiopathic arthritis (JIA). Twelve patients (9-17 years of age), 7 girls, with JIA were compared to matched controls by age, pubertal stage and gender. After one night of habituation in the sleep laboratory, sleep measurements were obtained by standard polysomnography with conventional sleep scoring and additional CAP analyses. The sleep parameters of the JIA and control groups were similar for sleep efficiency (91.1 ± 6.7 vs 95.8 ± 4.0), sleep stage in minutes: stage 1 (16.8 ± 8.5 vs 17.8 ± 4.0), stage 2 (251.9 ± 41 vs 262.8 ± 38.1), stage 3 (17.0 ± 6.0 vs 15.1 ± 5.7), stage 4 (61.0 ± 21.7 vs 77.1 ± 20.4), and rapid eye movement sleep (82.0 ± 27.6 vs 99.0 ± 23.9), respectively. JIA patients presented nocturnal disrupted sleep, with an increase in short awakenings, but CAP analyses showed that sleep disruption was present even during SWS, showing an increase in the overall CAP rate (P < 0.01). Overall CAP rate during non-rapid eye movement sleep was significantly higher in pediatric patients who were in chronic pain. This is the first study of CAP in pediatric patients with chronic arthritis showing that CAP analyses can be a powerful tool for the investigation of disturbance of SWS in children, based on sleep EEG visual analysis.