Artificial Intelligence Models for the Automation of Standard Diagnostics in Sleep Medicine-A Systematic Review.

Sleep disorders, prevalent in the general population, present significant health challenges. The current diagnostic approach, based on a manual analysis of overnight polysomnograms (PSGs), is costly and time-consuming. Artificial intelligence has emerged as a promising tool in this context, offering a more accessible and personalized approach to diagnosis, particularly beneficial for under-served populations. This is a systematic review of AI-based models for sleep disorder diagnostics that were trained, validated, and tested on diverse clinical datasets. An extensive search of PubMed and IEEE databases yielded 2114 articles, but only 18 met our stringent selection criteria, underscoring the scarcity of thoroughly validated AI models in sleep medicine. The findings emphasize the necessity of a rigorous validation of AI models on multimodal clinical data, a step crucial for their integration into clinical practice. This would be in line with the American Academy of Sleep Medicine's support of AI research.

Sleep duration and all-cause mortality among stroke survivors.

BACKGROUND: Post stroke sleep duration could increase the risk of death. This study tested the hypothesis that inadequate sleep duration is associated with increased mortality among stroke survivors. METHODS: The REasons for Geographic And Racial Differences in Stroke (REGARDS), a national population-based longitudinal study, was the data source. Sleep duration was ascertained between 2013 and 2016 among stroke survivors who were subsequently followed up until death or censored on December 31, 2022. Sleep duration was estimated as the difference between wake-up time and bedtime to which was subtracted the time spent in bed without sleep. Cox proportional hazards regression models were employed to investigate the association between sleep duration and all-cause mortality adjusting for demographic factors, socioeconomic factors, behavioral factors, and co-morbidities. RESULTS: A total of 468 non-Hispanic Black and White stroke survivors were included in this analysis. The mean age was 76.3 years, 52.6% were females and 56.0% were non-Hispanic White individuals. The distribution of short (≤6 h), adequate (7.0-8.9 h), and long sleep (≥9 h) was 30.3%, 44.7%, and 25%, respectively. Over a mean follow-up of 5.0 years, 190 (40.6%) deaths occurred. Compared to stroke survivors with adequate sleep (7.0-8.9 h), stroke survivors with long sleep (≥9 h) were at increased risk of all-cause mortality (HR=1.46, 95% CI=1.01, 2.12). However, short sleep (≤6 h) was not significantly associated with an increased risk of all-cause mortality (HR=1.31, 95% CI=0.90, 1.91). Subgroup analyses indicated higher risk in the age <75 years, females, non-Hispanic Black individuals, and those living in the Stroke Belt region, but those differences were not statistically significant. CONCLUSION: In this study of stroke survivors, 9 hours or more of sleep per day was associated with an increased risk of all-cause mortality. This finding suggests that excessive sleep duration may be a warning sign of poor life expectancy in stroke survivors.

Association Between Insomnia Symptoms and Trajectory With the Risk of Stroke in the Health and Retirement Study.

BACKGROUND AND OBJECTIVES: Insomnia is a common condition affecting more than a third of the US population. However, the link between insomnia symptoms and stroke is understudied and the underlying mechanism remains unclear. This study aimed to investigate the relationship between insomnia symptoms and the incidence of stroke. METHODS: The Health and Retirement Study, a survey of Americans older than 50 years and their spouses, from 2002 to 2020 was used as the data source. Only those who were stroke-free at baseline were included in this study. The exposure variable was insomnia symptoms and was derived from self-reported sleep-related factors including difficulty initiating sleep, difficulty maintaining sleep, waking up too early, and nonrestorative sleep. Repeated-measures latent class analysis was used to identify insomnia trajectories over time. To investigate the relationship between insomnia symptoms and stroke events reported during the follow-up period, Cox proportional hazards regression models were used. Mediation analyses of comorbidities were performed using causal mediation within a counterfactual framework. RESULTS: A total of 31,126 participants were included with a mean follow-up of 9 years. The mean age was 61 years (SD = 11.1) and 57% were females. Insomnia symptom trajectories remained constant over time. Compared with those with no insomnia symptoms, an increased risk of stroke was observed for those with insomnia symptom scores ranging from 1 to 4 and 5 to 8 (hazard ratio (HR) = 1.16, 95% confidence interval (CI) 1.02-1.33) and (HR = 1.51, 95% CI 1.29-1.77), respectively, indicating a dose-response relationship. The association was stronger in participants younger than 50 years (HR = 3.84, 95% CI 1.50-9.85) than in those aged 50 years and older (HR = 1.38, 95% CI 1.18-1.62), comparing those with insomnia symptoms ranging from 5 to 8 with those with no insomnia symptoms. This association was mediated by diabetes, hypertension, heart disease, and depression. DISCUSSION: Insomnia symptoms were associated with an increased risk of stroke, especially in adults younger than 50 years, and the risk was mediated by certain comorbidities. Increased awareness and management of insomnia symptoms may contribute to the prevention of stroke occurrence.

Gastroesophageal reflux disease as an etiology of sleep disturbance in subjects with insomnia and minimal reflux symptoms: a pilot study of prevalence and response to therapy.

BACKGROUND: Gastroesophageal reflux disease (GERD) is a well-recognized cause of impaired sleep in patients with frequent GERD symptoms, as well as those with sleep apnea. GERD's role in sleep disturbance of minimally symptomatic patients with poor sleep quality is less clear. AIM: We aimed to define the prevalence of GERD-related sleep disturbance in minimally-symptomatic subjects with demonstrated insomnia, and to assess the changes in sleep efficiency in these subjects after vigorous acid suppression. METHODS: We recruited subjects aged 18-75 years reporting at least 6 months of insomnia, and sleep difficulty at least three nights per week. Subjects with a BMI > 30, a history of snoring or ongoing use of proton pump inhibitor or H2 receptor antagonist were excluded. Subjects underwent concurrent sleep study with dual channel 24-h pH study. Sleep efficiency, defined as the percentage of time after sleep initiation that the subject actually slept, and spontaneous arousal index, defined as the number of arousals per hour, were calculated. Those with a sleep study demonstrating poor sleep quality (sleep efficiency of < 83%, and > 10 arousals/h for those aged < 45, and > 15 for those who were 45 or older) and no obstructive sleep apnea were treated with rabeprazole 20 mg PO BID x 14 days. After 14 days, the subjects underwent repeat sleep study with pH monitoring. The GERD Symptom Assessment Scale (GSAS), the Epworth Sleepiness Scale (ESS) and the Functional Outcomes of Sleep Questionnaire (FOSQ) were administered to subjects at study inception and after 2 weeks of therapy. RESULTS: Twenty-four subjects reporting insomnia were enrolled, and 20 met criteria for disordered sleep and no OSA. Seventeen completed both the first and second studies, and 16 were adequate for analysis. Baseline GSAS demonstrated trivial or no reflux symptoms in the cohort (no subject scored > 8 out of 45 on GSAS, corresponding to a median rating of reflux symptoms of "not at all"). Four of 16 subjects (25%) demonstrated abnormal pH studies at baseline. All four had normalization of acid exposures on PPI. After 2 weeks of treatment, three of these four subjects had normalization of sleep efficiency, compared to 4 of 12 of the subjects with normal Johnson-DeMeester scores. Repeated measures analysis showed significant improvement in spontaneous arousal index between the first and second study for the whole group (P < 0.0035). Pre- and post-therapy ESS and FOSQ scores were not significantly different. CONCLUSIONS: Despite the lack of GERD symptoms, a significant minority of subjects with sleep disturbance have abnormal acid exposures. These preliminary data suggest that aggressive treatment of GERD in such patients may result in improvement in sleep efficiency.

Sleep problems in primary care: a North Carolina Family Practice Research Network (NC-FP-RN) study.

BACKGROUND: The prevalence and nature of sleep disorders in primary care has not been widely studied. As part of a survey conducted in 5 family practice offices in North Carolina, we screened adult patients for sleep syndromes and sought to ascertain which demographic status and health status were associated with these disorders. METHODS: We approached 2963 consecutive adults who presented for office visits to the 5 study practices. The 4-page study questionnaire, which was available in English and Spanish, included items on insomnia, excessive daytime sleepiness, obstructive sleep apnea syndrome, and restless legs syndrome. Analyses evaluated the relationship between sleep syndromes and demographic factors, health status, and disability. RESULTS: We enrolled 1935 patients (65.3% response rate). More than half reported excessive daytime sleepiness, one third had insomnia, more than 25% had symptoms of restless legs syndrome, and 13% to 33% reported obstructive sleep apnea syndrome symptoms. Participants who rated their health as poor reported significantly higher rates of all sleep disturbance items. Patients with hypertension, pain syndromes, and depression had a significantly increased risk for all sleep complaints. Patients who reported limited activity had a significant risk of restless legs syndrome. CONCLUSION: Sleep complaints are highly prevalent in primary care populations. Patients with the highest risk for sleep disturbance are those with pain, mental illness, limited activity, and overall "poor physical and mental health." Because sleep disorders are associated with a significant health impact, positive responses to questions regarding sleep symptoms should prompt further diagnostic inquiry.