Study on the impact of COVID-19 pandemic on the mental health of Chinese college students: a cross-sectional analysis.

Background: The COVID-19 pandemic has significantly impacted the mental health of college students, prompting the need for universities to implement measures to mitigate these adverse effects. This study aims to assess the mental health status and mitigation measures of college students, identify the primary factors contributing to their mental health challenges, and provide suggestions for educational institutions to reduce negative psychological impacts. Methods: In February 2023, a questionnaire survey was conducted among 1,445 college students. Statistical analysis was performed on the survey results, and multiple regression models were used to identify significant influencing factors and optimize the model. Results: The study revealed correlations between factors affecting mental health during the pandemic, with interactions observed among some factors. Significant differences in mental health status were found among different groups of college students based on their information-sharing habits through apps and engagement in thesis research. Multiple regression analysis indicated that conducting academic research related to COVID-19 significantly increased the psychological stress of college students during the pandemic (p = 0.043). Among all mitigation measures, playing games demonstrated significant effectiveness in model analysis (p = 0.047). The optimization of the model showed that the multiple regression model considering the interaction of factors was more effective. Conclusion: Our research identifies crucial factors influencing the mental health of college students and investigates the mental health status of various student groups. We recommend that educational institutions adopt proactive strategies and a multifaceted approach to support the mental health of college students and address potential issues that may arise.

Efficacy of Oral Appliance for Mild, Moderate, and Severe Obstructive Sleep Apnea: A Meta-analysis.

OBJECTIVE: Oral appliances (OA) are the recommended first-line option for mild-to-moderate obstructive sleep apnea (OSA)-hypopnea. However, there is a lack of evidence to compare the effectiveness of OA in different severities of OSA. The purpose of this study was to investigate the therapeutic effects of preferred OA (tongue retention devices [TRD] and mandibular advancement device [MAD]) in different severities of OSA. DATA SOURCES: PubMed/MEDLINE, The Cochrane Library, and Web of Science. REVIEW METHODS: Concentrating on the efficacy of OA, 2 authors searched 3 databases up to November 10, 2022, independently and systematically, following the requirements and steps of the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. RESULTS: Ultimately, 42 studies with 2265 patients met the criteria for inclusion in OA. Overall, the apnea-hypopnea index improved by 48% (5.6), 67% (14.92), and 62% (32.1) in mild, moderate, and severe OSA, respectively. Subgroup analysis showed a significant difference between MAD and TRD efficacy in mild OSA (58% vs 21%). However, no significant difference was seen between MAD and TRD efficacy in moderate (67% vs 66%) and severe OSA (66% vs 51%). There was no significant difference across groups in the Epworth Sleepiness Scale, oxygen desaturation index (ODI), and lowest oxygen saturation (LSAT). CONCLUSION: Overall, both TRD and MAD are effective treatments for moderate and severe OSA. MAD is efficacious in mild OSA, while TRD requires further validation. Furthermore, mild-moderate and severe OSA received similar improvements in sleepiness, ODI, and LSAT. This study complements the evidence for the efficacy of OA.

Association between overnight repetitive respiratory events and the accumulation of genioglossus fatigue in male patients with severe obstructive sleep apnea.

PURPOSE: To evaluate the correlation between median frequency (MF) as a measure of genioglossus (GG) fatigue and overnight repetitive respiratory events in male patients with severe obstructive sleep apnea (OSA). METHODS: GG electromyography (EMG) data were collected synchronously with polysomnography (PSG). Overnight respiratory events were divided based on whether they occurred during the first or second halves of the total number of overnight respiratory events, and differences in MF in the respiratory phase were compared in the same segments. Events were then sampled in pairs to compare MF. The correlation between MF and the order of respiratory events, as well as interindividual differences, were analyzed. RESULTS: Twenty-two male patients were enrolled in this study and 2210 respiratory events were recorded. Before and during respiratory events, MF decreased significantly in the second half, especially during the inspiratory phase (segments 1-4: P = 0.014, P < 0.001, P < 0.001, P < 0.001, respectively). This trend was observed in non-rapid eye movement sleep and lateral position, but not in rapid eye movement sleep or the supine position, and remained after pairing for duration, stage, and position. MF correlated negatively with the order of respiratory events during the inspiratory phase. The trend of decrease in MF only existed in patients with apnea-hypopnea index > 30 events/h. CONCLUSION: Overnight repetitive respiratory events were associated with increased GG fatigue, influenced by sleep stage and body position in male patients with severe OSA. GG fatigue depends on the order and frequency of respiratory events.

Automatic prediction of obstructive sleep apnea event using deep learning algorithm based on ECG and thoracic movement signals.

BACKGROUND: Obstructive sleep apnea (OSA) is a sleeping disorder that can cause multiple complications. AIMS/OBJECTIVE: Our aim is to build an automatic deep learning model for OSA event detection using combined signals from the electrocardiogram (ECG) and thoracic movement signals. MATERIALS AND METHODS: We retrospectively obtained 420 cases of PSG data and extracted the signals of ECG, as well as the thoracic movement signal. A deep learning algorithm named ResNeSt34 was used to construct the model using ECG with or without thoracic movement signal. The model performance was assessed by parameters such as accuracy, precision, recall, F1-score, receiver operating characteristic (ROC), and area under the ROC curve (AUC). RESULTS: The model using combined signals of ECG and thoracic movement signal performed much better than the model using ECG alone. The former had accuracy, precision, recall, F1-score, and AUC values of 89.0%, 88.8%, 89.0%, 88.2%, and 92.9%, respectively, while the latter had values of 84.1%, 83.1%, 84.1%, 83.3%, and 82.8%, respectively. CONCLUSIONS AND SIGNIFICANCE: The automatic OSA event detection model using combined signals of ECG and thoracic movement signal with the ResNeSt34 algorithm is reliable and can be used for OSA screening.

Associations Between Sleep Spindle Metrics, Age, Education and Executive Function in Young Adult and Middle-Aged Patients with Obstructive Sleep Apnea.

Purpose: This study aimed to investigate the association between sleep spindle metrics and executive function in individuals with obstructive sleep apnea (OSA). Furthermore, we examined the association of age and education on executive function. Patients and Methods: A total of 230 (40.90 ± 8.83 years, F/M = 45/185) participants were enrolled. Overnight electroencephalogram (C3-M2) recording detected sleep spindles by a novel U-Net-type neural network that integrates temporal information with time-frequency images. Sleep spindle metrics, including frequency (Hz), overall density (events/min), fast density (events/min), slow density (events/min), duration (sec) and amplitude (µV), were measured. Executive function was assessed using standardized neuropsychological tests. Associations between sleep spindle metrics, executive function, and demographic factors were analyzed using multivariate linear regression. Results: In fully adjusted linear regression models, higher overall sleep spindle density (TMT-A, B=-1.279, p=0.009; TMT-B, B=-1.813, p=0.008), fast sleep spindle density (TMT-A, B=-1.542, p=0.048; TMT-B, B=-2.187, p=0.036) and slow sleep spindle density (TMT-A, B=-1.731, p=0.037; TMT-B, B=-2.449, p=0.034) were associated with better executive function. And the sleep spindle duration both during N2 sleep time (TMT-A, B=-13.932, p=0.027; TMT-B, B=-19.001, p=0.034) and N3 sleep time (TMT-B, B=-29.916, p=0.009; Stroop-incongruous, B=-21.303, p=0.035) was independently associated with better executive function in this population. Additionally, age and education were found to be highly associated with executive function. Conclusion: Specific sleep spindle metrics, higher overall density, fast density and slow density during N2 sleep time, and longer duration during N2 and N3 sleep time, are independent and sensitive indicators of better executive function in young adult and middle-aged patients with OSA. Further research is needed to explore the underlying mechanisms and clinical implications of these findings.

Sleep disorders causally affect the brain cortical structure: A Mendelian randomization study.

BACKGROUND: s: Previous studies have reported that patients with sleep disorders have altered brain cortical structures. However, the causality has not been determined. We performed a two-sample Mendelian randomization (MR) to reveal the causal effect of sleep disorders on brain cortical structure. METHODS: We included as exposures 11 phenotypes of sleep disorders including subjective and objective sleep duration, insomnia symptom and poor sleep efficiency, daytime sleepiness (narcolepsy)/napping, morning/evening preference, and four sleep breathing related traits from nine European-descent genome-wide association studies (GWASs). Further, outcome variables were provided by ENIGMA Consortium GWAS for full brain and 34 region-specific cortical thickness (TH) and surface area (SA) of grey matter. Inverse-variance weighted (IVW) was used as the primary estimate whereas alternative MR methods were implemented as sensitivity analysis approaches to ensure results robustness. RESULTS: At the global level, both self-reported or accelerometer-measured shorter sleep duration decreases the thickness of full brain both derived from self-reported data (ßIVW = 0.03 mm, standard error (SE) = 0.02, P = 0.038; ßIVW = 0.02 mm, SE = 0.01, P = 0.010). At the functional level, there were 66 associations of suggestive evidence of causality. Notably, one robust evidence after multiple testing correction (1518 tests) suggests the without global weighted SA of superior parietal lobule was influenced significantly by sleep efficiency (ßIVW = -285.28 mm2, SE = 68.59, P = 3.2 × 10-5). CONCLUSIONS: We found significant evidence that shorter sleep duration, as estimated by self-reported interview and accelerometer measurements, was causally associated with atrophy in the entire human brain.

Sleep medicine in China: current clinical practice.

The modern practice of sleep medicine in China was introduced in the 1980s and has undergone significant development over the past few decades. The field has witnessed an increase in sleep laboratories, the publication of guidelines for the diagnosis and treatment of sleep disorders, and the establishment of several sleep medicine associations. Despite these achievements, there is still much to be done in this field. By utilizing original national survey data, this study comprehensively discusses the current practice of sleep medicine in China, including sleep medicine training, diagnostic capacity, multidisciplinary care, clinical competence and standardization, sleep telemedicine, barriers to the practice of sleep medicine, and costs of sleep medicine. It is imperative for the Chinese health care authorities to enhance their attention and investment in the field of sleep medicine. Urgent implementation of standardized training programs and accreditation systems is necessary to promote the diagnosis and treatment of sleep disorders in China. CITATION: Xu S, Li Y, Ye J, Han D. Sleep medicine in China: current clinical practice. J Clin Sleep Med. 2023;19(12):2125-2131.

Facial Contour Features Measured on CT Reflects Upper Airway Morphology in Patients With OSA.

OBJECTIVE: To determine facial contour features, measured on computed tomography (CT), related to upper airway morphology in patients with obstructive sleep apnea (OSA); certain phenotype of facial abnormalities implying restriction of craniofacial skeleton and adipose tissue nimiety has predicted the value of the severity of OSA. MATERIALS AND METHOD: Sixty-four male patients with OSA [apnea-hypopnea index (AHI) ≥10/h] who had upper airway CT were randomly selected to quantitatively measure indicators of facial contour and upper airway structures. Pearson correlation analyses were performed. Partial correlation procedure was used to examine correlations while controlling body mass index (BMI). RESULTS: Upper airway anatomy can nearly all be reflected in the face, except retroglossal airway. Upper face width can be measured to assess the overall skeletal structures of the airway. Lower face width can be used to represent how much adipose tissue deposited. Hard palate, retropalatal, and hypopharyngeal airways have corresponding face indicators respectively. Midface width is a better predictor of AHI severity and minimum blood oxygen even than neck circumference because it contains the most anatomical information about the airway, including RP airway condition, soft palate length, tongue volume, etc. These correlations persisted even after correction for BMI. CONCLUSIONS: All anatomical features of the upper airway except retroglossal airway can be reflected in the face, and midface width is the best predictor of AHI severity and minimum blood oxygen, even better than neck circumference and BMI.

Optimized concurrent hearing and genetic screening in Beijing, China: A cross-sectional study.

Concurrent screening has been proven to provide a comprehensive approach for management of congenital deafness and prevention of ototoxicity. The SLC26A4 gene is associated with late-onset hearing loss and is of great clinical concern. For much earlier detection of newborns with deafness-causing mutations in the SLC26A4 gene, the Beijing Municipal Government launched a chip for optimized genetic screening of 15 variants of 4 genes causing deafness based on a chip to screen for 9 variants of 4 genes, and 6 variants of the SLC26A4 gene have now been added. To ascertain the advantage of a screening chip including 15 variants of 4 genes, the trends in concurrent hearing and genetic screening were analyzed in 2019 and 2020. Subjects were 76,460 newborns who underwent concurrent hearing and genetic screening at 24 maternal and child care centers in Beijing from January 2019 to December 2020. Hearing screening was conducted using transiently evoked otoacoustic emissions (TEOAEs), distortion product otoacoustic emissions (DPOAE), or the automated auditory brainstem response (AABR). Dried blood spots were collected for genetic testing and 15 variants of 4 genes, namely GJB2, SLC26A4, mtDNA 12S rRNA, and GJB3, were screened for using a DNA microarray platform. The initial referral rate for hearing screening decreased from 3.60% (1,502/41,690) in 2019 to 3.23% (1,124/34,770) in 2020, and the total referral rate for hearing screening dropped form 0.57% (236/41,690) in 2019 to 0.54% (187/34,770) in 2020, indicating the reduced false positive rate of newborn hearing screening and policies to prevent hearing loss conducted by the Beijing Municipal Government have had a significant effect. Positivity according to genetic screening was similar in 2019 (4.970%, 2,072/41,690) and 2020 (4.863%,1,691/34,770), and the most frequent mutant alleles were c.235 del C in the GJB2 gene, followed by c.919-2 A > G in the SLC26A4 gene, and c.299 del AT in the GJB2 gene. In this cohort study, 71.43% (5/7) of newborns with 2 variants of the SLC26A4 gene were screened for newly added mutations, and 28.57% (2/7) of newborns with 2 variants of the SLC26A4 gene passed hearing screening, suggesting that a screening chip including 15 variants of 4 genes was superior at early detection of hearing loss, and especially in early identification of newborns with deafness-causing mutations in the SLC26A4 gene. These findings have clinical significance.

Value of Opening the Middle Meatus in Patients With Nasal Airway Obstruction.

OBJECTIVES: While surgeries to correct the anatomical malformations that cause nasal airway obstruction (NAO) are generally successful, the outcomes of such procedures are often unsatisfactory. The aim of the present study was to assess the value of opening the middle meatus in patients with NAO. METHODS: Thirty-four patients with nasal obstruction due to nasal septal deviation were included in this study. After randomization, the middle meatus was either opened or not opened during septoplasty. The patients were evaluated through pre- and postoperative rhinomanometry and acoustic rhinometry. The Visual Analog Scale (VAS) scores of subjective symptoms along with responses to the 20-item Sinonasal Outcome Test (SNOT-20) were obtained before surgery and three months after surgery. RESULTS: The VAS scores and SNOT-20 responses improved significantly in both groups after surgery. The effective treatment rate based on the nasal congestion score (NCS) was 64.7% in the single group (septoplasty alone) and 100% in the combined group (septoplasty in conjunction with opening the middle meatus), and the difference was statistically significant (P = .018). In both groups, surgery significantly improved nasal flow, resistance, minimal cross-sectional area, cross-sectional area 6 cm (CA6) from the anterior nostril and nasal volume. Nasal volume and CA6 after surgery were statistically different between the 2 groups (P = .004 and .019, respectively). CONCLUSIONS: Opening the middle meatus may further improve the subjective perception of patency on the basis of septoplasty.

Clinical Characteristics Combined with Craniofacial Photographic Analysis in Children with Obstructive Sleep Apnea.

Purpose: Distinguishing obstructive sleep apnea (OSA) in a high-risk population remains challenging. This study aimed to investigate clinical features to identify children with OSA combined with craniofacial photographic analysis. Methods: One hundred and forty-five children (30 controls, 62 with primary snoring, and 53 with OSA) were included. Differences in general demographic characteristics and surface facial morphology among the groups were compared. Risk factors and prediction models for determining the presence of OSA (obstructive sleep apnea-hypopnea index>1) were developed using logistic regression analysis. Results: The BMI (z-score), tonsil hypertrophy, and lower face width (adjusted age, gender, and BMI z-score) were showed significantly different in children with OSA compared with primary snoring and controls (adjusted p<0.05). The screening model based on clinical features and photography measurements correctly classified 79.3% of the children with 64.2% sensitivity and 89.1% specificity. The area under the curve of the model was 81.0 (95% CI, 73.5-98.4%). Conclusion: A screening model based on clinical features and photography measurements would be helpful in clinical decision-making for children with highly suspected OSA if polysomnography remains inaccessible in resource-stretched healthcare systems.

Prediction of obstructive sleep apnea using deep learning in 3D craniofacial reconstruction.

Background: Obstructive sleep apnea (OSA) is a common sleep disorder. However, current diagnostic methods are labor-intensive and require professionally trained personnel. We aimed to develop a deep learning model using upper airway computed tomography (CT) to predict OSA and to warn the medical technician if a patient has OSA while the patient is undergoing any head and neck CT scan, even for other diseases. Methods: A total of 219 patients with OSA [apnea-hypopnea index (AHI) ≥10/h] and 81 controls (AHI <10/h) were enrolled. We reconstructed each patient's CT into 3 types (skeletal structures, external skin structures, and airway structures) and captured reconstructed models in 6 directions (front, back, top, bottom, left profile, and right profile). The 6 images from each patient were imported into the ResNet-18 network to extract features and output the probability of OSA using two fusion methods: Add and Concat. Five-fold cross-validation was used to reduce bias. Finally, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were calculated. Results: All 18 views with Add as the feature fusion performed better than did the other reconstruction and fusion methods. This gave the best performance for this prediction method with an AUC of 0.882. Conclusions: We present a model for predicting OSA using upper airway CT and deep learning. The model has satisfactory performance and enables CT to accurately identify patients with moderate to severe OSA.

Analysis of nasal resistance regulation mechanism during postural changes in patients with obstructive sleep apnea by measuring heart rate variability.

STUDY OBJECTIVES: Changes in nasal resistance (NR) during postural changes are influenced by venous filling pressure and autonomic nervous system mediation, and heart rate variability (HRV) can reflect changes in the autonomic nervous system. This study aimed to explore the regulatory mechanisms of NR in patients with obstructive sleep apnea (OSA) during postural changes. METHODS: Healthy controls (apnea-hypopnea index < 5 events/h) and patients with OSA were recruited. NR and electrocardiogram data were collected in sitting, supine, left-lateral, and right-lateral postures. HRV parameters were obtained by analyzing the electrocardiogram data from each posture. Subgroups were divided according to sitting-supine NR changes, and HRV parameters were compared between different postures and groups/subgroups. RESULTS: In total, 34 healthy controls and 39 patients with OSA (mean apnea-hypopnea index 34.34 ± 22.44 events/h) were recruited. During sitting-supine postural changes, the NR increased in the control group but did not change significantly in the OSA group. None of the autonomic nervous system-related HRV parameters changed significantly. After the groups were divided into NR-elevated and NR-unchanged subgroups, sympathetic activity-related HRV parameters were higher in the NR-unchanged subgroup but only statistically significant in the OSA group. When comparing the left and right postures, there was no significant change in NR; however, the OSA group had lower parasympathetic activity-related HRV parameters when in the right posture. CONCLUSIONS: During postural changes from the sitting to supine positions, the total NR increases, and this increment is smaller in patients with OSA. This is likely due to overregulation of sympathetic activity, which may occur in patients with OSA. CITATION: Shi Y, Lou H, Wang H, et al. Analysis of nasal resistance regulation mechanism during postural changes in obstructive sleep apnea patients by measuring heart rate variability. J Clin Sleep Med. 2023;19(4):643-650.

Evaluation of obstructive sleep apnea: an analysis based on aberrant genes.

BACKGROUND: Obstructive sleep apnea (OSA) is an upstream disorder that frequently causes multisystem disorders. Much research has revealed the pathogenesis of OSA, but there is still a lack of research on the complications caused by OSA. METHODS: The mRNA expression and methylation dataset based on peripheral blood mononuclear cells (PBMCs) were downloaded from the Gene Expression Omnibus (GEO) database. All differential expressed genes (DEGs) were ranked using the Robust Rank Aggregation (RRA) algorithm. A weighted gene co-expression network analysis (WGCNA) was constructed. Subsequently, we used immune infiltration, enrichment analysis, and least absolute shrinkage and selection operator (LASSO) regression analysis for apnea and hypopnea index (AHI) and hypertension and excessive daytime sleepiness (EDS) and constructed diagnostic model using random forest algorithm. RESULTS: In the present study, we identified 318 DEGs in PBMCs involved in pathogenesis or continuous positive airway pressure (CPAP) therapy. Pathway enrichment identified DEGs associated with protein regulation and metabolism. Notably, through intra group analysis, we found that the immune disorder was more significant for OSA in males, non-daytime sleepy, or non-hypertensive OSA. The area under the ROC curve of model for EDS prediction is 0.889 and 0.852 for hypertension. Notably, we found that the diagnostic model had a high linear predictive value for AHI. CONCLUSIONS: Our results indicate that PBMCs are a significant component of alterations in OSA and are expected to explain the mechanism of multisystem diseases caused by OSA. The present study provides new insights for symptom evaluation, classification and treatment of OSA from the molecular level.

Causal associations between obstructive sleep apnea and COVID-19: A bidirectional Mendelian randomization study.

BACKGROUNDS: The COVID-19 pandemic has caused significant impact on human health. Whether obstructive sleep apnea (OSA) increases the risk of COVID-19 remains unclear. We sought to clarify this issue using two-sample Mendelian randomization (TSMR) analysis in large cohorts. METHODS: Bidirectional two-sample Mendelian randomization (MR) was used to evaluate the potential causality between OSA and COVID-19 by selecting single-nucleotide polymorphisms (SNPs) as instrumental variables (IVs) from genome-wide association studies (GWAS). The inverse-variance weighted (IVW) method was selected as the main approach for data analysis to estimate the possible causal effects. Alternative methods such as MR-Egger, the MR pleiotropy residual sum and outlier (MR-PRESSO), and leave-one-out analysis methods were implemented as sensitivity analysis approaches to ensure the robustness of the results. RESULTS: All forward MR analyses consistently indicated the absence of a causal relationship between OSA and any COVID-19 phenotype. In the reverse MR analysis, the IVW mode demonstrated that severe respiratory confirmed COVID-19 was correlated with a 4.9% higher risk of OSA (OR, 1.049; 95%CI, 1.018-1.081; P = 0.002), consistent in MR-PRESSO (OR = 1.049, 95%CI 1.018-1.081, P = 0.004), weighted median (OR = 1.048, 95%CI 1.003-1.095, P = 0.035), and MR-Egger (OR = 1.083, 95%CI 1.012-1.190, P = 0.041) methods. CONCLUSIONS: There is no significant evidence supporting a causal association between OSA and any COVID phenotype, while we identified potential evidence for a causal effect of severe COVID-19 on an increased risk of OSA.

Influence of postural changes on nasal resistance in patients with obstructive sleep apnea.

OBJECTIVE: Increased nasal resistance (NR) can augment upper airway collapse in patients with obstructive sleep apnea (OSA). Posture change can lead to altered nasal resistance. Our study aimed to investigate the influence of posture changes on NR in patients with OSA. METHODS: Healthy controls without subjective nasal obstruction (apnea-hypopnea index (AHI) < 5 events/h), patients with OSA and subjective nasal obstruction, and patients with OSA and no subjective nasal obstruction were recruited. NR was measured by active anterior rhinomanometry in sitting, supine, left-lateral, and right-lateral postural positions. Total NR and postural change-related NR increments were calculated and compared among groups. RESULTS: In total, 26 healthy controls and 72 patients with OSA (mean AHI 39.7 ± 24.8 events/h) were recruited. Of patients with OSA, 38/72 (53%) had subjective nasal obstruction. Compared with controls, patients with OSA and no subjective nasal obstruction had lower total NR (inspiration, p = 0.037; expiration, p = 0.020) in the supine postural position. There was no difference in sitting, left-lateral, and right-lateral total NR among groups. Total NR was higher in lateral compared to sitting posture in both patients with OSA and in controls. The NR increment for sitting to supine postural change was significantly lower in patients with OSA (inspiration, p = 0.003; expiration, p = 0.005) compared with controls. The change in NR showed no statistically significant difference among groups in supine-left or supine-right postural change. CONCLUSION: Patients with OSA had lower supine total NR and lower total NR increment in the sitting to supine postural change, which may be related to a different posture-related NR regulatory mechanism. This study provides a new exploratory direction for the compensatory mechanism of the upper airway to collapse during sleep.

Obstructive Sleep Apnea Detection Based on Sleep Sounds via Deep Learning.

Purpose: This study aimed to propose a novel deep-learning method for automatic sleep apneic event detection and thus to estimate the apnea hypopnea index (AHI) and identify obstructive sleep apnea (OSA) in an event-by-event manner solely based on sleep sounds obtained by a noncontact audio recorder. Methods: We conducted a cross-sectional study of participants with habitual snoring or heavy breathing sounds during sleep to train and test a deep convolutional neural network named OSAnet for the detection of OSA based on sleep sounds. Polysomnography (PSG) was conducted, and sleep sounds were recorded simultaneously in a regular room without noise attenuation. The study was conducted in two phases. In phase one, eligible participants were enrolled and randomly allocated into training and validation groups for deep learning algorithm development. In phase two, eligible patients were enrolled in a test group for algorithm assessment. Sensitivity, specificity, accuracy, unweighted Cohen kappa coefficient (κ) and the area under the curve (AUC) were calculated using PSG as the reference standard. Results: A total of 135 participants were randomly divided into a training group (n, 116) and a validation group (n, 19). An independent test group of 59 participants was subsequently enrolled. Our algorithm achieved a precision of 0.81 and sensitivity of 0.78 in the test group for overall sleep event detection. The algorithm exhibited robust diagnostic performance to identify severe cases with a sensitivity of 95.6% and specificity of 91.6%. Conclusion: Our results showed that a deep learning algorithm based on sleep sounds recorded by a noncontact voice recorder served as a feasible tool for apneic event detection and OSA identification. This technique may hold promise for OSA assessment in the community in a relatively comfortable and low-cost manner. Further studies to develop a tool based on a home-based setting are warranted.

Automatic classification of the obstruction site in obstructive sleep apnea based on snoring sounds.

OBJECTIVES: Snoring is a common symptom of obstructive sleep apnea (OSA) which is considered to be potential predictors of the obstruction site. Successful treatment of OSA depend on the determination the types of obstruction site. This study aimed to develop a machine learning-based model to detect obstruction site using snoring sound. METHODS: Patients with OSA underwent drug-induced sleep endoscopy (DISE) and the snoring sounds were recorded simultaneously. We extracted acoustic features based on Mel-frequency cepstral coefficients (MFCC). A k-nearest neighbors (KNN) was used for snore classification. RESULTS: Total 42 patients with OSA were enrolled. The accuracy of model was 85.55 %, F1 score was 85.04. With combined age, gender and Body Mass Index (BMI), the accuracy of model was 87.98 %, and F1 score was 87.96. The model exhibited accuracies of 83 %, 93 % and 92 %; an AUC of 85.88, 89.22 and 88.17 in detecting retropalatal, retrolingual and multilevel obstructions. CONCLUSION: Our results suggest that combing snoring sound with age, gender and BMI, the machine learning based model can help automatically assess obstruction site. The model may have potential utility as a clinical tool to help for clinical decision-making.

Childhood asthma and type 1 diabetes mellitus: A meta-analysis and bidirectional Mendelian randomization study.

BACKGROUND: Worldwide incidence and prevalence of both asthma and type 1 diabetes mellitus (T1DM) in children have been increasing in past decades. Association between the two diseases has been found in some but not in other studies. OBJECTIVE: We conducted a meta-analysis to verify such an association, and bidirectional Mendelian randomization analysis to examine the potential cause-effect relationships. METHODS: Three databases (PubMed, Embase, and Web of Science) were searched from their inception to February 1, 2021. Pooled hazard ratios (HR) or odds ratios (OR), and 95% confidence intervals, were calculated. Associations between single-nucleotide polymorphisms with childhood asthma and T1DM were selected based on genome-wide association studies. The outcome datasets were obtained from FinnGen study. We used the inverse-variance-weighted (IVW), weighted median and MR-Egger methods to estimate causal effects. To assess robustness and horizontal pleiotropy, MR-Egger regression and MR pleiotropy residual sum and outlier test were conducted. RESULTS: In meta-analysis, childhood asthma was associated with an increased risk of T1DM (HR = 1.30, 95% CI 1.05-1.61, P = .014), whereas T1DM was not associated with the risk of asthma (HR = 0.98, 95% CI 0.64-1.51, P = .941; OR = 0.84, 95% CI 0.65-1.08, P = .168). MR analysis indicated increased genetic risk of T1DM in children with asthma (OR = 1.308; 95% CI 1.030-1.661; P = .028). Analysis using the IVW method indicated no association between T1DM and genetic risk of asthma (OR = 1.027, 95%CI 0.970-1.089, P = .358). CONCLUSION: Both meta-analysis and MR study suggested that childhood asthma was a risk factor for T1DM. No epidemiological or genetic evidence was found for an association of T1DM with asthma incidence. Further studies could be carried out to leverage this newfound insight into better clinical and experimental research in asthma and T1DM.

Classification of facial phenotypes in Asian patients with obstructive sleep apnea.

BACKGROUD: The facial phenotypes of Asian obstructive sleep apnea (OSA) patients remain unclear. OBJECTIVES: (1) To describe the facial features of OSA patients. (2) To develop a model based on facial contour indicators to predict OSA. (3) To classify the facial phenotypes of Asian OSA patients. MATERIALS AND METHODS: 110 patients with OSA (apnea-hypopnea index [AHI] ≥ 10/h) and 50 controls (AHI< 10/h) were selected to measure facial contour indicators. Indicators were compared between OSA patients and the control group. We used multivariable linear regression analysis to predict OSA severity and K-means cluster analysis to classify OSA patients into different phenotypes. RESULTS: We built a model to predict OSA which explained 49.1% of its variance and classified OSA patients into four categories. Cluster 1 (Skeletal type) had the narrowest facial width indicators with narrowing of the retroglossal airway. Cluster 2 (Obese type) had the widest face, and narrowest hard palate, retropalatal, and hypopharyngeal airways. Cluster 3 (Nose type) had the narrowest nasal cavity. Cluster 4 (Long type) had the longest airway length. CONCLUSIONS AND SIGNIFICANCE: Patients with OSA were classified into four categories, each of which identified different anatomic risk factors that can be used to select the treatment.