Search

Introduction A polysomnography or home sleep apnea study provides multiple pieces of information to diagnose obstructive sleep apnea (OSA), but the tests are costly with limited access. The study aims to use an automated AHI model with only PPG signals and can be applied to a wearable device. Methods We have included patients with different OSA severity to build an algorithm detecting ODI based on the scoring criteria with varying sizes of windows ranging from 10 to 60 seconds. For patients without ODI events, the automated CPC for detecting low-frequency oscillation is included to support the automated AHI model. Results The automated ODI and the combination of automated CPC are highly correlated with the AHI. When a CPC is detected without the ODI, the low-frequency coupling can assist in detecting AHI. The accuracy of the automated AHI is 86% compared to the actual AHI, with the sensitivity, specificity and precision at 92%, 73% and 89%, respectively. Conclusion The automated AHI algorithm with PPG signals as input can have a high sensitivity and accuracy in screening patients with OSA (AHI≥5), which can considerably be implied in a PPG wearable device. Support (if any)

Introduction Stressful conditions can impact our judgement and increase the risk of adverse events such as workplace accidents, falls, and automobile collisions, among others. Sleep quality has also been shown to influence stress responses and heart rate variability (HRV) has proved to be a useful indicator of stress in a variety of scenarios. Therefore, wearable devices such as heart rate monitors may be able to reduce adverse events by measuring stress responses. Among subjects reporting different sleep quality, we measured heart rate variability (HRV) before and after different stressors to understand whether these stressful events could be detected by a wearable device. Methods Twenty-four subjects (male = 10; female = 14) with no known health conditions participated in this study. Subjects were divided into good sleep quality (N = 12) and bad sleep quality (N = 12) groups based on their Pittsburgh Sleep Quality Index (PSQI) score. Each subject engaged in a physical stress and mental stress on separate days and HRV was assessed before and after each stress intervention. T-tests were used to assess the change in HRV from pre- to post stress condition for each group. Results In the good sleep quality group we found several statistically significant differences across several HRV frequency-domains before and after physical stress (Low frequency: p = 0.0292; High frequency: p = 0.0287; Low frequency/ High frequency ratio: p = 0.0245) and mental stress (Low frequency: p = 0.0394; High frequency: p = 0.0387; Low frequency/ High frequency ratio: p = 0.0373). Conclusion Our findings indicate that sleep quality may influence the HRV response to different stressors. This suggests that HRV measured via wearable device may be used to provide warning under stressful conditions that could help prevent accidents and other adverse events. Support (if any)

Study Objectives Polysomnography is the gold standard for diagnosis of obstructive sleep apnea (OSA) but it is costly and access is often limited. The aim of this study is to develop a clinically useful support vector machine (SVM)-based prediction model to identify patients with high probability of OSA for nonsleep specialist physician in clinical practice. Methods The SVM model was developed using the features routinely collected at the clinical evaluation from 6,875 Chinese patients referred to sleep clinics for suspected OSA. Three apnea-hypopnea index (AHI) cutoffs, ≥5/h, ≥15/h, and ≥30/h were used to define the severity of OSA. The continuous and categorized features were selected separately and were further selected through stepwise forward feature selection. The modeling was achieved through fivefold cross-validation. The model discriminative ability was evaluated for the whole data set and four subgroups categorized with gender and age ( Results Two features were selected to predict AHI cutoff ≥5/h with six features selected for ≥15/h, and six features selected for ≥30/h, respectively, to reach Area under the Receiver Operating Characteristic (AUROC) 0.82, 0.80, and 0.78, respectively. The sensitivity was 74.14%, 75.18%, and 70.26%, while the specificity was 74.71%, 68.73%, and 70.30%, respectively. Compared to logistic regression, Berlin questionnaire, NoSAS Score, and Supersparse Linear Integer Model (SLIM) scoring system, the SVM model performs better with a more balanced sensitivity and specificity. The discriminative ability was best for male Conclusion Our model provides a simple and accurate modality for early identification of patients with OSA and may potentially help prioritize them for sleep study.

Introduction The study aims to validate the automatic sleep staging system (ASSS) with photoplethysmography (PPG) and accelerometers embedded in smart watches in community-based population Methods 75 healthy subjects were randomly recruited form 304 staffs in an industrial firm who volunteered for this study. A four-stage classifier was designed based on Linear Discriminant Analysis using PPG and accelerometers. To better validate the system performance, the leave-one-out approach was applied in this study. The performance of ASSS was assessed with the epoch-by-epoch and whole-night agreement for sleep staging against manual scoring of overnight polysomnography. Results The mean agreement of four stages across all subjects was 61.1% (95% CI, 58.9-63.2) with kappa 0.55 (0.52-0.58). The mean agreement for wake, light sleep (LS), deep sleep (DS), and REM was 53.4%, 84.1%, 40.3%, 75.6%, respectively. The whole-night agreement was good-excellent (Intra-class correlation coefficient, 0.74 to 0.84) for total sleep time, sleep efficiency, wake after sleep onset, and duration of wake and REM. The agreement was fair for sleep onset and LS duration, but poor for DS duration. Conclusion Our result showed that PPG and accelerometers based smart watches have proper validity for automatic sleep staging in the community-based population. Support (if any) “Center for electronics technology integration (NTU-107L900502, 108L900502, 109-2314-B-002-252)” from the Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan; MediaTek Inc (201802034 RIPD).

Introduction Obstructive sleep apnea (OSA) is a condition characterized by repeated episodes of partial or complete obstruction of the respiratory passages during the sleep. Traditional polysomnography (PSG) for OSA estimation is bulky and time-consuming for daily use. Therefore, this study aims to develop a novel photoplethysmography (PPG) and accelerometer based smart watch for OSA detection, in which a high-performance and low-complexity automated OSA detection was embedded for long-term in-home measurement. Methods The developed watch measured PPG signals from wrist radial artery and body motion from accelerometer as well. 121 patients (92 males, 29 females) were recruited from the normal community and Center of Sleep, National Taiwan University Hospital, Taiwan in this study. All OSA scoring were analyzed by three registered PSG technologists. The AHI of the cohort was 10.1±18.3 (0 to 82.7). An automated OSA detection algorithm was designed based on machine-learning (ML) technique, in which was iteratively updated according to each 30-second epoch of the collected data. Subsequently, obstructive and hypopnea events were detected according to the OSA detection algorithm. Results To better valid the effectiveness, this study focused on the estimation performance of the subjects with AHI>15. Based on hold-out validation, the average sensitivity and precision in the AHI>15 cohort were 77.2% and 58.6%, respectively, with a Cohen’s kappa of 0.46. The interclass correlation between the watch and technologists was 0.81 (95%CI: 0.61-0.91). The result showed that the proposed automated OSA detection could achieve consistent result with technologist during standard sleep testing. Conclusion This study developed a wrist-based watch based on ML technique to assess OSA severity. We compared the performance with clinical technologists for the OSA detection. Further, the sensitivity and precision were generally acceptable while subjects were with AHI>15. The proposed wrist-based watch could provide reliable performance for OSA estimation, and may be of a light for future in-home sleep studies. Support This study was supported by Mediatek Inc.

Introduction Excessive daytime sleepiness (EDS) is a common symptom that patients with obstructive sleep apnea (OSA) seek medical attention for. Prevalence ranged from 20% to 60%. Previous studies reported factors associated with EDS included age, body mass index (BMI), depression, and OSA severity. In most studies, the sample size was small, participants having specific co-morbidities, and the definitions of EDS was heterogeneous. Moreover, the association between anxiety, depression, habitual sleep pattern, and EDS has not been widely studied. Therefore, the present study aims to investigate the prevalence of EDS and associated factors, especially anxiety, depression, and habitual sleep pattern, in a large-scale clinical sample. Methods Data was prospectively collected from 8,081 adult patients who underwent initial overnight polysomnography (PSG) for the first time were referred from 2009 to 2016. Patients with total recording time less than 240 minutes and missing data were excluded. Data collected include demographics, anthropometrics, co-morbidities, and self-reported habitual sleep patterns. Subjective sleepiness was assessed by the Epworth Sleepiness Scale (ESS) with EDS defined as ESS ≥10. Anxiety and depression were assessed by the Hospital Anxiety and Depression Scale (HADS). The stepwise forward Logistic regression was used to identified predictors for EDS. Results In 5,780 (82.6%) patients with OSA (apnea-hypopnea index, AHI≥5/h), mean age was 63.9±0.2 yr, BMI was 27.7±0.1, and ESS was 10.4±0.1 and 80.5% were male. Prevalence of EDS in all OSA patients was 55.1% where the patients with severe OSA had higher prevalence (59.8%) than that in mild (49.5%) and moderate OSA (51.8%). Anxiety (OR: 2.036, 95% CI:1.153-1.502), depression (OR: 1.159, 95% CI:1.01-1.33), and short sleep ( Conclusion Anxiety, depression, short sleep are predictive and OSA severity are predictive of EDS while long sleep was associated with lower risk. Support National Taiwan University (NTU-EPR-104R8951-1; 105R8951-1; 106R880301), Center of electronics technology integration (NTU-107L900502, 108L900502) by the Ministry of Education in Taiwan and MediaTek Inc (201802034RIND).