Accurate Deep Learning-Based Sleep Staging in a Clinical Population with Suspected Obstructive Sleep Apnea

Abstract Study Objectives To assess the relationship between obstructive sleep apnea (OSA) severity and sleep fragmentation, accurate differentiation between sleep and wakefulness is needed. Sleep staging is usually performed manually using electroencephalography (EEG). This is time-consuming due to complexity of EEG setup and the amount of work in manual scoring. In this study, we aimed to develop an automated deep learning-based solution to assess OSA-related sleep fragmentation based on photoplethysmography (PPG)-signal. Methods A combination of convolutional and recurrent neural networks was used for PPG-based sleep staging. The models were trained using two large clinical datasets from Israel (n=2149) and Australia (n=877) and tested separately on three-class (wake/NREM/REM), four-class (wake/N1+N2/N3/REM), and five-class (wake/N1/N2/N3/REM) classification. The relationship between OSA severity categories and sleep fragmentation was assessed using survival analysis of mean continuous sleep. Overlapping PPG epochs were applied to artificially obtain denser hypnograms for better identification of fragmented sleep. Results Automatic PPG-based sleep staging showed accuracy of 83.3% on three-class, 74.1% on four-class, and 68.7% on five-class models. The hazard ratios for decreased mean continuous sleep compared to the non-OSA group obtained with Cox proportional hazards models with 5-second epoch-to-epoch intervals were 1.70, 3.30, and 8.11 for mild, moderate, and severe OSA, respectively. With manually scored EEG-based hypnograms, the corresponding hazard ratios were 1.18, 1.78, and 2.90. Conclusions PPG-based automatic sleep staging can be used to differentiate between OSA severity categories based on sleep continuity. The differences between the OSA severity categories become more apparent when a shorter epoch-to-epoch interval is used.

Download Full-text

407 Explanatory analysis of polysomnography for the identification of sleep apnea hypopnea events using deep learning neural network

SLEEP ◽

10.1093/sleep/zsab072.406 ◽

2021 ◽

Vol 44 (Supplement_2) ◽

pp. A161-A162

Author(s):

Soonhyun Yook ◽

Chaitanya Gupte ◽

Zhixian Han ◽

Eun Yeon Joo ◽

Hea Ree Park ◽

...

Keyword(s):

Obstructive Sleep Apnea ◽

Deep Learning ◽

Sleep Apnea ◽

Learning Algorithms ◽

Sleep Apnea Syndrome ◽

Learning Approaches ◽

Explanatory Analysis ◽

Technology Advancement ◽

Obstructive Sleep ◽

Deep Learning Neural Network

Abstract Introduction Using deep learning algorithms, we investigated univariate and multivariate effects of four polysomnography features including heart rate (HR), electrocardiogram (ECG), oxygen saturation (SpO2) and nasal air flow (NAF) on the identification of sleep apnea and hypopnea events. This explanatory analysis that may clarify the sensitivity and specificity of those features to SAs and SHs have not been probed. Methods We studied 804 polysomonography samples from 704 patients with obstructive sleep apnea and 100 controls. The input data were converted into scalograms as 4-channel 2D images to train Xception networks. For training, 77,638 patches were sampled from the original 6-hour sleep data with 30-second time width. A 10% of these patches were segregated as the test-set. With each feature sets, we tested the following classifications: 1) normal vs apnea vs hypopnea; 2) normal vs. apnea+hypopnea; 3) normal vs. apnea; and 4) normal vs. hypopnea. Results SpO2 classified normal vs. apnea most accurately (98%), followed by NAF (85%), ECG (77%), and HR (63%). SpO2 also showed the highest accuracy in classifying normal vs. hypopnea (87%), and normal vs. apnea+hypopnea (96%) and three groups (82%). When the combination of four features were used, the classification accuracies were generally improved compared to use of SpO2 only (normal vs. apnea 99%; vs. hypopnea 89%; vs. apnea+hypopnea: 94%; three groups: 86%). Conclusion Deep learning with SpO2 or NAF feature most accurately classified apneas from normal sleep events, suggesting these features’ characterization of sleep apnea events. Oxygen desaturation, which is a typical pattern of hypopnea, was only the feature showing reliable accuracy in classifying hypopnea vs. normal. Nevertheless, combination of four polysomnography features could improve the identification of sleep apnea and hypopnea. Furthermore, classifying normal vs. apnea+hypopnea was more accurate than separately classifying three groups, suggesting deep learning approaches as the primary screen tool. Since the classification accuracy of using SpO2 was higher than any other features, developing a portable equipment measuring SpO2 and running deep learning algorithms has the potential for inexpensive, accurate diagnostics of obstructive sleep apnea syndrome. Support (if any) This study was supported by USC STEVENS CENTER FOR INNOVATION TECHNOLOGY ADVANCEMENT GRANTS (TAG), BrightFocus Foundation Award (A2019052S).

Download Full-text

Classification and Detection of Breathing Patterns with Wearable Sensors and Deep Learning

Sensors ◽

10.3390/s20226481 ◽

2020 ◽

Vol 20 (22) ◽

pp. 6481

Author(s):

Kristin McClure ◽

Brett Erdreich ◽

Jason H. T. Bates ◽

Ryan S. McGinnis ◽

Axel Masquelin ◽

...

Keyword(s):

Obstructive Sleep Apnea ◽

Deep Learning ◽

Sleep Apnea ◽

Wearable Sensors ◽

Rapid Assessment ◽

Central Sleep Apnea ◽

Breathing Patterns ◽

Movement Data ◽

Normal Breathing ◽

Obstructive Sleep

Rapid assessment of breathing patterns is important for several emergency medical situations. In this research, we developed a non-invasive breathing analysis system that automatically detects different types of breathing patterns of clinical significance. Accelerometer and gyroscopic data were collected from light-weight wireless sensors placed on the chest and abdomen of 100 normal volunteers who simulated various breathing events (central sleep apnea, coughing, obstructive sleep apnea, sighing, and yawning). We then constructed synthetic datasets by injecting annotated examples of the various patterns into segments of normal breathing. A one-dimensional convolutional neural network was implemented to detect the location of each event in each synthetic dataset and to classify it as belonging to one of the above event types. We achieved a mean F1 score of 92% for normal breathing, 87% for central sleep apnea, 72% for coughing, 51% for obstructive sleep apnea, 57% for sighing, and 63% for yawning. These results demonstrate that using deep learning to analyze chest and abdomen movement data from wearable sensors provides an unobtrusive means of monitoring the breathing pattern. This could have application in a number of critical medical situations such as detecting apneas during sleep at home and monitoring breathing events in mechanically ventilated patients in the intensive care unit.

Download Full-text

Deep learning applied to polysomnography to predict blood pressure in obstructive sleep apnea and obesity hypoventilation: a proof-of-concept study

Journal of Clinical Sleep Medicine ◽

10.5664/jcsm.8608 ◽

2020 ◽

Vol 16 (10) ◽

pp. 1797-1803 ◽

Cited By ~ 1

Author(s):

Bharati Prasad ◽

Chirag Agarwal ◽

Elan Schonfeld ◽

Dan Schonfeld ◽

Babak Mokhlesi

Keyword(s):

Blood Pressure ◽

Obstructive Sleep Apnea ◽

Deep Learning ◽

Sleep Apnea ◽

Proof Of Concept ◽

Concept Study ◽

Obstructive Sleep ◽

Obesity Hypoventilation

Download Full-text

Characteristics of Patients with Obstructive Sleep Apnea at High Risk for Cardiovascular Disease

Medicina ◽

10.3390/medicina57111265 ◽

2021 ◽

Vol 57 (11) ◽

pp. 1265

Author(s):

Izolde Bouloukaki ◽

Michail Fanaridis ◽

Georgios Stathakis ◽

Christina Ermidou ◽

Eleftherios Kallergis ◽

...

Keyword(s):

Cardiovascular Disease ◽

Obstructive Sleep Apnea ◽

Depressive Symptoms ◽

Sleep Apnea ◽

Odds Ratio ◽

Clinical Population ◽

Clinical Patient ◽

Athens Insomnia Scale ◽

Obstructive Sleep ◽

And Gender

Background and Objectives: To evaluate the influence of obstructive sleep apnea (OSA)-related symptoms on prevalent cardiovascular disease (CVD) in a large clinical population of patients. Materials and Methods: A total of 2127 patients (mean age 55 years, 24% women) underwent diagnostic polysomnography and were evaluated using the Epworth sleepiness scale (ESS), the Athens Insomnia Scale (AIS), and the Beck Depression Inventory (BDI). We investigated the predictive value of OSA-associated symptoms for prevalent cardiovascular disease, after adjustment for relevant confounding factors including age, obesity, and co-morbidities. Results: Patients with OSA and CVD were older and had a higher Body Mass Index (BMI); the percentage of obese patients was also higher (83% vs. 70%, p < 0001). They also had greater neck, waist, and hip circumferences and a higher waist-to-hip ratio. Excessive daytime sleepiness (ESS ≥ 10) [odds ratio (95% CI) 1.112 (0.708–1.748), p = 0.64], insomnia symptoms (AIS ≥ 6) [odds ratio (95% CI) 0.748 (0.473–1.184), p = 0.21], frequent awakenings [odds ratio (95% CI) 1.599 (1.019–2.508), p = 0.06], and nocturia [odds ratio (95% CI) 1.359 (0.919–2.009), p = 0.124] were not associated with CVD after adjustment for the previous confounders. On the other hand, depressive symptoms (BDI ≥ 10) independently predicted prevalent CVD [odds ratio (95% CI) 1.476 (1.154–1.887), p = 0.002]. Further analysis in subgroups stratified by age, BMI, and gender demonstrated that depressive symptoms predicted prevalent CVD but only in the subgroup of younger (age group < 60 years), obese (BMI group ≥ 30), and male (OR = 1.959, 95% CI = 1.209–3.175, p = 0.006) OSA patients. Conclusions: OSA patients with CVD were more likely to complain of less typical OSA symptoms and depressive symptoms compared to patients without CVD in this large clinical patient cohort, supportingthecomplexity and heterogeneityof OSA.

Download Full-text

Differentiating Sleepy and non-sleepy obstructive sleep apnea patients using nocturnal pulse oximetry and deep learning

Sleep Medicine ◽

10.1016/j.sleep.2019.11.508 ◽

2019 ◽

Vol 64 ◽

pp. S184

Author(s):

S. Kainulainen ◽

J. Töyräs ◽

A. Oksenberg ◽

H. Korkalainen ◽

I. Afara ◽

...

Keyword(s):

Obstructive Sleep Apnea ◽

Deep Learning ◽

Sleep Apnea ◽

Pulse Oximetry ◽

Obstructive Sleep

Download Full-text

0616 Validation of a Sleep Headband for Detecting Obstructive Sleep Apnea

SLEEP ◽

10.1093/sleep/zsaa056.613 ◽

2020 ◽

Vol 43 (Supplement_1) ◽

pp. A236-A236

Author(s):

A Guillot ◽

T Moutakanni ◽

M Harris ◽

P J Arnal ◽

V Thorey

Keyword(s):

Obstructive Sleep Apnea ◽

Deep Learning ◽

Sleep Apnea ◽

Cross Validation ◽

Apnea Hypopnea Index ◽

Learning Approach ◽

Specificity And Sensitivity ◽

Obstructive Sleep ◽

Fold Cross Validation ◽

At Home

Abstract Introduction Polysomnography (PSG) is the gold-standard to diagnose obstructive sleep apnea (OSA). OSA severity diagnosis is defined by the apnea-hypopnea index (AHI) defined as the number of apnea and hypopnea events measured per hour of sleep. The Dreem2 headband (DH) is a self-administered, easy to use device that measure EEG, breathing frequency, heart rate and sound at-home. In our study, we assessed the performance of the DH to automatically detects OSA compared to 3 sleep’s experts scoring on PSG. Methods 41 subjects (8 females, 42.6 ± 13.7 y.o.) having a suspicion of OSA performed a night at-home wearing both a PSG and the DH. Each PSG record was scored for apnea and hypopnea events by 3 independent trained sleep experts following AASM guidelines. The deep learning approach DOSED, was trained on the DH signals using the manual apnea scoring. 10-fold cross-validation was used to provide predictions for each of the 41 subjects with the DH. Results We observed an average AHI expert’s scoring of 13.6 ± 10.1 CI[10.5, 16.5] compared to 12.9 ± 10.3 CI[9.6, 15.8] for the DH. Both, the correlation between the 3 scorers (r= 0.88, p < 0.001) and the DH and the scorers (r=0.79, p< 0.001) were significant. The specificity and sensitivity to detect mild OSA (AHI ≤ 5) was 84.4 % and 96.4 % for the DH and 86.5 % and 86.0% for the scorers. Conclusion The results show that the DH using deep learning can detect OSA with an accuracy similar to the sleep experts. The use of DH paves the way for longitudinal monitoring of patients with a suspicion of OSA and its accessibility could lead to better screening of the general population. Support This Study has been supported by Dreem sas.

Download Full-text