Machine Learning–based Sleep Staging in Patients with Sleep Apnea Using a Single Mandibular Movement Signal

Le-Dong, Nhât-Nam; Martinot, Jean-Benoît; Coumans, Nathalie; Cuthbert, Valérie; Tamisier, Renaud; Bailly, Sébastien; Pépin, Jean-Louis

doi:10.1164/rccm.202103-0680le

Cited by 26 publications

(14 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our results show that, for patients at OSA risk and without an OSA diagnosis, the Berlin score is a simple questionnaire with a high sensitivity but a lack of specificity, and it can be easily used by dieticians at the beginning of a weight-loss program. Moreover, a new home sleep test has also been developed, and it can be repeated over the course of weight loss; this tool should be disseminated in physicians’ offices and obesity clinics in the future for the better management of OSA [ 34 , 35 , 36 ].…”

Section: Discussionmentioning

confidence: 99%

Impact of a Weight-Loss Rehabilitation Program on Sleep Apnea Risk and Subjective Sleepiness in Patients with Overweight/Obesity: The DietSleep Study

Bailly

Fabre

Cals-Maurette

et al. 2022

JCM

Self Cite

View full text Add to dashboard Cite

Obstructive sleep apnea (OSA) is one of the most frequent chronic diseases, and comorbid obesity occurs in more than 60% of cases. Variations in body weight influence both OSA severity and OSA-related symptoms. We prospectively assessed the impact of a weight-loss program using the Berlin score to reflect OSA risk, and we also used the Epworth Sleepiness Scale (ESS) to assess daytime sleepiness. DietSleep was a prospective multicentric cohort study investigating OSA risk and daytime sleepiness before and after weight-loss intervention. One hundred and twenty-seven patients were included (initial OSA risk 36%), most of whom were women (85.8%) with a median body mass index (BMI) of 29.7 kg/m2, and the interquartile range was (27.6; 34). The diet-based weight-loss program induced a median decrease in BMI of 3.7 kg/m2 (−5; −2.9) (body weight~12.1% (−16.0; −8.8)) over a period of 171 days (114; 269). Changes in anthropometric values were similar regarding OSA risk after adjusting for initial values. Berlin scores significantly improved from 3 (1; 5) to 1 (0; 2), p < 0.01; the proportion of patients with a Berlin score ≥2 decreased from 36% to 7% after the intervention. The proportion of patients with ESS ≥11 decreased from 13% to 2%. These results confirm that a weight-loss program produces clinically relevant weight loss and a significant improvement in both OSA and subjective daytime sleepiness.

show abstract

Section: Discussionmentioning

confidence: 99%

Impact of a Weight-Loss Rehabilitation Program on Sleep Apnea Risk and Subjective Sleepiness in Patients with Overweight/Obesity: The DietSleep Study

Bailly

Fabre

Cals-Maurette

et al. 2022

JCM

Self Cite

View full text Add to dashboard Cite

show abstract

“…To identify wake, the algorithm tested whether MJM signals were fast, irregular and non-predictable [ 31 ]. For the identification of arousal movements, the algorithm detected brisk MJM of large amplitude indicating the abrupt closure of the mouth characteristic of arousals [ 31 , 32 ].…”

Section: Methodsmentioning

confidence: 99%

Respiratory effort during sleep and prevalent hypertension in obstructive sleep apnoea

Martinot

Le-Dong²,

Malhotra

et al. 2022

Eur Respir J

Self Cite

View full text Add to dashboard Cite

Mechanisms underlying blood pressure changes in obstructive sleep apnoea (OSA) are incompletely understood. Increased respiratory effort (RE) is one of the main features of OSA and is associated with sympathetic overactivity, leading to increased vascular wall stiffness and remodelling. This study investigated associations between a new measure of RE (percentage of sleep time spent with increased RE based on measurement of mandibular jaw movements [MJM]; REMOV, %TST) and prevalent hypertension in adults referred for evaluation of suspected OSA. A machine learning model was built to predict hypertension from clinical data, conventional polysomnography (PSG) indices, and MJM-derived parameters (including REMOV, %TST). The model was evaluated in a training subset and a test subset. The analysis included 1127 patients, 901 (80%) in the training subset and 226 (20%) in the test subset. The prevalence of hypertension was 31% and 30%, respectively, in the training and test subsets. A risk stratification model based on eighteen input features including REMOV had good accuracy for predicting prevalent hypertension (sensitivity 0.75, specificity 0.83). Using the Shapley additive explanation (SHAP) method, REMOV was the best predictor of hypertension after clinical risk factors (age, sex, body mass index, neck circumference) and time with oxygen saturation <90%, ahead òf standard PSG metrics (including the apnoea-hypopnoea index and oxygen desaturation index). The proportion of sleep time spent with increased RE automatically derived from MJM was identified as a potential new reliable metric to predict prevalent hypertension in patients with OSA.

show abstract

“…Martinot et al (2019) and Pépin et al (2020) have shown that mandibular movements measured using midsagittal mounted magnetic sensors on the chin and the forehead, successfully differentiated obstructive and central events (Martinot et al, 2019) and when the signals were combined with machine learning could successfully discriminate controls from apnoea patients (RDI ≥ 5) with an area under the receiver operating characteristic curve (AUC-ROC) of 0.95 (Pépin et al, 2020). More recent work by the same group (Le-Dong et al, 2021) demonstrated that machine learning also enabled accurate sleep staging to be performed with an AUC for wake of 0.98, N1/N2 sleep of 0.86, N3 sleep of 0.97, and REM sleep of 0.96. Additionally, mandibular jaw movements (MJM) can be used as a surrogate measure for nocturnal respiratory effort (RE), since the slight protrusions of the mandible during sleep are reflective of respiratory drive (Martinot et al, 2022).…”

Section: Biological Signalsmentioning

confidence: 99%

Innovations in mandibular advancement splint therapy for obstructive sleep apnoea

Mohammadieh

Tong²,

Chazal³

et al. 2023

Front. Sleep

View full text Add to dashboard Cite

Mandibular advancement splint (MAS) therapy emerged as an effective therapy for obstructive sleep apnoea (OSA) in the mid 1990s, and is now the leading treatment alternative for OSA. Since its inception, the field has seen a suite of revisions and advances in relation to design and customisation, fabrication, titration methods, response prediction models and the integration of data collection technology. This paper reviews these current and emerging innovations in MAS therapy and their impact upon sleep apnoea management.

show abstract

Machine Learning–based Sleep Staging in Patients with Sleep Apnea Using a Single Mandibular Movement Signal

Cited by 26 publications

References 8 publications

Impact of a Weight-Loss Rehabilitation Program on Sleep Apnea Risk and Subjective Sleepiness in Patients with Overweight/Obesity: The DietSleep Study

Impact of a Weight-Loss Rehabilitation Program on Sleep Apnea Risk and Subjective Sleepiness in Patients with Overweight/Obesity: The DietSleep Study

Respiratory effort during sleep and prevalent hypertension in obstructive sleep apnoea

Innovations in mandibular advancement splint therapy for obstructive sleep apnoea

Contact Info

Product

Resources

About