Prediction of Obstructive Sleep Apnea Based on Respiratory Sounds Recorded Between Sleep Onset and Sleep Offset

Enhanced machine learning approaches for OSA patient screening: model development and validation study

Age, gender, body mass index (BMI), and mean heart rate during sleep were found to be risk factors for obstructive sleep apnea (OSA), and a variety of methods have been applied to predict the occurrence of OSA. This study aimed to develop and evaluate OSA prediction models using simple and accessible parameters, combined with multiple machine learning algorithms, and integrate them into a cloud-based mobile sleep medicine management platform for clinical use. The study data were obtained from the clinical records of 610 patients who underwent polysomnography (PSG) at the Sleep Medicine Center of the Second Affiliated Hospital of Fujian Medical University between January 2021 and December 2022. The participants were randomly divided into a training-test group (80%) and an independent validation group (20%). The logistic regression, artificial neural network, naïve Bayes, support vector machine, random forest, and decision tree algorithms were used with age, gender, BMI, and mean heart rate during sleep as predictors to build a risk prediction model for moderate-to-severe OSA. To evaluate the performance of the models, we calculated the area under the receiver operating curve (AUROC), accuracy, recall, specificity, precision, and F1-score for the independent validation set. In addition, the calibration curve, decision curve, and clinical impact curve were generated to determine clinical usefulness. Age, gender, BMI, and mean heart rate during sleep were significantly associated with OSA. The artificial neural network model had the best efficacy compared with the other prediction algorithms. The AUROC, accuracy, recall, specificity, precision, F1-score, and Brier score were 80.4% (95% CI 76.7-84.1%), 69.9% (95% CI 69.8-69.9%), 86.5% (95% CI 81.6-91.3%), 61.5% (95% CI 56.6-66.4%), 53.2% (95% CI 47.7-58.7%), 65.9% (95% CI 60.2-71.5%), and 0.165, respectively, for the artificial neural network model. The AUROCs for the LR, NB, SVM, RF, and DT models were 80.2%, 79.7%, 79.2%, 78.4%, and 70.4%, respectively. The six models based on four simple and easily accessible parameters effectively predicted moderate-to-severe OSA in patients with PSG screening, with the artificial neural network model having the best performance. These models can provide a reliable tool for early OSA diagnosis, and their integration into a cloud-based mobile sleep medicine management platform could improve clinical decision making.

In-Home Smartphone-Based Prediction of Obstructive Sleep Apnea in Conjunction With Level 2 Home Polysomnography

Importance

Consumer-level sleep analysis technologies have the potential to revolutionize the screening for obstructive sleep apnea (OSA). However, assessment of OSA prediction models based on in-home recording data is usually performed concurrently with level 1 in-laboratory polysomnography (PSG). Establishing the predictability of OSA using sound data recorded from smartphones based on level 2 PSG at home is important.

Objective

To validate the performance of a prediction model for OSA using breathing sound recorded from smartphones in conjunction with level 2 PSG at home.

Design, Setting, and Participants

This diagnostic study followed a prospective design, involving participants who underwent unattended level 2 home PSG. Breathing sounds were recorded during sleep using 2 smartphones, one with an iOS operating system and the other with an Android operating system, simultaneously with home PSG in participants' own home environment. Participants were 19 years and older, slept alone, and had either been diagnosed with OSA or had no previous diagnosis. The study was performed between February 2022 and February 2023.

Main Outcomes and Measures

Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of the predictive model based on the recorded breathing sounds.

Results

Of the 101 participants included during the study duration, the mean (SD) age was 48.3 (14.9) years, and 51 (50.5%) were female. For the iOS smartphone, the sensitivity values at apnea-hypopnea index (AHI) levels of 5, 15, and 30 per hour were 92.6%, 90.9%, and 93.3%, respectively, with specificities of 84.3%, 94.4%, and 94.4%, respectively. Similarly, for the Android smartphone, the sensitivity values at AHI levels of 5, 15, and 30 per hour were 92.2%, 90.0%, and 92.9%, respectively, with specificities of 84.0%, 94.4%, and 94.3%, respectively. The accuracy for the iOS smartphone was 88.6%, 93.3%, and 94.3%, respectively, and for the Android smartphone was 88.1%, 93.1%, and 94.1% at AHI levels of 5, 15, and 30 per hour, respectively.

Conclusions and Relevance

This diagnostic study demonstrated the feasibility of predicting OSA with a reasonable level of accuracy using breathing sounds obtained by smartphones during sleep at home.

Application of artificial intelligence in the diagnosis of sleep apnea

STUDY OBJECTIVES

Machine learning (ML) models have been employed in the setting of sleep disorders. This review aims to summarize the existing data about the role of ML techniques in the diagnosis, classification, and treatment of sleep-related breathing disorders.

METHODS

A systematic search in Medline, EMBASE, and Cochrane databases through January 2022 was performed.

RESULTS

Our search strategy revealed 132 studies that were included in the systematic review. Existing data show that ML models have been successfully used for diagnostic purposes. Specifically, ML models showed good performance in diagnosing sleep apnea using easily obtained features from the electrocardiogram, pulse oximetry, and sound signals. Similarly, ML showed good performance for the classification of sleep apnea into obstructive and central categories, as well as predicting apnea severity. Existing data show promising results for the ML-based guided treatment of sleep apnea. Specifically, the prediction of outcomes following surgical treatment and optimization of continuous positive airway pressure therapy can be guided by ML models.

CONCLUSIONS

The adoption and implementation of ML in the field of sleep-related breathing disorders is promising. Advancements in wearable sensor technology and ML models can help clinicians predict, diagnose, and classify sleep apnea more accurately and efficiently.

CITATION

Bazoukis G, Bollepalli SC, Chung CT, et al. Application of artificial intelligence in the diagnosis of sleep apnea. J Clin Sleep Med. 2023;19(7):1337-1363.

Multi-feature fusion and improved BO and IGWO metaheuristics based models for automatically diagnosing the sleep disorders from sleep sounds

A night of regular and quality sleep is vital in human life. Sleep quality has a great impact on the daily life of people and those around them. Sounds such as snoring reduce not only the sleep quality of the person but also reduce the sleep quality of the partner. Sleep disorders can be eliminated by examining the sounds that people make at night. It is a very difficult process to follow and treat this process by experts. Therefore, this study, it is aimed to diagnose sleep disorders using computer-aided systems. In the study, the used data set contains seven hundred sound data which has seven different sound class such as cough, farting, laugh, scream, sneeze, sniffle, and snore. In the model proposed in the study, firstly, the feature maps of the sound signals in the data set were extracted. Three different methods were used in the feature extraction process. These methods are MFCC, Mel-spectrogram, and Chroma. The features extracted in these three methods are combined. Thanks to this method, the features of the same sound signal extracted in three different methods are used. This increases the performance of the proposed model. Later, the combined feature maps were analyzed using the proposed New Improved Gray Wolf Optimization (NI-GWO), which is the improved version of the Improved Gray Wolf Optimization (I-GWO) algorithm, and the proposed Improved Bonobo Optimizer (IBO) algorithm, which is the improved version of the Bonobo Optimizer (BO). In this way, it is aimed to run the models faster, reduce the number of features, and obtain the most optimum result. Finally, Support Vector Machine (SVM) and k-nearest neighbors (KNN) supervised shallow machine learning methods were used to calculate the metaheuristic algorithms' fitness values. Different types of metrics such as accuracy, sensitivity, F1 etc., were used for the performance comparison. Using the feature maps optimized by the proposed NI-GWO and IBO algorithms, the highest accuracy value was obtained from the SVM classifier with 99.28% for both metaheuristic algorithms.

Validity of non-contact methods for diagnosis of Obstructive Sleep Apnea: a systematic review and meta-analysis

STUDY OBJECTIVE

Obstructive Sleep Apnea (OSA) is associated with increased perioperative cardiac, respiratory and neurological complications. Pre-operative OSA risk assessment is currently done through screening questionnaires with high sensitivity but poor specificity. The objective of this study was to evaluate the validity and diagnostic accuracy of portable, non-contact devices in the diagnosis of OSA as compared with polysomnography.

DESIGN

This study is a systematic review of English observational cohort studies with meta-analysis and risk of bias assessment.

SETTING

Pre-operative, including in the hospital and clinic setting.

PATIENTS

Adult patients undergoing sleep apnea assessment using polysomnography and an experimental non-contact tool.

INTERVENTIONS

A novel non-contact device, which does not utilize any monitor that makes direct contact with the patient's body, in conjunction with polysomnography.

MEASUREMENTS

Primary outcomes included pooled sensitivity and specificity of the experimental device in the diagnosis of obstructive sleep apnea, in comparison to gold-standard polysomnography.

RESULTS

Twenty-eight of 4929 screened studies were included in the meta-analysis. A total of 2653 patients were included with the majority being patients referred to a sleep clinic (88.8%). Average age was 49.7(SD±6.1) years, female sex (31%), average body mass index of 29.5(SD±3.2) kg/m², average apnea-hypopnea index (AHI) of 24.7(SD±5.6) events/h, and pooled OSA prevalence of 72%. Non-contact technology used was mainly video, sound, or bio-motion analysis. Pooled sensitivity and specificity of non-contact methods in moderate to severe OSA diagnosis (AHI > 15) was 0.871 (95% CI 0.841,0.896, I² 0%) and 0.8 (95% CI 0.719,0.862), respectively (AUC 0.902). Risk of bias assessment showed an overall low risk of bias across all domains except for applicability concerns (none were conducted in the perioperative setting).

CONCLUSION

Available data indicate contactless methods have high pooled sensitivity and specificity for OSA diagnosis with moderate to high level of evidence. Future research is needed to evaluate these tools in the perioperative setting.

A Classifying Model of Obstructive Sleep Apnea Based on Heart Rate Variability in a Large Korean Population

BACKGROUND

The majority of patients with obstructive sleep apnea do not receive timely diagnosis and treatment because of the complexity of a diagnostic test. We aimed to predict obstructive sleep apnea based on heart rate variability, body mass index, and demographic characteristics in a large Korean population.

METHODS

Models of binary classification for predicting obstructive sleep apnea severity were constructed using 14 features including 11 heart rate variability variables, age, sex, and body mass index. Binary classification was conducted separately using apnea-hypopnea index thresholds of 5, 15, and 30. Sixty percent of the participants were randomly allocated to training and validation sets while the other forty percent were designated as the test set. Classifying models were developed and validated with 10-fold cross-validation using logistic regression, random forest, support vector machine, and multilayer perceptron algorithms.

RESULTS

A total of 792 (651 men and 141 women) subjects were included. The mean age, body mass index, and apnea-hypopnea index score were 55.1 years, 25.9 kg/m², and 22.9, respectively. The sensitivity of the best performing algorithm was 73.6%, 70.7%, and 78.4% when the apnea-hypopnea index threshold criterion was 5, 10, and 15, respectively. The prediction performances of the best classifiers at apnea-hypopnea indices of 5, 15, and 30 were as follows: accuracy, 72.2%, 70.0%, and 70.3%; specificity, 64.6%, 69.2%, and 67.9%; area under the receiver operating characteristic curve, 77.2%, 73.5%, and 80.1%, respectively. Overall, the logistic regression model using the apnea-hypopnea index criterion of 30 showed the best classifying performance among all models.

CONCLUSION

Obstructive sleep apnea was fairly predicted by using heart rate variability, body mass index, and demographic characteristics in a large Korean population. Prescreening and continuous treatment monitoring of obstructive sleep apnea may be possible simply by measuring heart rate variability.

Accurate Contactless Sleep Apnea Detection Framework with Signal Processing and Machine Learning Methods

The detection of sleep apnea is critical for assessing sleep quality. It is also a proven biometric in diagnosing cardiovascular and other diseases. Recent studies have shown that radar-based non-contact vital sign monitoring system can effectively detect sleep apnea. However, the detection accuracy in the current study still needs to be improved. In this paper, we propose a sleep apnea detection framework based on FMCW radar. First, the radar system is employed to record the sleep data throughout the night with polysomnography (PSG) comparison. Then, in order to extract more accurate respiratory signal from the raw radar data, the signal processing methods are investigated to solve the observed discontinuity phenomenon. Finally, machine learning methods are adopted. The apneic and not-apneic events are classified accurately by selecting effective features of respiratory signal. As shown in the experimental results, the proposed system could achieve a good classification performance with an accuracy of 95.53%, a sensitivity of 72.60%, a specificity of 97.32%, a Kappa of 0.68, and an F-score of 0.84.

Neurobiology of dream activity and effects of stimulants on dreams

The sleep-wake cycle is the result of the activity of a multiple neurobiological network interaction. Dreaming feature is one interesting sleep phenomena that represents sensorial components, mostly visual perceptions, accompanied with intense emotions. Further complexity has been added to the topic of the neurobiological mechanism of dreams generation by the current data that suggests the influence of drugs on dream generation. Here, we discuss the review on some of the neurobiological mechanism of the regulation of dream activity, with special emphasis on the effects of stimulants on dreaming.

Evaluating Prediction Models of Sleep Apnea From Smartphone-Recorded Sleep Breathing Sounds

Importance

Breathing sounds during sleep are an important characteristic feature of obstructive sleep apnea (OSA) and have been regarded as a potential biomarker. Breathing sounds during sleep can be easily recorded using a microphone, which is found in most smartphone devices. Therefore, it may be easy to implement an evaluation tool for prescreening purposes.

Objective

To evaluate OSA prediction models using smartphone-recorded sounds and identify optimal settings with regard to noise processing and sound feature selection.

Design, Setting, and Participants

A cross-sectional study was performed among patients who visited the sleep center of Seoul National University Bundang Hospital for snoring or sleep apnea from August 2015 to August 2019. Audio recordings during sleep were performed using a smartphone during routine, full-night, in-laboratory polysomnography. Using a random forest algorithm, binary classifications were separately conducted for 3 different threshold criteria according to an apnea hypopnea index (AHI) threshold of 5, 15, or 30 events/h. Four regression models were created according to noise reduction and feature selection from the input sound to predict actual AHI: (1) noise reduction without feature selection, (2) noise reduction with feature selection, (3) neither noise reduction nor feature selection, and (4) feature selection without noise reduction. Clinical and polysomnographic parameters that may have been associated with errors were assessed. Data were analyzed from September 2019 to September 2020.

Main Outcomes and Measures

Accuracy of OSA prediction models.

Results

A total of 423 patients (mean [SD] age, 48.1 [12.8] years; 356 [84.1%] male) were analyzed. Data were split into training (n = 256 [60.5%]) and test data sets (n = 167 [39.5%]). Accuracies were 88.2%, 82.3%, and 81.7%, and the areas under curve were 0.90, 0.89, and 0.90 for an AHI threshold of 5, 15, and 30 events/h, respectively. In the regression analysis, using recorded sounds that had not been denoised and had only selected attributes resulted in the highest correlation coefficient (r = 0.78; 95% CI, 0.69-0.88). The AHI (β = 0.33; 95% CI, 0.24-0.42) and sleep efficiency (β = -0.20; 95% CI, -0.35 to -0.05) were found to be associated with estimation error.

Conclusions and Relevance

In this cross-sectional study, recorded sleep breathing sounds using a smartphone were used to create reasonably accurate OSA prediction models. Future research should focus on real-life recordings using various smartphone devices.

Prediction of Oxygen Desaturation by Using Sound Data From a Noncontact Device: A Proof-of-Concept Study

OBJECTIVES/HYPOTHESIS

Prediction of the apnea-hypopnea index (AHI) from breathing sounds during sleep could be used to prescreen for obstructive sleep apnea (OSA). In addition, the oxygen desaturation index (ODI) is a known risk factor for developing cardiovascular disease in OSA patients. This study focused on estimation of ODI from a noncontact manner from sleep breathing sounds.

STUDY DESIGN

Retrospective study.

METHODS

Patients who visited the sleep center due to snoring or sleep apnea underwent polysomnography in lab overnight. Sound recordings were made during polysomnography using a microphone. After noise reduction, the sound data were segmented into 5 seconds windows and features were extracted. Binary classification and regression analyses were performed to estimate the ODI during sleep (model 1). This was re-tested after inclusion of body mass index (BMI) and age as additional features (model 2: BMI only, model 3: BMI and age).

RESULTS

We included 116 patients. The mean age and AHI of all patients were 50.4 ± 16.7 years and 23.0 ± 24.0 events/hr. In binary classification, for ODI cutoff values of 5, 15, and 30 events/hr, the areas under the curve were 0.88, 0.93, 0.91, respectively, and accuracies were 85.34, 86.21, and 87.07, respectively. In regression analysis, the correlation coefficient and mean absolute error were 0.80 and 9.60 events/hr, respectively. In models 2 and 3, the correlation coefficient and mean absolute error were 0.82, 9.44 events/hr and 0.81, 9.6 events/hr, respectively.

CONCLUSION

Prediction of ODI from sleep sound seems to be feasible. Additional clinical feature such as BMI may increase overall predictability.

LEVEL OF EVIDENCE

IV Laryngoscope, 2021.

Application of substitution box of present cipher for automated detection of snoring sounds

BACKGROUND AND PURPOSE

Snoring is one of the sleep disorders, and snoring sounds have been used to diagnose many sleep-related diseases. However, the snoring sound classification is done manually which is time-consuming and prone to human errors. An automated snoring sound classification model is proposed to overcome these problems.

MATERIAL AND METHOD

This work proposes an automated snoring sound classification method using three new methods. These methods are maximum absolute pooling (MAP), the nonlinear present pattern, and two-layered neighborhood component analysis, and iterative neighborhood component analysis (NCAINCA) selector. Using these methods, a new snoring sound classification (SSC) model is presented. The MAP decomposition model is applied to snoring sounds to extract both low and high-level features. The presented model aims to attain high performance for SSC problem. The developed present pattern (Present-Pat) uses substitution box (SBox) and statistical feature generator. By deploying these feature generators, both textural and statistical features are generated. NCAINCA chooses the most informative/valuable features, and these selected features are fed to k-nearest neighbor (kNN) classifier with leave-one-out cross-validation (LOOCV). The Present-Pat based SSC system is developed using Munich-Passau Snore Sound Corpus (MPSSC) dataset comprising of four categories.

RESULTS

Our model reached an accuracy and unweighted average recall (UAR) of 97.10 % and 97.60 %, respectively, using LOOCV. Moreover, a nocturnal sound dataset is used to show the universal success of the presented model. Our model attained an accuracy of 98.14 % using the used nocturnal sound dataset.

CONCLUSIONS

Our developed classification model is ready to be tested with more data and can be used by sleep specialists to diagnose the sleep disorders based on snoring sounds.

Real-time prediction of upcoming respiratory events via machine learning using snoring sound signal

STUDY OBJECTIVES

The aim of the study was to inspect acoustic properties and sleep characteristics of pre-apneic snoring sound. The feasibility of forecasting upcoming respiratory events by snoring sound was also investigated.

METHODS

Participants with habitual snoring or heavy breathing sound during sleep were recruited consecutively. Polysomnography was conducted and snoring related breathing sound was recorded simultaneously. Acoustic features and sleep features were extracted from 30-second samples and a machine learning algorithm was used to establish two prediction models.

RESULTS

A total of 74 eligible participants were included. Model 1 tested by five-fold cross validation achieved the accuracy of 0.92 and area under the curve of 0.94 for respiratory event prediction. model 2 with acoustic features and sleep information tested by Leave-One-Out cross validation had the accuracy of 0.78 and area under the curve of 0.80. Sleep position was found to be the most important amongst all sleep features contributing to the performance.

CONCLUSIONS

Pre-apneic sound presented unique acoustic characteristics and snoring related breathing sound could be deployed as a real-time apneic event predictor. The model combined with sleep information served as a promising tool for an early warning system to forecast apneic events.

Non-contact diagnosis of obstructive sleep apnea using impulse-radio ultra-wideband radar

While full-night polysomnography is the gold standard for the diagnosis of obstructive sleep apnea, its limitations include a high cost and first-night effects. This study developed an algorithm for the detection of respiratory events based on impulse-radio ultra-wideband radar and verified its feasibility for the diagnosis of obstructive sleep apnea. A total of 94 subjects were enrolled in this study (23 controls and 24, 14, and 33 with mild, moderate, and severe obstructive sleep apnea, respectively). Abnormal breathing detected by impulse-radio ultra-wideband radar was defined as a drop in the peak radar signal by ≥30% from that in the pre-event baseline. We compared the abnormal breathing index obtained from impulse-radio ultra-wideband radar and apnea–hypopnea index (AHI) measured from polysomnography. There was an excellent agreement between the Abnormal Breathing Index and AHI (intraclass correlation coefficient = 0.927). The overall agreements of the impulse-radio ultra-wideband radar were 0.93 for Model 1 (AHI ≥ 5), 0.91 for Model 2 (AHI ≥ 15), and 1 for Model 3 (AHI ≥ 30). Impulse-radio ultra-wideband radar accurately detected respiratory events (apneas and hypopneas) during sleep without subject contact. Therefore, impulse-radio ultra-wideband radar may be used as a screening tool for obstructive sleep apnea.

Prediction of Apnea-Hypopnea Index Using Sound Data Collected by a Noncontact Device

OBJECTIVE

To predict the apnea-hypopnea index (AHI) in patients with obstructive sleep apnea (OSA) using data from breathing sounds recorded using a noncontact device during sleep.

STUDY DESIGN

Prospective cohort study.

SETTING

Tertiary referral hospital.

SUBJECT AND METHODS

Audio recordings during sleep were performed using an air-conduction microphone during polysomnography. Breathing sounds recorded from all sleep stages were analyzed. After noise reduction preprocessing, the audio data were segmented into 5-second windows and sound features were extracted. Estimation of AHI by regression analysis was performed using a Gaussian process, support vector machine, random forest, and simple linear regression, along with 10-fold cross-validation.

RESULTS

In total, 116 patients who underwent attended, in-laboratory, full-night polysomnography were included. Overall, random forest resulted in the highest performance with the highest correlation coefficient (0.83) and least mean absolute error (9.64 events/h) and root mean squared error (13.72 events/h). Other models resulted in somewhat lower but similar performances, with correlation coefficients ranging from 0.74 to 0.79. The estimated AHI tended to be underestimated as the severity of OSA increased. Regarding bias and precision, estimation performances in the severe OSA subgroup were the lowest, regardless of the model used. Among sound features, derivative of the area methods of moments of overall standard deviation demonstrated the highest correlation with AHI.

CONCLUSION

AHI was fairly predictable by using data from breathing sounds generated during sleep. The prediction model may be useful not only for prescreening but also for follow-up after treatment in patients with OSA.

The Endocannabinoid System May Modulate Sleep Disorders in Aging

Aging is an inevitable process that involves changes across life in multiple neurochemical, neuroanatomical, hormonal systems, and many others. In addition, these biological modifications lead to an increase in age-related sickness such as cardiovascular diseases, osteoporosis, neurodegenerative disorders, and sleep disturbances, among others that affect activities of daily life. Demographic projections have demonstrated that aging will increase its worldwide rate in the coming years. The research on chronic diseases of the elderly is important to gain insights into this growing global burden. Novel therapeutic approaches aimed for treatment of age-related pathologies have included the endocannabinoid system as an effective tool since this biological system shows beneficial effects in preclinical models. However, and despite these advances, little has been addressed in the arena of the endocannabinoid system as an option for treating sleep disorders in aging since experimental evidence suggests that some elements of the endocannabinoid system modulate the sleep-wake cycle. This article addresses this less-studied field, focusing on the likely perspective of the implication of the endocannabinoid system in the regulation of sleep problems reported in the aged. We conclude that beneficial effects regarding the putative efficacy of the endocannabinoid system as therapeutic tools in aging is either inconclusive or still missing.