Automated Detection of Sleep Apnea-Hypopnea Events Based on 60 GHz Frequency-Modulated Continuous-Wave Radar Using Convolutional Recurrent Neural Networks: A Preliminary Report of a Prospective Cohort Study

A novel deep learning model for obstructive sleep apnea diagnosis: hybrid CNN-Transformer approach for radar-based detection of apnea-hypopnea events

STUDY OBJECTIVES

The demand for cost-effective and accessible alternatives to polysomnography (PSG), the conventional diagnostic method for obstructive sleep apnea (OSA), has surged. In this study, we have developed and validated a deep learning model for detecting apnea-hypopnea events using radar data.

METHODS

We conducted a single-center prospective cohort study, dividing participants with suspected sleep-disordered breathing into development and temporally independent test sets. Utilizing a hybrid CNN-Transformer architecture, we performed fivefold cross-validation on the development set to develop and subsequently validate the model. Evaluation metrics included sensitivity for event detection, mean absolute error (MAE), intraclass correlation coefficient (ICC), and Pearson correlation coefficient (r) for apnea-hypopnea index (AHI) estimation. Linearly weighted kappa statistics (κ) assessed OSA severity.

RESULTS

The development set comprised 54 participants (July 2021-May 2022), while the test set included 35 participants (June 2022-June 2023). In the test set, our model achieved an event detection sensitivity of 67.2% (95% CI = 65.8% to 68.5%) and demonstrated a MAE of 7.54 (95% CI = 5.36 to 9.72), indicating good agreement (ICC = 0.889 [95% CI = 0.792 to 0.942]) and a strong correlation (r = 0.892 [95% CI = 0.795 to 0.945]) with the ground truth for AHI estimation. Furthermore, OSA severity estimation showed substantial agreement (κ = 0.780 [95% CI = 0.658 to 0.903]).

CONCLUSIONS

Our study highlights radar sensors and advanced AI models' potential to improve OSA diagnosis, paving the path for future radar-based diagnostic models in sleep medicine research.

Combining a wireless radar sleep monitoring device with deep machine learning techniques to assess obstructive sleep apnea severity

STUDY OBJECTIVES

The gold standard for diagnosing obstructive sleep apnea (OSA) is polysomnography (PSG). However, PSG is a time-consuming method with clinical limitations. This study aimed to create a wireless radar framework to screen the likelihood of 2 levels of OSA severity (ie, moderate-to-severe and severe OSA) in accordance with clinical practice standards.

METHODS

We conducted a prospective, simultaneous study using a wireless radar system and PSG in a Northern Taiwan sleep center, involving 196 patients. The wireless radar sleep monitor, incorporating hybrid models such as deep neural decision trees, estimated the respiratory disturbance index relative to the total sleep time established by PSG (RDIPSG_TST), by analyzing continuous-wave signals indicative of breathing patterns. Analyses were performed to examine the correlation and agreement between the RDIPSG_TST and apnea-hypopnea index, results obtained through PSG. Cut-off thresholds for RDIPSG_TST were determined using Youden's index, and multiclass classification was performed, after which the results were compared.

RESULTS

A strong correlation (ρ = 0.91) and agreement (average difference of 0.59 events/h) between apnea-hypopnea index and RDIPSG_TST were identified. In terms of the agreement between the 2 devices, the average difference between PSG-based apnea-hypopnea index and radar-based RDIPSG_TST was 0.59 events/h, and 187 out of 196 cases (95.41%) fell within the 95% confidence interval of differences. A moderate-to-severe OSA model achieved an accuracy of 90.3% (cut-off threshold for RDIPSG_TST: 19.2 events/h). A severe OSA model achieved an accuracy of 92.4% (cut-off threshold for RDIPSG_TST: 28.86 events/h). The mean accuracy of multiclass classification performance using these cut-off thresholds was 83.7%.

CONCLUSIONS

The wireless-radar-based sleep monitoring device, with cut-off thresholds, can provide rapid OSA screening with acceptable accuracy and also alleviate the burden on PSG capacity. However, to independently apply this framework, the function of determining the radar-based total sleep time requires further optimizations and verification in future work.

CITATION

Lin S-Y, Tsai C-Y, Majumdar A, et al. Combining a wireless radar sleep monitoring device with deep machine learning techniques to assess obstructive sleep apnea severity. J Clin Sleep Med. 2024;20(8):1267-1277.

Developing a deep learning model for sleep stage prediction in obstructive sleep apnea cohort using 60 GHz frequency-modulated continuous-wave radar

Given the significant impact of sleep on overall health, radar technology offers a promising, non-invasive, and cost-effective avenue for the early detection of sleep disorders, even prior to relying on polysomnography (PSG)-based classification. In this study, we employed an attention-based bidirectional long short-term memory (Attention Bi-LSTM) model to accurately predict sleep stages using 60 GHz frequency-modulated continuous-wave (FMCW) radar. Our dataset comprised 78 participants from an ongoing obstructive sleep apnea (OSA) cohort, recruited between July 2021 and November 2022, who underwent overnight polysomnography alongside radar sensor monitoring. The dataset encompasses comprehensive polysomnography recordings, spanning both sleep and wakefulness states. The predictions achieved a Cohen's kappa coefficient of 0.746 and an overall accuracy of 85.2% in classifying wakefulness, rapid-eye-movement (REM) sleep, and non-REM (NREM) sleep (N1 + N2 + N3). The results demonstrated that the models incorporating both Radar 1 and Radar 2 data consistently outperformed those using only Radar 1 data, indicating the potential benefits of utilising multiple radars for sleep stage classification. Although the performance of the models tended to decline with increasing OSA severity, the addition of Radar 2 data notably improved the classification accuracy. These findings demonstrate the potential of radar technology as a valuable screening tool for sleep stage classification.

Comparative Effectiveness of Supine Avoidance versus Continuous Positive Airway Pressure for Treating Supine-isolated Sleep Apnea: A Clinical Trial

Rationale: About 20-35% of patients with obstructive sleep apnea (OSA) have supine-isolated OSA, for which supine sleep avoidance could be an effective therapy. However, traditional supine discomfort-based methods show poor tolerance and compliance to treatment and so cannot be recommended. Supine alarm devices show promise, but evidence to support favorable adherence to treatment and effectiveness at reducing excessive daytime sleepiness compared with continuous positive airway pressure (CPAP) remains limited. Objectives: To establish if alarm-based supine-avoidance treatment in patients with supine-isolated OSA is noninferior to CPAP in reducing daytime sleepiness. Methods: After baseline questionnaire administration and in-home supine-time and polysomnography assessments, patients with supine-isolated OSA and Epworth Sleepiness Scale scores ⩾8 were randomized to ⩾6 weeks of supine-avoidance or CPAP treatment, followed by crossover to the remaining treatment with repeat assessments. Noninferiority was assessed from change in Epworth Sleepiness Scale with supine avoidance compared with CPAP using a prespecified noninferiority margin of 1.5. Average nightly treatment use over all nights and treatment efficacy and effectiveness at reducing respiratory disturbances were also compared between treatments. Results: The reduction in sleepiness score with supine avoidance (mean [95% confidence interval], -1.9 [-2.8 to -1.0]) was noninferior to that with CPAP (-2.4 [-3.3 to -1.4]) (supine avoidance-CPAP difference, -0.4 [-1.3 to 0.6]), and the lower confidence limit did not cross the noninferiority margin of 1.5 (P = 0.021). Average treatment use was higher with supine avoidance compared with CPAP (mean ± standard deviation, 5.7 ± 2.4 vs. 3.9 ± 2.7 h/night; P < 0.001). Conclusions: In patients with supine-isolated OSA, vibrotactile supine alarm device therapy is noninferior to CPAP for reducing sleepiness and shows superior treatment adherence. Clinical trial registered with www.anzctr.org.au (ACTRN 12613001242718).

Technologies for sleep monitoring at home: wearables and nearables

Sleep is an essential part of our lives and daily sleep monitoring is crucial for maintaining good health and well-being. Traditionally, the gold standard method for sleep monitoring is polysomnography using various sensors attached to the body; however, it is limited with regards to long-term sleep monitoring in a home environment. Recent advancements in wearable and nearable technology have made it possible to monitor sleep at home. In this review paper, the technologies that are currently available for sleep stages and sleep disorder monitoring at home are reviewed using wearable and nearable devices. Wearables are devices that are worn on the body, while nearables are placed near the body. These devices can accurately monitor sleep stages and sleep disorder in a home environment. In this study, the benefits and limitations of each technology are discussed, along with their potential to improve sleep quality.

Convolutional Neural Networks for the Real-Time Monitoring of Vital Signs Based on Impulse Radio Ultrawide-Band Radar during Sleep

Vital signs provide important biometric information for managing health and disease, and it is important to monitor them for a long time in a daily home environment. To this end, we developed and evaluated a deep learning framework that estimates the respiration rate (RR) and heart rate (HR) in real time from long-term data measured during sleep using a contactless impulse radio ultrawide-band (IR-UWB) radar. The clutter is removed from the measured radar signal, and the position of the subject is detected using the standard deviation of each radar signal channel. The 1D signal of the selected UWB channel index and the 2D signal applied with the continuous wavelet transform are entered as inputs into the convolutional neural-network-based model that then estimates RR and HR. From 30 recordings measured during night-time sleep, 10 were used for training, 5 for validation, and 15 for testing. The average mean absolute errors for RR and HR were 2.67 and 4.78, respectively. The performance of the proposed model was confirmed for long-term data, including static and dynamic conditions, and it is expected to be used for health management through vital-sign monitoring in the home environment.

Sleep Apnea Detection Using Wavelet Scattering Transformation and Random Forest Classifier

Obstructive Sleep Apnea (OSA) is a common sleep-breathing disorder that highly reduces the quality of human life. The most powerful method for the detection and classification of sleep apnea is the Polysomnogram. However, this method is time-consuming and cost-inefficient. Therefore, several methods focus on using electrocardiogram (ECG) signals to detect sleep apnea. This paper proposed a novel automated approach to detect and classify apneic events from single-lead ECG signals. Wavelet Scattering Transformation (WST) was applied to the ECG signals to decompose the signal into smaller segments. Then, a set of features, including higher-order statistics and entropy-based features, was extracted from the WST coefficients to formulate a search space. The obtained features were fed to a random forest classifier to classify the ECG segments. The experiment was validated using the 10-fold and hold-out cross-validation methods, which resulted in an accuracy of 91.65% and 90.35%, respectively. The findings were compared with different classifiers to show the significance of the proposed approach. The proposed approach achieved better performance measures than most of the existing methodologies.