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 5-fold 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%, 68.5%) and demonstrated a MAE of 7.54 (95% CI: 5.36, 9.72), indicating good agreement (ICC = 0.889 [95% CI: 0.792, 0.942]) and a strong correlation (r = 0.892 [95% CI: 0.795, 0.945]) with the ground truth for AHI estimation. Furthermore, OSA severity estimation showed substantial agreement (κ = 0.780 [95% CI: 0.658, 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.

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.

Non-Contact VITAL Signs Monitoring of a Patient Lying on Surgical Bed Using Beamforming FMCW Radar.

Respiration and heartrates are important information for surgery. When the vital signs of the patient lying prone are monitored using radar installed on the back of the surgical bed, the surgeon's movements reduce the accuracy of these monitored vital signs. This study proposes a method for enhancing the monitored vital sign accuracies of a patient lying on a surgical bed using a 60 GHz frequency modulated continuous wave (FMCW) radar system with beamforming. The vital sign accuracies were enhanced by applying a fast Fourier transform (FFT) for range and beamforming which suppress the noise generated at different ranges and angles from the patient's position. The experiment was performed for a patient lying on a surgical bed with or without surgeon. Comparing a continuous-wave (CW) Doppler radar, the FMCW radar with beamforming improved almost 22 dB of signal-to-interference and noise ratio (SINR) for vital signals. More than 90% accuracy of monitoring respiration and heartrates was achieved even though the surgeon was located next to the patient as an interferer. It was analyzed using a proposed vital signal model included in the radar IF equation.

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.

Radar is a promising non-contact sensor for overnight polysomnography (PSG), the gold standard for diagnosing obstructive sleep apnea (OSA). This preliminary study aimed to demonstrate the feasibility of the automated detection of apnea-hypopnea events for OSA diagnosis based on 60 GHz frequency-modulated continuous-wave radar using convolutional recurrent neural networks. The dataset comprised 44 participants from an ongoing OSA cohort, recruited from July 2021 to April 2022, who underwent overnight PSG with a radar sensor. All PSG recordings, including sleep and wakefulness, were included in the dataset. Model development and evaluation were based on a five-fold cross-validation. The area under the receiver operating characteristic curve for the classification of 1-min segments ranged from 0.796 to 0.859. Depending on OSA severity, the sensitivities for apnea-hypopnea events were 49.0-67.6%, and the number of false-positive detections per participant was 23.4-52.8. The estimated apnea-hypopnea index showed strong correlations (Pearson correlation coefficient = 0.805-0.949) and good to excellent agreement (intraclass correlation coefficient = 0.776-0.929) with the ground truth. There was substantial agreement between the estimated and ground truth OSA severity (kappa statistics = 0.648-0.736). The results demonstrate the potential of radar as a standalone screening tool for OSA.

Sensor Fault Detection and Signal Restoration in Intelligent Vehicles.

This paper presents fault diagnosis logic and signal restoration algorithms for vehicle motion sensors. Because various sensors are equipped to realize automatic operation of the vehicle, defects in these sensors lead to severe safety issues. Therefore, an effective and reliable fault detection and recovery system should be developed. The primary idea of the proposed fault detection system is the conversion of measured wheel speeds into vehicle central axis information and the selection of a reference central axis speed based on this information. Thus, the obtained results are employed to estimate the speed for all wheel sides, which are compared with measured values to identify fault and recover the fault signal. For fault diagnosis logic, a conditional expression is derived with only two variables to distinguish between normal and fault; further, an analytical redundancy structure and a simple diagnostic logic structure are presented. Finally, an off-line test is conducted using test vehicle information to validate the proposed method; it demonstrates that the proposed fault detection and signal restoration algorithm can satisfy the control performance required for each sensor failure.

Analysis of adjacent fracture after percutaneous vertebroplasty: does intradiscal cement leakage really increase the risk of adjacent vertebral fracture?

OBJECTIVE: The purpose of this study was to evaluate the incidence and risk factors associated with adjacent vertebral fracture after percutaneous vertebroplasty (PVP) to treat osteoporotic vertebral compression fractures. We also investigated the effect of intradiscal cement leakage on adjacent vertebral fracture formation after PVP. MATERIALS AND METHODS: From January 2003 to March 2009, 188 patients (163 women, 25 men; mean age, 70.9 years; range, 42-92 years) who underwent 214 PVP sessions at 351 levels for osteoporotic vertebral compression fractures were retrospectively enrolled in this study. The effect of intradiscal cement leakage on new adjacent vertebral fracture formation after PVP was evaluated. Possible other risk factors were also analyzed using univariate and multivariate methods. The risk factors included age, gender, mean bone mineral density (BMD), the vertebral level treated, presence of an intravertebral cleft or cyst before treatment, kyphosis angle, wedge angle, and the injected cement volumes. RESULTS: During the follow-up periods, new adjacent vertebral fractures developed in 36 (10.3%) of 351 treated levels. For 91 (25.9%) levels, intradiscal cement leakage was detected on procedural fluoroscopic radiographs. There was no statistically significant association between intradiscal cement leakage and new adjacent vertebral compression fracture (p = 0.789). Among the other risk factors, only the vertebral levels treated, especially the thoracolumbar junction, showed a significant relationship to new adjacent vertebral fractures (univariate analysis, p = 0.037; multivariate analysis, p = 0.043). CONCLUSIONS: Intradiscal cement leakage does not seem to be related to subsequent adjacent vertebral compression fracture in patients who underwent PVP for treatment of an osteoporotic compression fracture. The thoracolumbar location of the initial compression fracture is the only factor correlated with an adjacent vertebral fracture after PVP.