Yuxi Luo

Polysomnography (PSG), the reference standard for sleep assessment, has limited scalability for large‐scale screening, driving the development of wearable and unobtrusive monitors. However, the physiological signals these devices capture are physiologically correlated with PSG channels but are not completely equivalent. Their reliability and validity for clinical use therefore require careful evaluation. The advent of artificial intelligence (AI) provides a technical bridge to close the representation gap between PSG signals and wearable/unobtrusive signals. This review outlines the core role of PSG signals in sleep assessment and sleep disorder diagnosis, including sleep structure analysis, respiratory event detection, and abnormal behavior identification. It then explores the current research progress of wearable and unobtrusive sleep monitoring devices, such as smartwatches, sleep monitoring pads, and radars, for sleep staging and the diagnosis of common sleep disorders such as sleep apnea, and discusses the use of AI in such devices, with a particular focus on building interpretable models to enhance the clinical trustworthiness of out‐of‐hospital sleep monitoring. This article aims to provide valuable insights into the development of low‐burden, easy‐to‐operate, trustworthy assessment tools for out‐of‐hospital sleep monitoring, with the goal of facilitating early detection, personalized management, and long‐term follow‐up of patients with sleep disorders.

Current ventilators rely on wall outlets or cylinders for oxygen supply, which limits their continuous use in the field or emergencies. In this study, we proposed a ventilator prototype that can achieve stand-alone oxygenated respiratory support, by designing and integrating a high-performance oxygen generator, and optimizing the control strategies of the whole system. Based on the designed oil-free scroll compressor and pressure swing adsorption (PSA) system, we first realized a mobile high-flow oxygen generator, which achieved an output flow greater than 17 L/min with an oxygen concentration of 93% ± 3%. The ventilator was also designed to synchronize with the respiratory state, to optimize the trigger performance for the pressure support of early inspiration, and reduce the gas supply in the late inspiratory phase to avoid pressure overshoot in the early expiratory phase. The respiratory synchronization of the integrated ventilator was estimated by the recorded chest movement of the subjects. Satisfactory respiratory synchronization was realized with an inspiratory trigger delay (ITD) time of less than 200 ms and sound respiratory waveform tracking. By regulating the PSA strategy, the oxygen generation and utilization efficiencies could be further improved. Ultimately, under the setting of inspiratory positive airway pressure (IPAP) at 10 cmH₂O, and expiratory positive airway pressure (EPAP) at 4 cmH₂O, we achieved non-invasive ventilation with a maximum oxygen concentration of 58% ± 1.75%. In conclusion, the proposed oxygen generator integrated ventilator could provide reliable oxygenated respiratory support in emergencies, such as on-site first aid, patient transport, and military field environments.

In recent years, substantial research has focused on automated systems for assessing depression levels using different types of data, such as audio and visual inputs. However, signals recorded from individuals with depression can be influenced by external factors, such as the recording equipment and environment, making it essential to create a system that is resilient to these interferences to maintain accuracy. This study introduces a fused-attention model for evaluating depression severity using enhanced multi-modal data inputs. Applying several pre-trained advanced models, this article incorporates audiovisual sequences with augmentation. The framework includes two novel components, which we termed as the FIE and VIE blocks, for extracting detailed facial and vocal features. The FIE block utilizes ResNet-18 to enhance the feature representation of video frames and integrates two types of attention mechanisms to capture spatial-temporal patterns. Meanwhile, the VIE block processes the Mel spectrogram of the audio signal, followed by an optimized Swin transformer block to extract auditory features. The model demonstrates strong performance, accurately identifying depression severity in 3-second audiovisual sequences with an 81.4% accuracy rate on the AVEC2014 dataset, and achieves a Kappa score of 0.731 and an MF1 index of 0.798. Furthermore, it shows high resilience to noise, underscoring its ability to mitigate the effects of recording equipment and environmental conditions in depression level estimation.

Single-channel EEG based sleep staging is of interest to researchers due to its broad application prospect in daily sleep monitoring recently. We proposed using contextual scalograms as input and developed a convolutional neural network with attention modules named Co-ScaleNet for sleep staging. The contextual scalograms were obtained by combining the same color channels of three original RGB scalograms from consecutive epochs, and a simple and efficient data augmentation was designed according to their various forms. The Co-ScaleNet consists of two main parts. Firstly, three parallel convolutional branches with attention modules correspondingly extract and fuse features from contextual scalograms at the top layers. The remaining part is a stack of lightweight blocks. We achieved an overall accuracy of 87.0% for healthy individuals, 84.7% for depressed patients. And we obtained comparable performance on the public Sleep-EDFx (82.8%), ISRUC (84.6%) and SHHS datasets (87.7%), including a high recall of N1. The contextual scalograms of R channel as input achieved the best performance, which conform to the features of interest in visual scoring. The attention modules improved the recall of N1 and N3. Overall, the contextual scalograms provided a novel scheme for both contextual information extraction and data augmentation. Our study successfully expanded its application to depression datasets, as well as patients with sleep apnea, demonstrating its wide applicability.

Introduction: A comfortable mattress should improve sleep quality. In this study, we sought to investigate the specific sleep parameters that could be affected by a mattress and explore any potential differences between the effects felt by each sex.Methods: A total of 20 healthy young adults (10 females and 20 males; 22.10 ± 1.25 years) participated in the experiments. A smart adjustable zoned air mattress was designed to maintain comfortable support, and an ordinary mattress was used for comparison. The participants individually spent four nights on these two mattresses in four orders for polysomnography (PSG) scoring. Sleep architecture, electroencephalogram (EEG) spectrum, and heart rate variability (HRV), which reflect the central and autonomic nervous activities, were used to compare the difference between the two mattresses.Results: An individual difference exited in sleep performance. The modes of influence of the mattresses were different between the sexes. The adjustable air mattress and the increase in experimental nights improved female participants' sleep efficiency, while male participants exhibited a smaller response to different mattresses. With an increasing number of experiment nights, both sexes showed increased REM and decreased N2 proportions; the N3 sleep proportion decreased in the male participants, and the heart rate decreased in both sexes. The performance of the EEG spectrum supports the above results. In addition, the adjustable air mattress weakened automatic nerve activity during N3 sleep in most participants. The female participants appeared to be more sensitive to mattresses. Experiment night was associated with psychological factors. There were differences in the results for this influence between the sexes.Conclusion: This study may shed some light on the differences between the ideal sleep environment of each sex.

Sleep state transitions are closely related to insomnia, drowsiness, and sleep maintenance. However, how the cortical network varies during such a transition process remains unclear. Changes in the cortical interaction during the short‐term process of sleep stage transitions were investigated. In all, 40 healthy young participants underwent overnight polysomnography. The phase transfer entropy of six frequency bands was obtained from 16 electroencephalography channels to assess the strength and direction of information flow between the cortical regions. Differences in the cortical network between the first and the last 10 s in a 40‐s transition period across wakefulness, N1, N2, N3, and rapid eye movement were, respectively, studied. Various frequency bands exhibited different patterns during the sleep stage transitions. It was found that the mutual transitions between the sleep stages were not necessarily the opposite. More significant changes were observed in the sleep deepening process than in the process of sleep awakening. During sleep stage transitions, changes in the inflow and outflow strength of various cortical regions led to regional differences, but for the entire sleep progress, such an imbalance did not intensify, and a dynamic balance was instead observed. The detailed findings of variations in cortical interactions during sleep stage transition promote understanding of sleep mechanism, sleep process, and sleep function. Additionally, it is expected to provide helpful clues for sleep improvement, like reducing the time required to fall asleep and maintaining sleep depth.

Depression, a common mental illness that seriously affects the psychological health of patients, is also thought to be associated with abnormal brain functional connectivity. This study aimed to explore the differences in the sleep-state functional network topology in depressed patients. A total of 25 healthy participants and 26 depressed patients underwent overnight 16-channel electroencephalography (EEG) examination. The cortical networks were constructed by using functional connectivity metrics of participants based on the weighted phase lag index (WPLI) between the EEG signals. The results indicated that depressed patients exhibited higher global efficiency and node strength than healthy participants. Furthermore, the depressed group indicated right-lateralization in the δ band. The top 30% of connectivity in both groups were shown in undirected connectivity graphs, revealing the distinct link patterns between the depressed and control groups. Links between the hemispheres were noted in the patient group, while the links in the control group were only observed within each hemisphere, and there were many long-range links inside the hemisphere. The altered sleep-state functional network topology in depressed patients may provide clues for a better understanding of the depression pathology. Overall, functional network topology may become a powerful tool for the diagnosis of depression.