Retrospective validation of automatic sleep analysis with grey areas model for human-in-the-loop scoring approach

State-of-the-art automatic sleep staging methods have demonstrated comparable reliability and superior time efficiency to manual sleep staging. However, fully automatic black-box solutions are difficult to adapt into clinical workflow due to the lack of transparency in decision-making processes. Transparency would be crucial for interaction between automatic methods and the work of sleep experts, i.e., in human-in-the-loop applications. To address these challenges, we propose an automatic sleep staging model (aSAGA) that effectively utilises both electroencephalography and electro-oculography channels while incorporating transparency of uncertainty in the decision-making process. We validated the model through extensive retrospective testing using a range of datasets, including open-access, clinical, and research-driven sources. Our channel-wise ensemble model, trained on both electroencephalography and electro-oculography signals, demonstrated robustness and the ability to generalise across various types of sleep recordings, including novel self-applied home polysomnography. Additionally, we compared model uncertainty with human uncertainty in sleep staging and studied various uncertainty mapping metrics to identify ambiguous regions, or "grey areas", that may require manual re-evaluation. The validation of this grey area concept revealed its potential to enhance sleep staging accuracy and to highlight regions in the recordings where sleep experts may struggle to reach a consensus. In conclusion, this study provides a technical basis and understanding of automatic sleep staging uncertainty. Our approach has the potential to improve the integration of automatic sleep staging into clinical practice; however, further studies are needed to test the model prospectively in real-world clinical settings and human-in-the-loop scoring applications.

Bridging AI and Clinical Practice: Integrating Automated Sleep Scoring Algorithm with Uncertainty-Guided Physician Review

Purpose

This study aims to enhance the clinical use of automated sleep-scoring algorithms by incorporating an uncertainty estimation approach to efficiently assist clinicians in the manual review of predicted hypnograms, a necessity due to the notable inter-scorer variability inherent in polysomnography (PSG) databases. Our efforts target the extent of review required to achieve predefined agreement levels, examining both in-domain (ID) and out-of-domain (OOD) data, and considering subjects' diagnoses.

Patients and Methods

A total of 19,578 PSGs from 13 open-access databases were used to train U-Sleep, a state-of-the-art sleep-scoring algorithm. We leveraged a comprehensive clinical database of an additional 8832 PSGs, covering a full spectrum of ages (0-91 years) and sleep-disorders, to refine the U-Sleep, and to evaluate different uncertainty-quantification approaches, including our novel confidence network. The ID data consisted of PSGs scored by over 50 physicians, and the two OOD sets comprised recordings each scored by a unique senior physician.

Results

U-Sleep demonstrated robust performance, with Cohen's kappa (K) at 76.2% on ID and 73.8-78.8% on OOD data. The confidence network excelled at identifying uncertain predictions, achieving AUROC scores of 85.7% on ID and 82.5-85.6% on OOD data. Independently of sleep-disorder status, statistical evaluations revealed significant differences in confidence scores between aligning vs discording predictions, and significant correlations of confidence scores with classification performance metrics. To achieve κ ≥ 90% with physician intervention, examining less than 29.0% of uncertain epochs was required, substantially reducing physicians' workload, and facilitating near-perfect agreement.

Conclusion

Inter-scorer variability limits the accuracy of the scoring algorithms to ~80%. By integrating an uncertainty estimation with U-Sleep, we enhance the review of predicted hypnograms, to align with the scoring taste of a responsible physician. Validated across ID and OOD data and various sleep-disorders, our approach offers a strategy to boost automated scoring tools' usability in clinical settings.

Single-channel EOG sleep staging on a heterogeneous cohort of subjects with sleep disorders

Objective.Sleep staging based on full polysomnography is the gold standard in the diagnosis of many sleep disorders. It is however costly, complex, and obtrusive due to the use of multiple electrodes. Automatic sleep staging based on single-channel electro-oculography (EOG) is a promising alternative, requiring fewer electrodes which could be self-applied below the hairline. EOG sleep staging algorithms are however yet to be validated in clinical populations with sleep disorders.Approach.We utilized the SOMNIA dataset, comprising 774 recordings from subjects with various sleep disorders, including insomnia, sleep-disordered breathing, hypersomnolence, circadian rhythm disorders, parasomnias, and movement disorders. The recordings were divided into train (574), validation (100), and test (100) groups. We trained a neural network that integrated transformers within a U-Net backbone. This design facilitated learning of arbitrary-distance temporal relationships within and between the EOG and hypnogram.Main results.For 5-class sleep staging, we achieved median accuracies of 85.0% and 85.2% and Cohen's kappas of 0.781 and 0.796 for left and right EOG, respectively. The performance using the right EOG was significantly better than using the left EOG, possibly because in the recommended AASM setup, this electrode is located closer to the scalp. The proposed model is robust to the presence of a variety of sleep disorders, displaying no significant difference in performance for subjects with a certain sleep disorder compared to those without.Significance.The results show that accurate sleep staging using single-channel EOG can be done reliably for subjects with a variety of sleep disorders.

InsightSleepNet: the interpretable and uncertainty-aware deep learning network for sleep staging using continuous Photoplethysmography

BACKGROUND

This study was conducted to address the existing drawbacks of inconvenience and high costs associated with sleep monitoring. In this research, we performed sleep staging using continuous photoplethysmography (PPG) signals for sleep monitoring with wearable devices. Furthermore, our aim was to develop a more efficient sleep monitoring method by considering both the interpretability and uncertainty of the model's prediction results, with the goal of providing support to medical professionals in their decision-making process.

METHOD

The developed 4-class sleep staging model based on continuous PPG data incorporates several key components: a local attention module, an InceptionTime module, a time-distributed dense layer, a temporal convolutional network (TCN), and a 1D convolutional network (CNN). This model prioritizes both interpretability and uncertainty estimation in its prediction results. The local attention module is introduced to provide insights into the impact of each epoch within the continuous PPG data. It achieves this by leveraging the TCN structure. To quantify the uncertainty of prediction results and facilitate selective predictions, an energy score estimation is employed. By enhancing both the performance and interpretability of the model and taking into consideration the reliability of its predictions, we developed the InsightSleepNet for accurate sleep staging.

RESULT

InsightSleepNet was evaluated using three distinct datasets: MESA, CFS, and CAP. Initially, we assessed the model's classification performance both before and after applying an energy score threshold. We observed a significant improvement in the model's performance with the implementation of the energy score threshold. On the MESA dataset, prior to applying the energy score threshold, the accuracy was 84.2% with a Cohen's kappa of 0.742 and weighted F1 score of 0.842. After implementing the energy score threshold, the accuracy increased to a range of 84.8-86.1%, Cohen's kappa values ranged from 0.75 to 0.78 and weighted F1 scores ranged from 0.848 to 0.861. In the case of the CFS dataset, we also noted enhanced performance. Before the application of the energy score threshold, the accuracy stood at 80.6% with a Cohen's kappa of 0.72 and weighted F1 score of 0.808. After thresholding, the accuracy improved to a range of 81.9-85.6%, Cohen's kappa values ranged from 0.74 to 0.79 and weighted F1 scores ranged from 0.821 to 0.857. Similarly, on the CAP dataset, the initial accuracy was 80.6%, accompanied by a Cohen's kappa of 0.73 and weighted F1 score was 0.805. Following the application of the threshold, the accuracy increased to a range of 81.4-84.3%, Cohen's kappa values ranged from 0.74 to 0.79 and weighted F1 scores ranged from 0.813 to 0.842. Additionally, by interpreting the model's predictions, we obtained results indicating a correlation between the peak of the PPG signal and sleep stage classification.

CONCLUSION

InsightSleepNet is a 4-class sleep staging model that utilizes continuous PPG data, serves the purpose of continuous sleep monitoring with wearable devices. Beyond its primary function, it might facilitate in-depth sleep analysis by medical professionals and empower them with interpretability for intervention-based predictions. This capability can also support well-informed clinical decision-making, providing valuable insights and serving as a reliable second opinion in medical settings.

Somnotate: A probabilistic sleep stage classifier for studying vigilance state transitions

Electrophysiological recordings from freely behaving animals are a widespread and powerful mode of investigation in sleep research. These recordings generate large amounts of data that require sleep stage annotation (polysomnography), in which the data is parcellated according to three vigilance states: awake, rapid eye movement (REM) sleep, and non-REM (NREM) sleep. Manual and current computational annotation methods ignore intermediate states because the classification features become ambiguous, even though intermediate states contain important information regarding vigilance state dynamics. To address this problem, we have developed "Somnotate"-a probabilistic classifier based on a combination of linear discriminant analysis (LDA) with a hidden Markov model (HMM). First we demonstrate that Somnotate sets new standards in polysomnography, exhibiting annotation accuracies that exceed human experts on mouse electrophysiological data, remarkable robustness to errors in the training data, compatibility with different recording configurations, and an ability to maintain high accuracy during experimental interventions. However, the key feature of Somnotate is that it quantifies and reports the certainty of its annotations. We leverage this feature to reveal that many intermediate vigilance states cluster around state transitions, whereas others correspond to failed attempts to transition. This enables us to show for the first time that the success rates of different types of transition are differentially affected by experimental manipulations and can explain previously observed sleep patterns. Somnotate is open-source and has the potential to both facilitate the study of sleep stage transitions and offer new insights into the mechanisms underlying sleep-wake dynamics.

Decentralized Data-Privacy Preserving Deep-Learning Approaches for Enhancing Inter-Database Generalization in Automatic Sleep Staging

Automatic sleep staging has been an active field of development. Despite multiple efforts, the area remains a focus of research interest. Indeed, while promising results have reported in past literature, uptake of automatic sleep scoring in the clinical setting remains low. One of the current issues regards the difficulty to generalization performance results beyond the local testing scenario, i.e. across data from different clinics. Issues derived from data-privacy restrictions, that generally apply in the medical domain, pose additional difficulties in the successful development of these methods. We propose the use of several decentralized deep-learning approaches, namely ensemble models and federated learning, for robust inter-database performance generalization and data-privacy preservation in automatic sleep staging scenario. Specifically, we explore four ensemble combination strategies (max-voting, output averaging, size-proportional weighting, and Nelder-Mead) and present a new federated learning algorithm, so-called sub-sampled federated stochastic gradient descent (ssFedSGD). To evaluate generalization capabilities of such approaches, experimental procedures are carried out using a leaving-one-database-out direct-transfer scenario on six independent and heterogeneous public sleep staging databases. The resulting performance is compared with respect to two baseline approaches involving single-database and centralized multiple-database derived models. Our results show that proposed decentralized learning methods outperform baseline local approaches, and provide similar generalization results to centralized database-combined approaches. We conclude that these methods are more preferable choices, as they come with additional advantages concerning improved scalability, flexible design, and data-privacy preservation.

Challenges of Applying Automated Polysomnography Scoring at Scale

Automatic polysomnography analysis can be leveraged to shorten scoring times, reduce associated costs, and ultimately improve the overall diagnosis of sleep disorders. Multiple and diverse strategies have been attempted for implementation of this technology at scale in the routine workflow of sleep centers. The field, however, is complex and presents unsolved challenges in a number of areas. Recent developments in computer science and artificial intelligence are nevertheless closing the gap. Technological advances are also opening new pathways for expanding our current understanding of the domain and its analysis.

The Effect of Coupled Electroencephalography Signals in Electrooculography Signals on Sleep Staging Based on Deep Learning Methods

The influence of the coupled electroencephalography (EEG) signal in electrooculography (EOG) on EOG-based automatic sleep staging has been ignored. Since the EOG and prefrontal EEG are collected at close range, it is not clear whether EEG couples in EOG or not, and whether or not the EOG signal can achieve good sleep staging results due to its intrinsic characteristics. In this paper, the effect of a coupled EEG signal in an EOG signal on automatic sleep staging is explored. The blind source separation algorithm was used to extract a clean prefrontal EEG signal. Then the raw EOG signal and clean prefrontal EEG signal were processed to obtain EOG signals coupled with different EEG signal contents. Afterwards, the coupled EOG signals were fed into a hierarchical neural network, including a convolutional neural network and recurrent neural network for automatic sleep staging. Finally, an exploration was performed using two public datasets and one clinical dataset. The results showed that using a coupled EOG signal could achieve an accuracy of 80.4%, 81.1%, and 78.9% for the three datasets, slightly better than the accuracy of sleep staging using the EOG signal without coupled EEG. Thus, an appropriate content of coupled EEG signal in an EOG signal improved the sleep staging results. This paper provides an experimental basis for sleep staging with EOG signals.

Interpretation and further development of the hypnodensity representation of sleep structure

Objective.The recently-introduced hypnodensity graph provides a probability distribution over sleep stages per data window (i.e. an epoch). This work explored whether this representation reveals continuities that can only be attributed to intra- and inter-rater disagreement of expert scorings, or also to co-occurrence of sleep stage-dependent features within one epoch.Approach.We proposed a simplified model for time series like the ones measured during sleep, and a second model to describe the annotation process by an expert. Generating data according to these models, enabled controlled experiments to investigate the interpretation of the hypnodensity graph. Moreover, the influence of both the supervised training strategy, and the used softmax non-linearity were investigated. Polysomnography recordings of 96 healthy sleepers (of which 11 were used as independent test set), were subsequently used to transfer conclusions to real data.Main results.A hypnodensity graph, predicted by a supervised neural classifier, represents the probability with which the sleep expert(s) assigned a label to an epoch. It thus reflects annotator behavior, and is thereby only indirectly linked to the ratio of sleep stage-dependent features in the epoch. Unsupervised training was shown to result in hypnodensity graph that were slightly less dependent on this annotation process, resulting in, on average, higher-entropy distributions over sleep stages (H_unsupervised= 0.41 versusH_supervised= 0.29). Moreover, pre-softmax predictions were, for both training strategies, found to better reflect the ratio of sleep stage-dependent characteristics in an epoch, as compared to the post-softmax counterparts (i.e. the hypnodensity graph). In real data, this was observed from the linear relation between pre-softmax N3 predictions and the amount of delta power.Significance.This study provides insights in, and proposes new, representations of sleep that may enhance our comprehension about sleep and sleep disorders.