MSED: A Multi-Modal Sleep Event Detection Model for Clinical Sleep Analysis

Olesen, Alexander Neergaard; Jennum, Poul; Mignot, Emmanuel; Sørensen, Helge Bjarup Dissing

doi:10.1109/tbme.2023.3252368

Cited by 4 publications

(2 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These should be mixed to reduce individual scorer bias. Other studies have shown that combining different micro-events during sleep improves results 21 . Therefore, we aim to further improve the results by adding other micro-events during sleep and sleep staging.…”

Section: Discussionmentioning

confidence: 97%

“…They improved this approach in 2020 20 by using different setups for their EEG channels, and achieved similar results on the test set with only a third of the training data. This approach was later extended by Zahid et al 21 by combining arousal detection with leg movement and sleep-disordered breathing, presumably using the same test set of 1000 male participants. They also included the correlation between the calculated ArI and the manually scored ArI in their evaluation.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

State-of-the-art Sleep Arousal Detection Evaluated on a Comprehensive Clinical Dataset

Ehrlich,

Sehr,

Brandt

et al. 2024

Preprint

View full text Add to dashboard Cite

Aiming to apply automatic arousal detection to support sleep laboratories, we evaluated an optimized, state-of-the-art approach using data from daily work in our university hospital sleep laboratory. Therefore, a machine learning algorithm was trained and evaluated on 3423 polysomnograms of people suffering from various sleep disorders. The model architecture is a U-net that accepts 50 Hz signals as input. We compared this algorithm with models trained on publicly available datasets, and evaluated these models using our clinical dataset, particularly with regard to the effects of different sleep disorders. In an effort to evaluate clinical relevance, we designed a metric based on the error of the predicted arousal index. Our models achieve an area under the precision recall curve (AUPRC) of up to 0.83 and F1 scores of up to 0.81. The model trained on our data showed no age or gender bias and no significant negative effect regarding sleep disorders on model performance compared to healthy sleep. In contrast, models trained on public datasets showed a small to moderate negative effect (calculated using Cohen's d) of sleep disorders on model performance. Therefore, we conclude that state-of-the-art arousal detection on our clinical data is possible with our model architecture. Thus, our results support the general recommendation to use a clinical dataset for training if the model is to be applied to clinical data.

show abstract

Section: Discussionmentioning

confidence: 97%

Section: Related Workmentioning

confidence: 99%

State-of-the-art Sleep Arousal Detection Evaluated on a Comprehensive Clinical Dataset

Ehrlich,

Sehr,

Brandt

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

State-of-the-art sleep arousal detection evaluated on a comprehensive clinical dataset

Ehrlich,

Sehr,

Brandt

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Aiming to apply automatic arousal detection to support sleep laboratories, we evaluated an optimized, state-of-the-art approach using data from daily work in our university hospital sleep laboratory. Therefore, a machine learning algorithm was trained and evaluated on 3423 polysomnograms of people with various sleep disorders. The model architecture is a U-net that accepts 50 Hz signals as input. We compared this algorithm with models trained on publicly available datasets, and evaluated these models using our clinical dataset, particularly with regard to the effects of different sleep disorders. In an effort to evaluate clinical relevance, we designed a metric based on the error of the predicted arousal index. Our models achieve an area under the precision recall curve (AUPRC) of up to 0.83 and F1 scores of up to 0.81. The model trained on our data showed no age or gender bias and no significant negative effect regarding sleep disorders on model performance compared to healthy sleep. In contrast, models trained on public datasets showed a small to moderate negative effect (calculated using Cohen's d) of sleep disorders on model performance. Therefore, we conclude that state-of-the-art arousal detection on our clinical data is possible with our model architecture. Thus, our results support the general recommendation to use a clinical dataset for training if the model is to be applied to clinical data.

show abstract