Goal and aims: Performance evaluation of automatic sleep staging on two-channel subcutaneous electroencephalography.
Focus technology: UNEEG medical's 24/7 electroencephalography SubQ (the SubQ device) with deep learning model U-SleepSQ.
Reference method/technology: Manually scored hypnograms from polysomnographic recordings.
Sample: Twenty-two healthy adults with 1-6 recordings per participant. The clinical study was registered at ClinicalTrials.gov with the identifier NCT04513743.
Design: Fine-tuning of U-Sleep in 11-fold cross-participant validation on 22 healthy adults. The resultant model was called U-SleepSQ.
Core analytics: Bland-Altman analysis of sleep parameters. Advanced multiclass model performance metrics: stage-specific accuracy, specificity, sensitivity, kappa (κ), and F1 score. Additionally, Cohen's κ coefficient and macro F1 score. Longitudinal and participant-level performance evaluation.
Additional analytics and exploratory analyses: Exploration of model confidence quantification. Performance vs. age, sex, body mass index, SubQ implantation hemisphere, normalized entropy, transition index, and scores from the following three questionnaires: Morningness-Eveningness Questionnaire, World Health Organization's 5-item Well-being Index, and Major Depression Inventory.
Core outcomes: There was a strong agreement between the focus and reference method/technology.
Important supplemental outcomes: The confidence score was a promising metric for estimating the reliability of each hypnogram classified by the system.
Core conclusion: The U-SleepSQ model classified hypnograms for healthy participants soon after implantation and longitudinally with a strong agreement with the gold standard of manually scored polysomnographics, exhibiting negligible temporal variation.