Scholarlink

Background <p>Frequent sleep disruption in preterm infants in neonatal intensive care units (NICU) is suspected to be associated with adverse neurodevelopmental outcomes. A&#xa0;continuous, noninvasive sleep monitoring solution could optimise care by aligning interventions with natural sleep–wake cycles. The algorithm for automatic sleep stage classification by Demme et&#xa0;al. [<CitationRef CitationID="CR5">5</CitationRef>] offers a&#xa0;promising approach, with an accuracy of 92.2 ± 0.01% compared to sleep stages measured in the sleep laboratory.</p> Objective <p>This study aims to evaluate the applicability of the sleep classification algorithm with regards to two main questions: (1)&#xa0;Does the algorithm’s performance vary with factors like pre-existing medical conditions or other patient characteristics in preterm infants? (2)&#xa0;How well does the algorithm capture sleep stage transitions compared to manual annotations?</p> Materials and methods <p>This study is a&#xa0;secondary analysis of data previously used by Demme et&#xa0;al. [<CitationRef CitationID="CR5">5</CitationRef>] to train models for automatic sleep stage classification in preterm infants. The models, which were trained on manual annotations of polysomnography (PSG) data, were applied to a&#xa0;diverse cohort of preterm infants. The performance was analysed in relation to medical conditions, sex, gestational age (GA), apnoea–hypopnea index (AHI), bodyweight at measurement, and Fleiss’ kappa (a&#xa0;measure of interrater variability). Sleep stage transition frequencies were compared between algorithmic and manual scoring.</p> Results <p>No significant performance differences could be observed with respect to any covariates, except for a&#xa0;reduction in performance among patients with low Fleiss’ kappa values, suggesting classification difficulties in cases in which manual raters have come to diverging assessments. Transition analysis revealed significant differences, likely due to the algorithm’s more sensitive ability to detect short sleep stage transitions.</p> Conclusion <p>The algorithm shows strong performance and generalisability. Despite some discrepancies in sleep stage transition detection, it appears suitable for noninvasive, continuous sleep monitoring in the NICU, supporting sleep-aware clinical care.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Impact of clinical covariates on the performance of an automatic sleep stage classification in preterm infants

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月20日

查看全文
  • Laura Schieder,
  • Nathalie Demme,
  • Maha Shadaydeh,
  • Claus Doerfel,
  • Stella Jähkel,
  • Joachim Denzler,
  • Knut Holthoff,
  • Hans Proquitté,
  • Jürgen Graf

摘要

Background

Frequent sleep disruption in preterm infants in neonatal intensive care units (NICU) is suspected to be associated with adverse neurodevelopmental outcomes. A continuous, noninvasive sleep monitoring solution could optimise care by aligning interventions with natural sleep–wake cycles. The algorithm for automatic sleep stage classification by Demme et al. [5] offers a promising approach, with an accuracy of 92.2 ± 0.01% compared to sleep stages measured in the sleep laboratory.

Objective

This study aims to evaluate the applicability of the sleep classification algorithm with regards to two main questions: (1) Does the algorithm’s performance vary with factors like pre-existing medical conditions or other patient characteristics in preterm infants? (2) How well does the algorithm capture sleep stage transitions compared to manual annotations?

Materials and methods

This study is a secondary analysis of data previously used by Demme et al. [5] to train models for automatic sleep stage classification in preterm infants. The models, which were trained on manual annotations of polysomnography (PSG) data, were applied to a diverse cohort of preterm infants. The performance was analysed in relation to medical conditions, sex, gestational age (GA), apnoea–hypopnea index (AHI), bodyweight at measurement, and Fleiss’ kappa (a measure of interrater variability). Sleep stage transition frequencies were compared between algorithmic and manual scoring.

Results

No significant performance differences could be observed with respect to any covariates, except for a reduction in performance among patients with low Fleiss’ kappa values, suggesting classification difficulties in cases in which manual raters have come to diverging assessments. Transition analysis revealed significant differences, likely due to the algorithm’s more sensitive ability to detect short sleep stage transitions.

Conclusion

The algorithm shows strong performance and generalisability. Despite some discrepancies in sleep stage transition detection, it appears suitable for noninvasive, continuous sleep monitoring in the NICU, supporting sleep-aware clinical care.