Abstract
The aim of this thesis was to take the important next steps to understand the role of sleep in preterm development. Subsequently, the groundwork is laid to design an intervention aimed to improve sleep of preterm infants admitted to neonatal intensive care units. As has been highlighted in chapter 7, sleep is an important behavioral state during development and a potential driver of early brain connectivity. The retrospective cohort study in chapter 8 built upon the existing animal research by proving a positive association between active sleep around 30 weeks postmenstrual age and a higher white matter volume at term equivalent age. Furthermore, in chapter 9 we suggest that active sleep might protect brain development in the most vulnerable preterm infants.
However, as shown in chapter 10, sleep is barely considered during most clinical handovers and ward rounds in the neonatal intensive care unit. Furthermore, healthcare professionals have trouble distinguishing active sleep from wake, which poses the risk that active sleep is disproportionately disturbed during standard care practices.
To improve sleep in the neonatal intensive care unit, we should first have a good clinical understanding of preterm sleep physiology and sleep quality. Subsequently, we can assess how sleep changes following an intervention. In order to do this, the current literature on sleep assessment in preterm infants was reviewed (chapter 2 and chapter 3) and combined into the first reliable and validated manual sleep classification system for preterm infants between 25 and 37 weeks postmenstrual age; the “Behavioral Sleep stage classification for Preterm Infants” (BeSSPI; chapter 4). Using the BeSSPI, two automated classifiers were developed to continuously assess sleep stages. The radar-based classifier (chapter 6) has currently only been used on retrospective data. However, the “Sleep Well Baby” classifier (chapter 5) is running bedside in our neonatal intensive care unit as a pilot.
However, as shown in chapter 10, sleep is barely considered during most clinical handovers and ward rounds in the neonatal intensive care unit. Furthermore, healthcare professionals have trouble distinguishing active sleep from wake, which poses the risk that active sleep is disproportionately disturbed during standard care practices.
To improve sleep in the neonatal intensive care unit, we should first have a good clinical understanding of preterm sleep physiology and sleep quality. Subsequently, we can assess how sleep changes following an intervention. In order to do this, the current literature on sleep assessment in preterm infants was reviewed (chapter 2 and chapter 3) and combined into the first reliable and validated manual sleep classification system for preterm infants between 25 and 37 weeks postmenstrual age; the “Behavioral Sleep stage classification for Preterm Infants” (BeSSPI; chapter 4). Using the BeSSPI, two automated classifiers were developed to continuously assess sleep stages. The radar-based classifier (chapter 6) has currently only been used on retrospective data. However, the “Sleep Well Baby” classifier (chapter 5) is running bedside in our neonatal intensive care unit as a pilot.
Original language | English |
---|---|
Awarding Institution |
|
Supervisors/Advisors |
|
Award date | 18 Jan 2024 |
Publisher | |
Print ISBNs | 978-94-6473-318-1 |
DOIs | |
Publication status | Published - 18 Jan 2024 |
Keywords
- sleep
- preterm
- NICU
- infant
- brain development
- neonate
- classifier
- machine learning
- developmental care
- hospital