Consent-driven, semi-automated data collection during birth and newborn resuscitation: Insights from the NewbornTime study.

Accurate observations at birth and during newborn resuscitation are fundamental for quality improvement initiatives and research. However, manual data collection methods often lack consistency and objectivity, are not scalable, and may raise privacy concerns. The NewbornTime project aims to develop an AI system that generates accurate timelines from birth and newborn resuscitation events by automated video recording and processing, providing a source of objective and consistent data. This work aims to describe the implementation of the data collection system that is necessary to support the project's purpose. Videos were recorded using thermal sensors in labor rooms and thermal and visible light cameras in resuscitation rooms. Consent from mothers was obtained before birth, and healthcare providers were given the option to delete videos by opting out on a case-by-case basis. The video collection process was designed to minimize interference with ongoing treatment and not impose unnecessary burden on healthcare providers. At Stavanger University Hospital, 1012 thermal videos of birth and 274 visible light videos of newborn stabilization and resuscitation have been collected during the data collection period from November 2021 to June 2025. The utilization of automated data collection and AI video processing around birth may allow for consistent and enhanced documentation, quality improvement initiatives, and research opportunities on sequence, timing and duration of treatment activities during acute events, with less effort needed for capturing data and improved privacy for participants.