How can we … use machine learning and computer vision to create a second set of eyes for NICU nurses?
6 May 2025
Premature and poorly newborns need round-the-clock care, with clinical decisions guided by measurements, such as vital sign monitoring, as well as subjective observations by clinical staff. These observations can provide crucial information, such as detection of seizures, lethargic behaviour and other abnormal movements. However, nurses in neonatal intensive care units (NICUs) care for multiple babies at once, who are often in covered incubators. Even with ‘continuous monitoring,’ significant events such as seizures can be missed due to their transient, fleeting nature. Furthermore, long-term changes, such as lethargic behaviour can be potentially indicative of sepsis, one of the most serious conditions for a newborn. Nursing schedules and shift patterns mean these trends are difficult to identify.