Smart Newborn: A Tool for Early Prediction of a Severe Preterm Newborn Illness

Abstract

Sepsis is a major cause of morbidity, mortality and increased healthcare costs among preterm babies. Infants are often diagnosed when seriously ill, which decreases the probability for prompt and complete recovery. To study on sepsis prediction, we developed a medical informatics system prototype, “Smart Newborn” that retains and analyses patient based data. As initial study, we illustrate the evolution of 30 minutes pulse rate (PR) histogram samples of a representative patient within 48 hours before sepsis suspicion, which point out that PR based analysis outcomes are in parallel with previous studies based on heart rate (HR) data obtained from ECG leads. Monitoring based on ECG leads for long time is not the preferred method in many Neonatal Intensive Care Units (NICUs) due to probable skin sensitivity reactions at contact points. Therefore, capturing and visualizing the same disease signs using PR data makes our process more practical and accessible.