A Rule-Based Temporal Analysis Method for Online Health Analytics and Its Application for Real-Time Detection of Neonatal Spells

Neonatal spells are cardiorespiratory events that occur in newborn infants with variable combinations of cessation of breathing, decrease in blood oxygen saturation and decrease in heart rate. A system using real-time temporal analysis of physiological data streams to accurately detect pauses in breathing and changes in heart rate and oxygen saturation for classifying neonatal spells is described. The system uses a multidimensional online health analytics environment that supports the acquisition, transmission and real-time processing of high volume, high rate data. A family of algorithms has been developed using IBM InfoSphere Streams, a scalable middleware component for analysing multiple streams of data in real-time. Respiratory pauses are identified by accurately detecting breaths and calculating time intervals between breaths. Changes in heart rate and blood oxygen saturation are identified by both threshold breaches and the detection of relative change by assessing a sliding baseline and generating alerts when values fall out of range. Events detected in individual signals are synced together based on timestamps and assessed using a classifier based on clinical rules to determine a classification of neonatal spells. The output of these algorithms has been shown, in a single use case study with 24 hours of patient data, to detect clinically significant events in heart rate, blood oxygen saturation and pauses in breathing. The accuracy for detecting these is 97.8%, 98.3% and 98.9% respectively. The accuracy for determining spells classifications is 98.9%. Future research will focus on the clinical validation of these algorithms.