1618 Identification of Noxious Events for Newborn Infants with a Neural Network. (October 2012)

• Record Type: Journal Article
• Authors: Bressan, N; McGregor, C; Blount, M; Ebling, M; Sow, D; James, A
• Is Part Of: Archives of disease in childhood. Volume 97(2012)Supplement 2
• Page Start: A458
• Page End: A458
• Publication Date: 2012-10
• Subjects: Children -- Diseases -- Periodicals; Infants -- Diseases -- Periodicals; 618.920005
• DOI: 10.1136/archdischild-2012-302724.1618
• Languages: English
• ISSNs: 0003-9888

Abstract

Abstract : Background: Recognition of pain experienced by immature and/or critically ill newborn infants in the Neonatal Intensive Care Unit remains a challenge despite the use of objective scoring systems that depend on physiological and behavioural parameters. We consider there is a need to identify pain using only physiological data streams. Methods: Data were collected from three preterm male, gestational age 27.25±0.95 weeks (mean±SD), birth weight 941.25±189.31 grams. Heart rate (HR), mean arterial pressure (MAP), respiratory rate (RR), blood oxygen saturation (SpO2 ) were considered for the NN Input Vector. NN's output were set to '1' for noxious stimuli pattern (NSP) define as: HR≥160 AND MAP≥55 AND RR≥40 AND SpO2 < < 90 1 . Artifact events were captured in previous study 2 and integrated with real-time physiological data streams. In this study we correlated the nociceptive event identified by NN with the artifact nociceptive event. Results: Events 'vascular access' and 'reintubation' statistically coincide with the NSP defined in 100%. Event 'routine care' coincides in 52.67% with NSP. HR≥160 was evident in 24.6% of the event occurrence. The events were successfully identified by the NN, shown figure 1 . Conclusion: This study showed correlation of artifact nociceptive event with the physiological data streams NN patterns verifying a positive relation between nociceptive response and non-invasive physiological response. NN developed previously proved to be an accurate tool for deployment in a clinical decision support system. References: Submitted to ESCTAIC 2012, Romania. 1st ACM International Health Informatics Symposium, 647–655.