Development of Dynamic Real-Time Robust Multivariable Monitoring Tool for Neonatology Data

• Activity Number: 470
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Joshua Callaway*+ and Keith Williams and Jeff Kaiser and Zoran Bursac and Morris Cranmer
• Companies: and University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences
• Keywords: Loess Regression ; Dynamic Graphics ; Real Time Analysis ; R ; Multivariable Analysis ; Autoregulation Index

Abstract

As real-time monitoring of salient variables has become increasingly important in many realms of statistics, use has not reached its full implementation in neonatal data. Premature infants constitute a population in which monitoring particular associations between certain variables serves a preventative need. Using R, we have developed a demonstration of how three crucial variables may be modeled through a real-time monitoring algorithm to produce the autoregulation index, a statistic which displays a vital physiologic status. Not only may the successful implementation of this tool save costs in the medical and health arenas, but it has potential for great flexibility. Any relationship in life that requires instantaneous measurement of association between two continuous variables at given levels of a third categorical variable can be computed through this model. Due to its utilization through R's open-source nature, the tool may pervade areas where real-time monitoring using more expensive and established software deems implementation uneconomical.