Intelligent Management System for Apnea Identification, 10-9037

Printer Friendly Version

Principal Investigators
Xiao Zhao
Keith A. Bartels

Inclusive Dates: 07/01/97 - 10/01/99

Background - Apnea, or the cessation of breathing, is recognized as a clinical problem affecting premature babies. Current technology for automatic apnea detection involves monitoring the baby's blood oxygen saturation percentage (SpO2). When the SpO2 level falls below a preset level for an extended time, the automatic monitoring system sounds an alarm to alert a care giver. A more desirable monitoring system would be one that predicts the onset of apnea before breathing stops so that the undesired drop in SpO2 can be avoided.

Approach - The vital signs of 12 premature infants with histories of apnea were monitored in the neonatal intensive care unit of a local hospital. The specific vital signs monitored were EKG, respiration, SpO2, heart rate, and respiration rate. The vital-sign information was obtained from the standard infant monitors in use at the hospital using an SwRI-designed and programmed data-acquisition system. Along with the baby's vital signs, measurements on the sound, light, and motion levels near the infant were also acquired using SwRI sensors and hardware designed specifically for this project.

The data-acquisition procedure was approved by the Institutional Review Board, and permission was obtained from the parents of each baby prior to data acquisition. Each of the babies had a history of apneic events, and four hours of data were obtained from each baby. After 12 data sets were acquired, the data were examined in the time and frequency domains for trends or events that could be used to predict the occurrence of apnea.

Accomplishments - A comprehensive review of all data sets showed no features, either in the time or frequency domains, that clearly predicted the onset of an apneic episode. However, one characteristic of the data was identified that may be indicative of problems in the development of a newborn baby's control of breathing, namely that all the babies showed a periodic change in their vital signs. For example, the magnitudes of the heart and respiration rates varied, sometimes by more than 100 percent, with a periodicity of about 2.5 minutes. This periodicity had not been previously reported in the scientific literature, and one hypothesis is that it may be an indicator of a developmental problem in breathing control. Because no control babies could be included in this initial study, the research team was unable to identify this feature as unique to premature infants with breathing problems.

Intelligent Systems, Advanced Computer and
Electronic Technology, and Automation Program
2000 IR&D Home SwRI Home