Recognition of cardiogenic artifact in pediatric capnograms

The capnograph is in regular use as a respiratory monitor. From measurements of its trace, calculations of breathing rate and end-tidal carbon dioxide can be made. Unfortunately, the reliability of these calculations depends on the quality of the signal. In the case of cardiogenic oscillations, the averaged results may be grossly inaccurate and, therefore, misleading. This paper describes a method for detecting such cardiogenic oscillations and removing their effect from the final results of capnogram analysis. The algorithm used resulted in a sensitivity of 99.6% and a specificity of 99.2%, when compared with manual scoring. The criteria could easily be incorporated into equipment software and are a step in the direction of "intelligent" monitoring.