Real-time automated computerized detection of venous air emboli in dogs

My objective was to develop a real-time pattern recognition system to monitor the precordial Doppler and end-tidal CO2 for characteristic changes of venous air emboli. The system also must check the adequacy of the input signals, to allow for unattended operation. The sensitivity of the precordial Doppler monitoring of the resulting system was the focus of this study.

The computerized system electronically sampled systolic sounds, the amplitude envelope of Doppler pulsations, and, optionally, end-tidal CO2. Features were defined and calculated from the samples, the means and standard deviations of which were also calculated. During real-time test administrations of intravenous air in anesthetized dogs, each new sample was compared with previous statistics and, when parameters changed beyond calculated limits, an alarm was activated.

The sensitivity of the on-line system to an intravenous air injection of 0.025 ml/kg was 33%; to 0.05 ml/kg, 73%; to 0.1 ml/kg, 90%; and to 0.2 ml/kg, 100%. A confounding factor, air lodging in the veins, was detected in the smaller injections; when this was corrected, the sensitivity of the system improved beyond these results.

An on-line, real-time system, developed for continuous observation of precordial Doppler, has a sensitivity comparable to human observers. This system may improve clinical monitoring particularly in situations where the occurrence of a venous air embolism is not a high probability and, therefore, monitoring is not currently used because of its requirement for human observation. Systems such as the one described may allow many more patients to be monitored for this complication.