Arrhythmia Compensation Of A Predictive Noninvasive Blood Pressure Algorithm

This paper presents an approach to the problem of arrhythmia management in noninvasive blood pressure (NIBP) measurement using the oscillometric technique. Algorithms which predict or match oscillometric pulse amplitudes as a function of cuff pressure may not respond to the large time variations induced by certain arrhythmias. A voting algorithm was developed which compensates for errors in prediction by scoring pulse characteristics based upon criteria of reasonableness which were derived by expert analysis of a large data base of atypical pulse waveforms. The algorithm has been implemented in a transport vital signs monitor. Clinical testing has demonstrated that the accuracy of the blood pressure measurements are well within AAMI specifications. Use in patient environments has demonstrated that the system is very robust in the presence of patient arrhythmias.