Suppress False Arrhythmia Alarms of ICU Monitors Using Heart Rate Estimation Based on Combined Arterial Blood Pressure and Ecg Analysis

Abstract

Intensive care unit (ICU) monitors generate a high rate of false alarms when physiological signals are severely corrupted by noise. To suppress the false life-threatening heart rate (HR)-related electrocardiogram arrhythmia alarms, data derived from arterial blood pressure (ABP) signal were used. A new ABP signal quality index (SQI) was designed based upon the combination of two previously reported signal quality measures. HR was then tracked based on beat detection from ABP and a Kalman filter with a SQI-modified update sequence. Using the SQI to reject noisy ABP data, false HR-related arrhythmia alarms were rejected by comparing the estimated ABP-derived HR to the monitor's HR threshold. The algorithm was evaluated on 707 episodes of extreme bradycardia and 1877 episodes of extreme tachycardia alarms produced by a commercial ICU monitoring system. The false alarm reduction rate was 74.13% and 53.81%, and the corresponding true alarm acceptance rate was 99.60% and 99.58% for extreme bradycardia and extreme tachycardia respectively. Combining ECG and ABP information therefore provides a significant reduction in false alarms with minimal impact on true alarms.