Reducing false arrhythmia alarms in the ICU using novel signal quality indices assessment method

The physiological signals such as the electrocardiogram (ECG) and arterial blood pressure (ABP) in the ICU are often severely corrupted by noise, artifact and missing data, producing large errors in the estimation of the characteristics of the signals values, leading to false alarms in ICU. In order to solve this problem, we started with the signal quality assessment of vital signals in intensive care patients using a derived signal quality index (SQI) to reveal the degree of signal quality. And then we use the SQI-weighted residual error of Kalman filters (KF) to complete the date fusion for evaluating the heart rate (HR). Finally, the algorithm of arrhythmia false alarm reduction in ICU monitors was developed based upon the method of combining SQIs and HR estimations derived from ECG waveform and ABP waveform recorded from ICU patients. Results show that the overall True Positive Rate (TPR), True Negative Rate (TNR) and overall score for the Event-1 are respectively 65%, 82%, and 53.19, for the Event-2, the TPR, TNR and overall score are 65%, 87%, and 54.64.