Bayesian Fusion of Multiple Sensors for Reliable Heart Rate Detection

Automatic patient monitoring is an essential resource in hospitals for a good health care management. While alarms due to abnormal physiological conditions are important to deliver fast treatment, it can be also a source of unnecessary noise due to false alarms caused by electromagnetic interference or motion artifacts. This condition leads to stress, sleep disorders in patients and desensitization in staff. One significant source of false alarms are those related to heart rate, which is triggered when the heart rhythm of the patient is too fast or too slow. Other types of cardiac arrhythmia alarms also relies on a good detection of the HR. In order to avoid false alarms, it is important to create systems for reliable heart rate calculation. In this paper, the fusion of different physiological sensors is used o create a robust heart rate estimation. The algorithm uses a bayesian approach to fuse information from electrocardiogram, arterial blood pressure and photoplethysmogram. To validate this system, it was used twenty selected recordings from MIMIC database. A white gaussian noise was added in each waveform to simulate the worst case scenario. The proposed algorithm was compared with two other techniques (HRV index and majority voter) in addition to the individual analysis of each source. Results show that this bayesian fusion presents the lower error of 23%, while the other evaluated techniques presents an error rate of 35% (ECG Only), 40% (ABP Only), 41% (PPG Only), 37% (HRV Index) and 31% (Majority Voter). Therefore, the system shows good performance and requires simple computations, so it is very useful for real-time applications.