Blood Pressure Estimation based on Electrocardiograms

Abstract To overcome limitations of currently used blood pressure measurement devices in accuracy, continuity and comfort, we propose an approach for estimation from electrocardiogram (ECG) signals only. Thereby, statistical signal features are extracted the ECG which, eventually, serve as input to a random forest regression. The method is trained tested on MIMIC III waveform data with large range values. It obtains mean absolute error ± standard deviation 3.73 5.19 mmHg diastolic (DBP) 5.92 7.23 systolic (SBP), Pearson coefficients Υ DBP =0.92 SBP =0.91 respectively.