A Novel Frequency Domain Method for Estimating Blood Pressure from Photoplethysmogram

A novel method of estimating blood pressure (BP) from Photoplethysmogram (PPG) is provided. The first 15 points of the discrete cosine transform (DCT) sequence of the PPG signal are trained as inputs of the Backpropagation neural network (BPNN), the systolic blood pressure (SBP) and diastolic blood pressure (DBP) extracted from the Arterial blood pressure (ABP) signal which is corresponding to the PPG signal are used as outputs of the BPNN. Combining the idea of AdaBoost algorithm, 10 BPNN with different initial values are chosen as "weak predictors" to form a "strong predictor" for predicting blood pressure. The PhysioNet/CinC Challenge 2010 data set was used in this work for training and testing and more than 10,000 separate PPG heartbeats with corresponding ABP signals were extracted from this data set. The experimental results show that this method provided in this article can predict blood pressure effectively.