The hypertension assessment based on features extraction using PPG signal and its derivatives.

The long-term abnormal blood pressure would lead to various cardiovascular diseases, therefore it is significant to assess the blood pressure status for prevention. In this study, the proposed feature-extraction based approach is performed on the open clinical trial dataset. Firstly, CEEMDAN algorithm and wavelet threshold analysis are applied to eliminate the noise interference from original PPG signal compared to other signal filters. Considering the strong connection between the hypertension and diabetes disease, ANOVA test with a 95% confidence interval is firstly carried out to select these leading extracted morphological features, which are uniquely related to hypertension, from PPG signal and its derivatives. Subsequently a variety of classification models are used to evaluate and compare in different blood pressure (BP) levels trial. The testing results is demonstrated that Support Vector Machine(SVM) classification model achieves a greater performance compared to other explored models in this paper with the accuracy of 78%, 87% and 88% for cases including normal versus prehypertension subjects, normotension versus hypertension subjects and not hypertension versus hypertension subjects, which further illustrates the proposed method holds the great potential in hypertension assessment.