Cuffless Hypertension Risk Assessment and the Significance of Calibration

Early detection of high blood pressure (BP) is of paramount relevance because hypertension is the main risk factor for many cardiovascular diseases. This work evaluates the need of per-subject calibration for discrimination between normotensive (NTS) and hypertensive (HTS) subjects. 668 electrocardiographic (ECG), photoplethysmographic (PPG) and BP recordings from 51 subjects were analyzed. After signal preprocessing and feature selection, 17 discriminatory features were obtained to train machine learning based classifiers. Previous persubject calibration relevance was evaluated by sequential validation, using both close and distant in time calibration measurements varying from less than 1 h to more than 24h with respect to test measurements. The k-nearest neighbors classifier provided an accuracy for new subjects before calibration of 56.79%. The inclusion of just one calibration measurement into the model improved classification accuracy by 30%, reaching gradually more than 97%. Classification accuracy decreased with distance to calibration, but remained well above 83% even days after the last calibration. Thus, discrimination of NTS and HTS subjects can be significantly improved combining PPG and ECG recordings with previous per-subject calibration and, therefore, could be used for the detection of hypertension implementing these techniques in wearable devices.