Cuffless blood-pressure estimation method using a heart-rate variability-derived parameter

Abstract

Objective: Cuffless blood-pressure (BP) estimation has attracted widespread interest owing to its potential for long-term, non-invasive BP monitoring. But it is still impractical in clinical settings, mainly owing to ongoing challenges with respect to accuracy in hypertensive patients. To better estimate the BP, the current study proposes a new cuffless estimation method that includes a sympathetic tone, which has been reported with a varied pattern in hypertensive patients. Approach: First, the heart-rate variability of all subjects is investigated, and a new parameter, the heart-rate power spectrum ratio (HPSR), is proposed to indicate BP dynamics under sympathetic regulation. Then, a new BP estimation model is constructed by integrating HPSP with the pulse transit time (PTT) and photoplethysmography intensity ratio. The estimation accuracy is further evaluated by making comparisons with the standard sphygmomanometer BP on 60 subjects (29 hypertensive and 31 normotensive). Main results: A significant increase in HPSR was observed in hypertensive patients compared to normotensive subjects. Of the 60 subjects, the estimation accuracy was 0.73 ± 10.04 mmHg for systolic BP (SBP) and 0.90 ± 7.10 mmHg for diastolic BP (DBP) in hypertensive patients, which is comparable to 0.54 ± 7.52 mmHg for SBP and 0.82 ± 6.20 mmHg for DBP in normotensive subjects. Furthermore, the proposed method overperformed the traditional PTT-based algorithm by reducing the 3 mmHg error in the standard deviation in hypertensive patients. Significance: The results of the current study demonstrate that the inclusion of factors associated with autonomic activities in the BP estimation model would improve the BP estimation accuracy, especially for hypertensive subjects.