New photoplethysmogram indicators for improving cuffless and continuous blood pressure estimation accuracy

Although hypertension is the most prevalent treatable vascular risk factor, how it causes end-organ damage and vascular events is poorly understood. Yet, a widespread belief exists that underlying usual blood pressure can alone account for all blood-pressure-related risk of vascular events and for the benefits of antihypertensive drugs, and this notion has come to underpin all major clinical guidelines on diagnosis and treatment of hypertension. Other potentially informative measures, such as variability in clinic blood pressure or maximum blood pressure reached, have been neglected, and effects of antihypertensive drugs on such measures are largely unknown. Clinical guidelines recommend that episodic hypertension is not treated, and the potential risks of residual variability in blood pressure in treated hypertensive patients have been ignored. This Review discusses shortcomings of the usual blood-pressure hypothesis, provides background to accompanying reports on the importance of blood-pressure variability in prediction of risk of vascular events and in accounting for benefits of antihypertensive drugs, and draws attention to clinical implications and directions for future research. Copyright 2010 Elsevier Ltd. All rights reserved.

In this paper we review recent developments in the methodology of non-invasive finger arterial pressure measurement and the information about arterial flow that can be obtained from it. Continuous measurement of finger pressure based on the volume-clamp method was introduced in the early 1980s both for research purposes and for clinical medicine. Finger pressure tracks intra-arterial pressure but the pressure waves may differ systematically both in shape and magnitude. Such bias can, at least partly, be circumvented by reconstruction of brachial pressure from finger pressure by using a general inverse anti-resonance model correcting for the difference in pressure waveforms and an individual forearm cuff calibration. The Modelflow method as implemented in the Finometer computes an aortic flow waveform from peripheral arterial pressure by simulating a non-linear three-element model of the aortic input impedance. The methodology tracks fast changes in stroke volume (SV) during various experimental protocols including postural stress and exercise. If absolute values are required, calibration against a gold standard is needed. Otherwise, Modelflow-measured SV is expressed as change from control with the same precision in tracking. Beat-to-beat information on arterial flow offers important and clinically relevant information on the circulation beyond what can be detected by arterial pressure.

This work presents a system for a simultaneous non-invasive estimate of the blood glucose level (BGL) and the systolic (SBP) and diastolic (DBP) blood pressure, using a photoplethysmograph (PPG) and machine learning techniques. The method is independent of the person whose values are being measured and does not need calibration over time or subjects. The architecture of the system consists of a photoplethysmograph sensor, an activity detection module, a signal processing module that extracts features from the PPG waveform, and a machine learning algorithm that estimates the SBP, DBP and BGL values. The idea that underlies the system is that there is functional relationship between the shape of the PPG waveform and the blood pressure and glucose levels. As described in this paper we tested this method on 410 individuals without performing any personalized calibration. The results were computed after cross validation. The machine learning techniques tested were: ridge linear regression, a multilayer perceptron neural network, support vector machines and random forests. The best results were obtained with the random forest technique. In the case of blood pressure, the resulting coefficients of determination for reference vs. prediction were R(SBP)(2)=0.91, R(DBP)(2)=0.89, and R(BGL)(2)=0.90. For the glucose estimation, distribution of the points on a Clarke error grid placed 87.7% of points in zone A, 10.3% in zone B, and 1.9% in zone D. Blood pressure values complied with the grade B protocol of the British Hypertension society. An effective system for estimate of blood glucose and blood pressure from a photoplethysmograph is presented. The main advantage of the system is that for clinical use it complies with the grade B protocol of the British Hypertension society for the blood pressure and only in 1.9% of the cases did not detect hypoglycemia or hyperglycemia. Copyright © 2011 Elsevier B.V. All rights reserved.