Blood Pressure Measurement at Home Á. Jobbágy 1 , P. Csordás 1 and A. Mersich 1 1 Budapest University of Technology and Economics/Dept. Measurement and Information Systems, Affiliation, Budapest, Hungary Abstract— Blood pressure varies during the day and also has beat-to-beat variations. Presently available meters determine momentary pressure values which are often misleading. Further drawback of the most frequently used oscillometric method is that it determines arterial mean pressure; the systolic and diastolic values are only derived from it. Though it would be necessary, it is impossible to use personalised constants for the calculation. Thus there are substantial deviations from reference values measured by trained medical operator manually. Measuring blood pressure at home is advantageous because the well-known white coat effect does not influence the result and measurements can be done always at the same phase of daily activity. Presently available automatic devices do not provide reliable results. A new method has been developed at the Department of Measurement and Information Systems, Budapest University of Technology and Economics. The method directly measures systolic pressure and calculates its short-term average. This eliminates the effect of breathing that can cause even 10 mmHg changes in momentary value. Diastolic pressure is also measured directly. The stress level of the tested person is determined and if it is found to be too high then the measurement is not started because its result would be misleading. Our method requires extra information: the Einthoven I lead ECG and photolethysmographic (PPG) signal from the index finger are used. The time delay between the ECG and PPG signal (ΔT EP ) depends partly on the systolic pressure. Short-term averaging is based on ΔT EP monitored for 30 seconds before inflating the cuff. The new method was tested by taking parallel measurements on healthy subjects with a COLIN CBM 7000 tonometer and good correlation was found. The technique is applied in a home-health monitoring device. Keywords— blood pressure, home health monitoring, PPG, pulse wave velocity. I. INTRODUCTION Conventional cuff-based blood pressure measurement does not give accurate and reproducible results [1], [2]. Momentary value is measured and displayed that would be correct if systolic and diastolic pressures were constant. On the contrary, there are short-term variations in blood pressure [3] resulting in a stochastic error. Even if the momentary blood pressure is accurately measured, this is not necessarily typical for the tested person. Blood pressure changes during the day. Knowing blood pressure changes during the day is valuable diagnostic information. 24-hour monitoring might be applied to assess the blood pressure profile. Even 24-hour monitoring does not give information on short-term variation in blood pressure which is not necessarily negligible. Beat-to-beat change in systolic pressure can reach 10 mmHg. Figure 1 shows a 5-s part of a recording taken from a young healthy subject at rest. Blood pressure was monitored continuously by a COLIN CBM 7000 tonometer. This device uses a pressure sensor fixed to the artery at the wrist. Pressure value is calibrated with the help of an upper arm cuff by determining the mean arterial pressure indicated by the maximal pressure oscillations. For 5 … 15 minutes following the calibration the accuracy of the tonometer is satisfactory, thus short-term variation can be assessed. Figure 1. Systolic and diastolic pressure substantially changes even within a short time. Recording was taken with a COLIN radial artery tonometer from a young healthy male subject. Both physical and psychical stress influences blood pressure. Its ranges considered as normal are valid only for relaxed persons. Occlusion of the arteries by a cuff changes the parameter to be measured. The most commonly used oscillometric method [4] measures the arterial mean pressure and only calculates the diastolic and systolic values. Averaging over a given time is proposed. Average value and standard deviation characterize blood pressure better than a single momentary value. Compared to the oscillometric measurement, the suggested method makes use of the extra information gained from an Einthoven I (or II) lead ECG and two PPG signals taken from the left and right index finger. Similar methods are reported in [5], [6], [7].