GB2516220A

GB2516220A - Heart and Blood Pressure Measurement Device

Info

Publication number: GB2516220A
Application number: GB1310785.9A
Authority: GB
Inventors: Dennis Majoe
Original assignee: Individual
Current assignee: Individual
Priority date: 2013-06-17
Filing date: 2013-06-17
Publication date: 2015-01-21
Also published as: GB201310785D0

Abstract

A ring to be worn on the finger, in which computing device and electronics 10 detect the electrical potential between contacts 1 and 3 in order to measure electro cardio activity of the heart when the user touches 3 with the non-ring wearing hand and where infra-red emitting LED 5 emits light via the finger flesh 6 to measure absorption of light via detector 7, the absorption being used to measure heart activity. The computing device calculates the heart rate and heart rate variability using the data from both electro cardio measures and flow rate measures by combining both sets of data available as well as using time dependent differences in measured heart activity to derive an estimate of systolic blood pressure that is further adjusted by environmental and personal factors measured using other devices such as an accelerometer.

Description

Heart and Blood Pressure Measurement Device This invention relates to a wearable ring used to measure and record the heart related activity of a human.

In many health related situations it is necessary to monitor a person's heart activity, blood pressure and physical activity. However in order to be easily worn and used a small device such as a finger ring would be more attractive to users.

Although many devices are available for measuring such physiological factors, in this invention the ergonomic ring embodiment is combined with data fusion of sensed data to provide more accurate measurements.

The present invention describes an electronic measurement device that is preferably worn as a ring around the middle or ring finger and estimates the heart activity, blood pressure, temperature, activity and stress levels of a human.

The ring device is small and light weight and non-obtrusive compared to other measurement devices.

The measurement of the user's electrocardiographic heart beat signal is ergonomically performed requiring the user to only touch the ring with a finger of the hand not wearing the ring.

A second heart measurement is achieved via measuring blood oxygen saturation. This is done without the user required to do anything.

The electrocardiographic heart beat signal measurement is mathematically fused with the heart measurement achieved via measuring blood oxygen saturation using a Central Limit Theorem filter approach which includes estimates for the current noise in the two signals therefore providing a fused and more accurate heart beat signal measurement.

The activity of the user is measured using an accelerometer and this activity measurement is used to provide both a noise estimate for the data fusion as well as a physical activity reference guide.

The ring will also measure the skin conductance of the user. This measure provides data for the heart signal noise estimation as well as a measure of user stress levels.

The ring will measure the time delay between the peak of the electrocardiographic heart beat signal and the peak of the oxygenated blood flow arriving at the ring. This delay provides an estimate of the user's blood pressure. This avoids the use of any uncomfortable and constrictive methods of blood pressure measurement.

The accelerometer sensor information is used to establish the position of the user's arms and this data is used to adjust the blood pressure estimate.

The activity of the user, estimated by sensing accelerometer data over time, provides further information used to adjust the blood pressure estimate, which rises for example during an activity such as running.

The stress level of the user sensed by skin conductance is also used to adjust the blood pressure estimate.

The skin temperature of the user is measured by the ring. This measurement provides an estimate for the general body temperature which relates to blood pressure. In addition it is necessary to measure this value to provide an estimate of capillary dilation which is a factor in the blood flow around the finger.

Thus Central Limit Theorem filtered heart rate and estimates for arm position, stress levels, activity levels, capillary dilation as well as time delay between the peak of the electrocardiographic heart beat signal and the peak of the oxygenated blood flow arriving at the ring are fused to derive a filtered estimate of blood pressure measurement. This measurement will be more robust than using a simple non adjusted method.

An example of the invention will now be described by referring to the accompanying drawings: * figure 1 shows a ring in cross section, showing the main active components * figure 2 shows the primary software flow and functions applied in the ring.

The ring comprises three conductive elements 1, 2 and 3 in figure 1.

Element 1 is a grounding contact to the palm side of the finger. The skin conductance is measured between elements 1 and 2. This is achieved by the application of a test volt voltage across the elements 1 and 2 which are connected to one side of a Wheatstone bridge circuit. Normal skin conductance lies between 100K and 400K Ohms. The bridge circuit therefore comprises three resistors of this approximate value and the conductance of the skin acting between contacts shown at 1 and 2. This results in a roughly balanced bridge and the skin conductance is measured by amplifying and digitizing the differential voltage across the balanced bridge.

In order to measure heart activity, the user must touch the ring element 3 using the other hand, which is not wearing the ring. In this way the ring can measure the electrical activity occurring between one hand and the other hand, via the chest. The electrophysiological voltage that exists across the chest due to the heart muscles during the heart beat is then measured between element 3 and element 1. The voltage is first amplified, offset dynamically to avoid saturation and then digitized using an analogue to digital converter.

Note that element 3 is insulated from the other elements via a padding 4.

The computing unit 10 measures the heart voltage by sampling the digitized values at over 300 samples per second. Software determines the peak of the heart beat signal and uses this as a fiducial point to mark the timing occurrence of the heartbeat. By timing the occurrence of the fiducial point, statistics may be calculated by the ring's computing device to obtain measures for the average heart rate and the heart rate variability.

The infra-red LED emitterS is positioned to illuminate the area of the finger 6 which contains blood vessels. Infra-red LED receiver] detects the scatter of infra-red light passing from 5 through 6 and arriving at 7. The light received at 7 depends on the blood flow in 6 which peaks when the pressure wave of blood arrives from the pumping heart. The arriving blood contains water, hemoglobin and oxyhemoglobin. These three components of blood combine to absorb infra-red light between 250 and l000nM wavelength of light. Therefore the amount of infra-red light output by the emitterS and received at 6 varies with the volume of blood in that zone and as a result when the peak flow of blood in area 6 occurs then the least amount of infra-red light reaches 7. Note that this blood flow is restricted more and more as the finger temperature falls due to the shrinking of capillaries to keep blood flow more central to the body.

The output of LED 7 is fed to a resistive network. The current flowing due to the emitted light is digitized by analogue to digital conversion of the resistor network voltage after amplification.

The software in the ring's computing unit may time the occurrence of the peak flow.

Therefore it may estimate the heart rate and heart rate variability based purely on blood oxygenation.

Therefore the computing unit may have two independently derived heart rate measures.

However each one is noisy due to external effects. Noise measures may be estimated for the electrocardiographic heart beat signal measurement based on the high frequency physical vibration and movement of the ring which degrades electrical contact and introduces electrical noise and this can be estimated from the standard deviation of accelerometer vibration readings. In addition if the skin conductance measure falls significantly then it is likely the surface resistance of the skin is high and this results in high levels of external electrical noise contaminating the heart electrical signal. Finally when the signal is processed the ratio of power spectral density in the 50 Hz region of the signal may be used as a further noise estimate. Therefore a noise measure may be derived from these three factors.

The heart measurement achieved via measuring blood oxygen saturation will also have an associated noise component. This noise will be dependent on the physical action of waving of the hand wearing the ring and changes in blood capillary dilation. Estimates for this noise may be derived from the accelerometer's sensed low frequency movement and the finger temperature. If the hand is being waved near the heart rate then this will confuse the heart rate signal. lithe fingers suddenly become very cold or hot then transients will be seen in the blood flow measures. The power in the low frequency spectrum of physical motion and the recorded ring temperature variation are the two factors combined to form a single noise measure.

Using a Central Limit Theorem filter approach which includes estimates for the current noise in the two signals and the two signals, a filtered heart rate and heart rate variability measure may be computed in real time to derive a more precise estimate in the presence of noise.

Pulse transit time is the time taken between the heart beating and the pressure wave of blood arriving at the finger and it has been widely published that this is linearly correlated with blood systolic pressure.

To measure Pulse transit time, the ring's software measures the time difference between the peaks of the electrophysiological voltage that exists across the chest due to the heart beating and the peaks in blood flow detected by the infrared light absorption. The time delay between these two events is related to the time it takes for blood to exit the heart and travel as a pressure wave down to the finger and may be used as the Pulse transit time value. Due to the elastic nature of the blood vessels, this time is inversely proportional to the blood pressure and also is related to the angle of the arm relative to the vertical. If the finger is raised well above the heart, the speed of blood flow is slower compared to if the finger is well below the level of the heart. For stable measurements the user is required to sit or stand with the arms at the same angle relative to vertical. The higher the blood pressure, the smaller is the time difference between the actual heart beat and the peak flow detection.

The ring preferably includes a three axis accelerometer to measure the activity of the user by sampling the accelerometer each half a second and recording the current acceleration values.

The accelerometerS is also used to estimate the angle of the arm. The accelerometer measures earth's gravity field acceleration in three axes and resolved forces provides an indication of the orientation of the finger and therefore an estimate of arm angle relative to vertical. This is used to identify which one of several angular positions relative to vertical the arm is in and correspond to calibration scaling factors accounted for when calibration of the ring with a reference blood pressure device was performed.

The temperature of the person wearing the ring may be measured using a positive temperature coefficient resistor 9 placed near the finger surface. The resistor may be used as part of a bridge resistor network whose output voltage is digitized using an ADC. The software may calibrate the measured values against a reference temperature.

The pulse transit time method for measuring systolic blood pressure has several sources of possible error. A better estimate for blood pressure may be achieved if the PU method is fused with a general model of blood pressure for a given user. In this invention the primary contributing factors for this model are described.

The arm position relative to vertical which is sensed by the ring accelerometer will provide a means to modify the pressure estimate since if the finger is above the head in a vertical angle then this will result in the longest delay, compared to when the arm is pointing down to the ground. This variation may be modelled using a trigonometric sine law in combination with several reference arm positions recorded during a calibration with a reference blood pressure device.

The emotional stress of a person will affect the blood pressure and in general high blood pressure is associated with high arousal which is associated with higher skin conductance.

Therefore the ring software may check that rising and falling skin conductance and rising and falling pressure measures are correlated and contrary variations in this rule may be used to modify the pressure readings if required.

The activity level of a person is strongly correlated with blood pressure. Major activity of the user measured through the accelerometer should correlate with increased blood pressure.

The model will check that rising and falling activity and rising and falling pressure measures are correlated and contrary variations in this rule may be used to modify the pressure readings if required.

The temperature of the user's hands near the blood flow sensors provides a measure for the dilation of the user's capillaries. If the capillaries are dilated then in general the delay time is smaller and if they are constricted then the delay will be longer. Therefore the temperature reading may be used to modify the pressure reading delays according to calibrated guidelines.

The above factors are combined into a model which generates an offset and scaling factor for the Pulse transit time derived blood pressure measure, such that that measure may be raised or lowered and scaled to match the environmental factors at the time. Arm angle and temperature will largely contribute to the scaling term. Stress and activity will largely contribute to the offset term.

Claims

Claims 1. A physiological measurement device worn on the finger as a ring that incorporates an electrical conductor contacting the finger and a second conductor insulated from the finger, the second conductor being brought in touch with a second part of the body skin, the electrical potential between the two contacts being used to measure the electro-cardio logical muscle activity of the heart and the digitised signal processed and used to estimate the heart rate and heart rate variability.
2. A physiological measurement device worn on the finger as a ring that incorporates an infra-red LED emitter and a wavelength matched infra-red LED receiver, the emitter being placed on one side of the ring, emitting light into the soft flesh of the palm side of the finger from one side, the receiver receiving the emitted light passing through the flesh on the other side, the emitted light being absorbed more when there is a larger amount of blood in the flesh, resulting in a lower amount of light received by the receiver and vice versa the absorption rates being measured by digitising the electrical flow in the receiver diode and the data processed to estimate the heart rate and heart rate variability.
3. A ring as in claim 1 and claim 2 in which the time difference between heart activity as detected by the electro-cardio logical muscle activity of the heart and absorption of emitted infra-red light by the blood flow in the finger is used to provide an estimate of the transit time of the pressure wave from the heart to the finger the time being used to estimate systolic blood pressure computations using blood pressure calibrated against an external blood pressure reference measure.
4. A ring as in claim 1,2 and 3 in which the heart rate and heart rate variability estimates are improved by fusing the results from the electro cardio reading and the blood flow reading using weighted averaging of the two values where the weightings are chosen a-priori or adapted in real time so as to prefer the less noisy signal and where the real time estimates of the noise are obtained from other sensor data measured from the ring wearer
5. A ring as in claim 4 in which the blood pressure estimate is adjusted according to a model of the ring wearer's personal and environmental status where the real time estimates for the status data are obtained from other sensor data measured from the ring wearer.