Patent Application
20230034358
The invention disclosed herein relates to improvements in the collection of personal health data. It further relates to a Personal Health Monitor (PHM), which may be a Personal Hand Held Monitor (PHHM), which incorporates a Signal Acquisition Device (SAD) and a processor with its attendant screen and other peripherals. The SAD is adapted to acquire signals which can be used to derive one or more measurements of parameters related to the health of a user. The computing and other facilities of the PHM with which the SAD is connected or integrated are adapted to control and analyse signals received from the SAD. The personal health data collected by the SAD may include data related to one or more of blood pressure, pulse rate, blood oxygen level (SpO2), body temperature, respiration rate, ECG, cardiac output, heart function timing, arterial stiffness, tissue stiffness, hydration, the concentration of constituents of the blood such as glucose or alcohol, blood viscosity, blood pressure variability and the identity of the user.
Many methods for measuring blood pressure are known, such as the oscillometric method (using measurements of pressure in a cuff), the PPG optical method (using a measurement of the absorption of light by blood in an artery), the auscultation method (using change in sound as blood flows through an artery) or a direct method (using ultrasound imaging or any other way of detecting the difference in luminal area or size of the artery as it changes from occluded to patent)
WO2013/001265 (PCT1) discloses a Personal Hand-Held Monitor (PHHM) in which a signal acquisition device (SAD) is integrated with a Personal Hand Held Computing Device (PHHCD), such as a cell phone, and is adapted for the measurement of, for example, blood pressure or one or more of several other health-related parameters. The SAD is adapted to be pressed against a body part or to have a body part pressed against it, for example, where the body part is the side of a finger. This permits a cuff-less occlusion measurement. The SAD also includes an electrical sensor which can be used to detect a 1-lead ECG between the two hands.
WO2014/125431 (PCT2) discloses several improvements of the invention described in PCT1, including the use of: a gel to measure pressure; a saddle-shape surface to interact with a body part; corrections for the actual position of an artery relative to the device; and the use of interactive instructions to the user.
WO/2014/125355 (PCT3) discloses improvements of the non-invasive blood analysis disclosed in PCT1 including improvements to the specificity and accuracy of the measurements.
WO2016/096919 (PCT4) discloses several further improvements of the inventions described in PCT1 and PCT2, including improvements to the gel and pressure sensing means, the use of mathematical procedures for extracting blood pressures and other signal processing inventions, a means for identifying the user, improvements to the electrical systems for measurement and several aspects of test and calibration of the device. PCT 4, page 11, lines 5 to 8 discloses that, in order to carry out a PWA, it is possible to use a camera to detect the arrival of the pulse by means of the change of colour of the skin. This allows, for example, the difference in pulse time at the face and finger to be found.
WO2017/140748 (PCT5) discloses further improvements to extracting blood pressure and several other health-related parameters that can be derived from the measured data.
WO2017/198981 (PCT6) discloses improvements to the inventions disclosed in PCT3 whereby the device can be built using small and inexpensive components.
WO2019/211807 (PCT7) discloses several improvements to the inventions disclosed above, including aspects where the device is adapted to use the fingertip or an artery in the cheek, both of which result in increased variation in applied pressure when compared to the earlier applications.
PCT1 to PCT7 are all in the name of Leman Micro Devices SA and are therefore collectively referred to as “the Leman applications”. The Leman applications are hereby incorporated into the present application in their entirety by reference.
The inventions disclosed by the Leman applications are effective, reasonably accurate and easy to use, and may be integrated into a cell phone. Cell phones are manufactured in quantities of hundreds of millions and the price of their components is critical. These inventions allow the cost of blood pressure measurement to be reduced to a level that is acceptable for a cell phone, in part because they eliminate the expensive and heavy components of conventional devices that ensure a constant applied pressure.
The area of an artery changes between the times of diastole and systole because there is a change of the difference between pressure of the blood inside the artery and the pressure in the tissue surrounding the artery. This causes the artery wall to stretch. If this stretching can be detected and measured, it is possible to find the diastolic and systolic pressures without using a cuff.
This measurement principle of cuff-less occlusion thus has many advantages to the user and to global health but it requires the user actively to press the device against a body part or the body part against the device in a controlled manner. The present invention reduces the disadvantages of this by allowing much shorter measurement times and by compensating for the fluctuations of pressure that arise from the action of pressing. It also discloses additional measurement capabilities using the same collection of sensing devices.
The present invention concerns several improvements to the measurements by creating novel features for the inventions described in the Leman applications.
There are three major classes of device for automatic non-invasive measurement of blood pressure: occlusion devices; pulse wave velocity (PWV) devices; and pulse wave analysis (PWA) devices.
Occlusion devices find the pressure that must be applied outside an artery to balance the pressure of the blood inside the artery. These give an absolute measurement of blood pressure that requires no personal calibration. Oscillometric automatic cuff devices use this occlusion principle. The Leman applications disclose a unique cuff-less occlusion device which is exploited in the first aspect of the current invention.
Pulse wave velocity (PWV) devices measure a property related to the rate at which the pressure wave propagates along the artery which, in turn, is related to the stiffness of the artery and the difference between the pressure of the blood inside the artery and the tissue outside the artery. These give a relative measure of blood pressure and so can be used to detect the change in blood pressure after they have been calibrated for the user. PWV devices are well-known and work by estimating the speed at which the pressure wave propagates along the artery by dividing the distance between two points in the arterial system by the time taken for the pulse to travel between the two points (the Pulse Transit Time PTT). There are two ways of doing this:
Pulse Wave Analysis (PWA) devices analyse the shape of the pulse wave measured at a remote point. This is usually measured by an optical sensor but may be measured using an ultrasound sensor, a sensor which detects displacement or a pressure sensor. The PWA is related to the pulse wave velocity. PWA devices analyse the waveform of a signal related to the area of the artery to infer blood pressure. They may do this by estimating the PWV explicitly (see, for example, Tavallali et al., Scientific Reports 8, Article number: 1014, 2018) or by direct analysis in which the contribution of velocity is implicit (see, for example, Gircys et al., Appl. Sci., 2019, 9, 2236; doi:10.3390/app9112236). PWA devices can be used to detect a change in blood pressure after they have been calibrated for the user.
A further feature of PWA and PWV devices is that they may be adapted to make an instantaneous estimate of the blood pressure on each heartbeat. This is useful both to provide biofeedback to the user and to allow the user to derive an estimate of short term blood pressure variability.
A first aspect of the present invention relates to novel ways of combining the strengths of occlusion devices, pulse wave velocity (PWV) devices and pulse wave analysis (PWA) devices to create a device that is more accurate and/or easier and quicker to use and/or offers additional measurement capabilities.
The Leman applications disclose devices that are capable of making occlusion, PWV and PWA measurements. The Leman applications disclose that two or more measurements may be combined to achieve greater accuracy. The table on page 16 of PCT1 states that:
The first aspect of the present invention goes beyond combining two values found independently and uses cooperation to make a measurement of blood pressure or another health-related parameter that is more accurate or that is easier to conduct or that is made more quickly than could be made by according to the disclosure in PCT1.
According to this aspect of the present invention, it has been discovered that it is possible to use occlusion devices, PWV devices and PWA devices in various combinations to reduce errors in the measurements of blood pressure and other health-related parameters that may result from their use and to make it easier and quicker to obtain such measurements.
Various kits and devices of the first aspect of the present invention are disclosed in independent claims 1, 2, 4, 5, 7, 8, 10 and 11 set forth below.
Preferred kits and devices of this aspect of the invention are set out in dependent claims 3, 6, 9 and 12 to 27 set forth below.
In respect of the claims to a kit comprising two devices and an analysing means, the analysing means may be an entirely separate item, present entirely as part of one, the other or both of the devices or present partly as part of one or both of the devices and partly in a separate item.
In any kit of the first aspect of the invention, the analysing means may receive signals from the other components in the kit or integrated device by any suitable means, such as via cables, Wifi, Bluetooth or any other suitable means, as will be well known to the person skilled in the art.
In particular, it has been found that devices of the type disclosed in the Leman applications can be adapted to take the necessary measurements for occlusion, PWV and PWA measurement and combine two or more of the measurements cooperatively to improve the accuracy of the measurement or to improve the speed of processing in the devices disclosed in the Leman applications.
It is preferred that the part of the body which contacts the contacting means is a finger, more preferably the end of a finger, but any part of the body with an accessible artery (e.g. face, neck, toe, wrist etc) may be used. Similarly, PWV and PWA measurements may be made between any two separated parts of the body, such as measuring the start time at the heart using an electrical sensor and the finish time at a finger using an optical sensor or measuring the start time at the face using a camera viewing the face and the finish time at a finger using an optical sensor.
Preferably, the contacting means does not include a cuff.
Preferably, the kit or device is adapted to provide instructions to the user to press harder or softer create a range of applied pressure.
The change in area of the artery may be found using an optical sensor to measure the absorption of light by the blood in the artery, in a manner analogous to a pulse oximeter.
In the devices disclosed by the Leman applications, the pressure that is used to occlude the artery is created by the muscular action of the user. Such action inevitably has fluctuations. Fluctuations between heart beats are not critical, because the algorithms of the LMD devices are able to accept data in any order, but fluctuations within a heartbeat can cause significant errors.
Conventional blood pressure measurement devices assume that the pressure in the tissue surrounding the artery is constant for the duration of a heartbeat, or at least for the period between diastole and systole. They also assume that the pressure in the tissue surrounding the artery is the same as the applied pressure. It is then simple to estimate blood pressure by plotting the value of the property related to the change in area of an artery against the applied pressure. Various proprietary algorithms are used to find diastolic and systolic blood pressures from this curve.
This approach is less effective, and therefore the measured blood pressures is less accurate, if the assumption that the pressure in the tissue surrounding the artery is constant is not valid. This may be because the change of arterial area during a heartbeat causes a significant change in the pressure in the tissue surrounding the artery. This will occur with any device but, in practice, is negligible for a device such as a conventional brachial cuff. However, it becomes more significant if the artery represents a significant fraction of the volume of the tissue being measured, such as if the body part being used for the measurement is the side of a finger. It may also occur if the mechanism for applying pressure to the tissue surrounding the artery does not apply a constant pressure, for example if it is effected by a person pressing the device against the body part or pressing the body part against the device; the person's muscles may shake with a time constant short compared to a heartbeat.
This limitation restricts the use of ways of measuring blood pressure that may be less expensive, more accurate, easier to use or smaller than the known devices, such as those ways disclosed in the Leman applications. In these, the pressure that is used to occlude the artery is created by the muscular action of the user. Such action inevitably has fluctuations. Fluctuations between heart beats are not critical, because the algorithms in the devices disclosed in the Leman applications are able to accept data in any order, but fluctuations within a heartbeat can cause significant errors.
The second aspect of the present invention overcomes or greatly mitigates the limitation due to changing pressure and so brings benefits to a range of ways of measuring blood pressure, such as those disclosed in the Leman applications.
A second aspect of the invention relates to a way of reducing the errors due to variations of the pressure between the device and the body part using isobaric analysis.
The second aspect of the present invention relates to a device for measuring blood pressure which measures a property related to the change in area of an artery as a function of the pressure applied to that artery and which can accommodate a change in applied pressure during the course of a single heartbeat.
The property related to the change in area may be the change in measured cuff pressure (the oscillometric method), the change in absorption of light (the optical method), a change in sound (the auscultation method) or the change in the luminal area or size of an artery as it changes from occluded to patent (a direct method).
Therefore, the second aspect of the present invention provides a device for the non-invasive measurement of blood pressure comprising a means for measuring the change in luminal arterial area during a heartbeat, a means for applying pressure to the body part containing the artery and a means for measuring the instantaneous pressure applied to the body part containing the artery, wherein:
Preferably, the means for measuring the change in luminal arterial area during a heartbeat is oscillometric.
Alternatively, the means for measuring the change in luminal arterial area during a heartbeat is optical.
Further alternatively, the means for measuring the change in luminal arterial area during a heartbeat is auscultation.
Yet further alternatively, the means for measuring the change in luminal arterial area during a heartbeat is ultrasound.
Preferably, the means for compensating for the change of the instantaneous pressure applied to the body part containing the artery during a heartbeat uses a separate analysis of the values found by the means for measuring the change in luminal arterial area during a heartbeat at diastole and at systole.
Preferably, the separate analysis uses a curve-fitting algorithm to create two parametric curves, representative of the values at diastole and at systole. The curve fitting algorithm may be a Loess algorithm.
Preferably, the difference between the two parametric curves is used to create a set of pseudo heartbeats that have the same instantaneous pressure applied to the body part containing the artery at diastole and systole and wherein the set of pseudo heartbeats may then be analysed in the same way as true beats would be analysed if the instantaneous pressure applied to the body part containing the artery were not to change significantly during a heartbeat.
Preferably, the blood pressure is found by analysing the changes in arterial area as a function of the instantaneous pressure applied to the body part containing the artery.
The first and second aspects of the present invention can act synergistically in that the improvement in accuracy of the second aspect allows greater reliance on the occlusion measurement for the first aspect. It is also apparent that, although the two aspects were invented in the context of the Leman applications, their utility extends more widely to other forms of blood pressure measurement device.
The devices disclosed by the Leman applications have several sensors that create a rich set of data. The sensors may be used cooperatively to improve the accuracy, ease of use or functionality of the device.
A third aspect of the invention uses features of the data that is or may be collected by the devices disclosed by the Leman applications to improve or extend the scope of the personal health data that is collected.
Two specific opportunities are:
PPG signals are strongly influenced by both the changes in luminal area of the artery during the heartbeat and the absorption of light by the tissue surrounding the artery, including the blood in the local blood vessels (micro-arteries and veins). When the pressure between the body part and the SAD changes, the tissue deforms and the blood flows in or out of the local blood vessels. The time constant for this deformation and blood flow is typically a few seconds.
The value of this time constant is dependent in part on the viscosity of the blood. The magnitude of the change of the PPG signal depends on the shape and constitution of the region of the body part that is illuminated and on the wavelength of the PPG light. The wavelength determines the relative absorption due to oxygenated blood, deoxygenated blood and tissue.
The third aspect of the present invention exploits both the high frequency PPG signal, resulting from the fluctuations due to variations of the luminal area of the arteries, and the low frequency PPG signal, obtained by filtering the high frequency signal. It also exploits the ability of the LMD devices to instruct the user to create a controlled applied pressure between the device and the body part and to vary that pressure on demand.
The relationship between the high and low frequency signals obtained by controlled applied pressures and the viscosity of the blood must be determined empirically. This may be done using supervised machine learning, whereby the training data set comprises:
The PPG optical system may also be used as a proximity detector. In devices according to the Leman applications, the LEDs that emit the light and the photodetector that detects the light are in general approximately 6 mm apart. If a reflecting or diffusive surface is moved towards the device when the LED is on, the received signal will peak at around that distance. At the same time, the ambient signal due to background light will fall as the device approaches the surface due to shadowing.
Features of the third aspect of the present invention are disclosed in independent claims 38 and 50 and preferred features of this aspect are disclosed in dependent claims 39 to 49 and 51 to 54, respectively.
Set out below, by way of example only, are examples of the aspects of the present invention. It will be understood that the present invention is not limited to these examples. The scope of the invention is set forth in the claims below.
In the examples, reference is made to the accompanying drawings, which are provided by way of illustration only and which do not limit the scope of the invention, in which:
A limitation of all PWV and PWA techniques is that they must be calibrated for each user. This requires the user to make several measurements of blood pressure using an occlusion (cuff) device at the same time as measurements of PWV or PWA. These allow the PWV or PWA to be calibrated and then used to detect changes in blood pressure from that measured with the cuff. The calibration typically remains valid for a period of time of days to weeks, after which it must be repeated. This limits the utility of the PWV or PWA technique since it also needs access to a cuff device.
The absolute measurement of blood pressure by cuff-less occlusion, using a separate device as part of a kit or as part of an integrated device, may be used as a calibration for the PWV or PWA measurement. The PWV or PWA measurement may then be used to make quick and easy measurements of blood pressure until it is necessary to recalibrate.
The calibration procedure may be further enhanced by making several calibrations under different conditions, such as for example at different times of day or before and after taking exercise. Such distributed calibrations may be exploited to improve the accuracy of the subsequent PWV or PWA measurements or to extend the period until it is necessary to recalibrate.
PWA is simple and easy to use but it is not easy to achieve adequate accuracy. One of the reasons is that the measured optical waveform depends on how hard the user presses the measurement device against the body part. The pressure sensor in the pressure means can be used with its optical sensor to generate PWA waveforms at a specific pressure by providing feedback to the user to press harder or softer. This can be used to ensure that the measurement pressure is the same as that used for calibration or can be used to provide the PWA analysis with a set of waveforms captured at different pressures.
Alternatively, the actual measured applied pressure may be used, without providing feedback to the user, as an input to the PWA algorithms so as to improve their accuracy and/or extend the time before needing re-calibration.
The PTT found using an electrical signal includes the PEP and so the estimated PWV will not be correct. PEP is fairly stable for an individual and so an occasional measurement of this can be used to correct the measured PTT.
PCT 4,
PWV is related to blood pressure via the arterial stiffness. If this stiffness is known, it is possible to make a more accurate estimate of that relationship and hence a more accurate estimate of blood pressure derived from PWV. Since the waveform analysed by PWA also depends on PWV, the stiffness can also be used to improve the accuracy of PWA.
PCT 4, page 11, lines 9 to 14 discloses that the local arterial stiffness can be measured directly by the Leman devices.
The effective stiffness of an artery also depends on the stiffness of tissue surrounding it. Aspect 5 of PCT 5 discloses that devices according to the Leman applications can make an estimate of the stiffness of that tissue, including its change due to hydration. This can also be used to improve the accuracy of PWV and PWA measurements, in a similar way to the fourth example above.
PCT 2, lines 24 to 30 discloses that an estimate of the arterial stiffness may be used by the cuff-less occlusion device to improve some of the techniques for extracting blood pressure from the occlusion data. It assumes that the estimate is found directly from the measured data but there is advantage in using an independent estimate, derived from a PWV or PWA measurement. This can contribute to both the accuracy of the result and the speed of processing.
The LMD applications disclose several techniques for extracting blood pressure from the data derived from the sensors that use a search or optimisation algorithm. These operate by searching the solution space, including searching for the diastolic and systolic blood pressures. An estimate of these values that is derived from PWV or PWA can be used to narrow the search space or at least to indicate starting values for the search. This reduces the time taken for the search and reduces the risk that the search selects a sub-optimum value of the solution.
Referring to
Note that the exact form and vertical scale of
It is clear that the value of deltaA depends on the pressure of the tissue surrounding the artery.
By plotting deltaA in
This is a mature technique for use in devices that measure blood pressure where the pressure in the surrounding tissue does not change significantly within one beat.
This aspect of the present invention does not use deltaA directly. Instead, it uses the following sequence of steps:
This set of pseudo heartbeats may then be analysed using any of the analytical methods that would otherwise be used for actual heartbeats, but with a known instantaneous applied pressure.
The smoothed curve may be found using curve-fitting techniques that are well-known by persons skilled in the art, such as the Loess algorithm. As well as providing a parametric model, the parameters of the curve-fitting technique may be selected to smooth the data and so reduce the effect of measurement noise.
If the instantaneous pressure of the tissue surrounding the artery lies between the diastolic blood pressure and the systolic blood pressure, the luminal area of the artery will increase rapidly as the instantaneous arterial pressure exceeds the instantaneous pressure of the tissue surrounding the artery. It will also fall rapidly as the instantaneous arterial pressure falls below the instantaneous pressure of the tissue surrounding the artery. The timing of these two events during the heartbeat may also be used to estimate the diastolic and systolic blood pressure, in a manner analogous to the use of deltaA. For example, in an ideal model with no noise, the interval between those two times is zero if the instantaneous pressure of the tissue surrounding the artery equals or exceeds the systolic blood pressure. Similarly, that interval equals the duration of the heartbeat TH if the instantaneous pressure of the tissue surrounding the artery equals or is less than the diastolic blood pressure.
Some techniques for finding diastolic and systolic blood pressure exploit that interval. They can be compensated for the effect of varying applied pressure by the same technique as is used for deltaA, where the equivalent steps are:
This set of pseudo heartbeats may then be analysed using any of the analytical methods that would otherwise be used for actual heartbeats, but with a known instantaneous applied pressure. It will be apparent to a person skilled in the art that other combinations of TF, TR and TH may be used, such as (TF−TR)/TH.
a term related to the integral of pressure;
a term related to the Systolic blood pressure, above which the arteries are occluded;
a linear variation in sensitivity due to deformation of the tissue; and
small terms related to the instantaneous pressure and rate of change of pressure.
Similar results may be obtained for other colours of PPG light, including but not restricted to red and infra-red, and also for the high frequency PPG signals.
The model parameters are used as inputs to the machine learning to find the combination of parameters that best predicts the viscosity of the blood.
It should be clearly understood that, for all of the aspects of the present invention, the examples and figures and the description thereof are provided purely by way of illustration and that the scope of the invention is not limited to this description of specific embodiments; the scope of the invention is set out in the attached claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 1918576.8 | Dec 2019 | GB | national |
| 1918579.2 | Dec 2019 | GB | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/IB2020/061894 | 12/14/2020 | WO |
