WO2016035041A1 - Apparatus and method for non-invasive pressure measurement of a fluid confined in a vessel with elastic or rigid walls fitted with an elastic window - Google Patents

Abstract

The present application discloses an apparatus and method for the non-invasive and continuous measurement of the internal pressure of a fluid confined in a vessel with elastic or rigid walls or fitted with an elastic window in the wall. The apparatus is based on a vibrator- accelerometer combination, mounted on a common support, which enables the accelerometer to sense the vibrations produced. The method for pressure measurement relies on the amplitude modulation effect that occurs when the fluid pressure transmitted through the vessel wall or window attenuates the vibrator oscillations, sensed by the accelerometer. This technology further comprise: 1) a method of simultaneously estimating the morphology of the pressure wave and other related parameters such as vessel wall compliance, distension and distension dynamics and 2) method of calibrating the estimated pressure waveform based on the re-establishment of the attenuation to a predefined reference amplitude value whatever the force exerted on the vessel wall may be. Although some other possible applications can be found, its main focus is the non- invasive assessment of arterial blood pressure and arterial pressure waveform, in humans.

Description

DESCRIPTION

"APPARATUS AND METHOD FOR NON-INVASIVE PRESSURE MEASUREMENT OF A FLUID CONFINED IN A VESSEL WITH ELASTIC OR RIGID WALLS FITTED WITH AN ELASTIC WINDOW"

Field of Invention

The present application discloses a non-invasive pressure sensing method and apparatus for continuously measuring the internal pressure of a fluid, confined in a vessel with elastic or rigid walls or fitted with an elastic window.

Background

The pressure of a fluid confined in an elastic vessel can be measured in a number of different ways. The standard method relies on the insertion of a pressure gauge through the vessel wall to provide a direct coupling from the enclosed fluid to the transducer's sensing surface. This method provides a direct, accurate and continuous measurement of fluid pressure; however in some systems, such as the biological ones, it is not always reasonable to apply such an invasive approach. The measurement of arterial blood pressure (ABP) in a living subject is perhaps the most significant case where the use of non^¬ invasive methods has several advantages: ease of use, convenience, speed, low-cost, lower risk to the patient and ability to be applied to large number of subjects.

At the moment, the most well-known non-invasive method of clinical measurement of ABP is based on a Riva-Rocci sphygmomanometer that is composed by an arm-encircling elastic cuff to occlude the blood vessel, a rubber bulb to inflate the cuff and an aneroid manometer to measure the blood pressure. When the measurement of ABP is obtained manually, a stethoscope is required, in order to auscultate the sounds generated, as the cuff is slowly deflated. These (Koroktoff) sounds - a complex series of audible frequencies produced by turbulent flow as the cuff is deflated - are clinically important for measuring systolic and diastolic blood pressures. When the measurement of ABP is obtained by automated non-invasive blood pressure devices (NIBP) , the stethoscope is not needed.

Most automated NIBP are based on the technique called oscillometry that measures the amplitude of oscillations which show up in the inner cuff pressure signal during cuff deflation, using a pressure transducer inside the monitor. Peak amplitude of arterial pulsations corresponds to the mean arterial pressure. Values for systolic and diastolic pressures are derived using specific and often proprietary algorithms that evaluate the rate of change of the pressure pulsations. Although other techniques have been described for automated NIBP measurement, e.g.: employing a microphone in the place of the stethoscope or employing a transcutaneous Doppler sensor to detect the motion of the blood vessel walls in the different occlusion states, none have yet succeeded the standard oscillometric technique. The non-invasive methods previously described, standard auscultatory method and oscillometry, are widely used in the clinical practice. However, they can only provide slow and time discrete measurements of ABP. Both methods typically require 20-45 seconds to obtain a valid measurement and a long period of time of more than 1 minute is necessary to start a new measurement, in order to avoid vessel wall damage. These methods have also other major drawbacks, such as: peripheral ABP measurement, dependence on an appropriate cuff size, dependence on operator's sensitivity (auscultatory method), no standardized algorithm for identifying either the systolic or diastolic ABP or inconsistent measured values due to the absence of systems validation (oscillometric method) .

To address the need for a continuously non-invasive ABP, several systems have been suggested. Those that are well- accepted among the medical community are based either on the Penaz method or on the applanation tonometry. The Penaz method is based on unloading the arterial wall to measure a calibrated waveform in a finger. A cuff is placed around the middle of a finger and blood pressure is determined by registering the cuff pressure needed to maintain a constant arterial volume, measured by light diodes or photoplethysmography . Applanation tonometry consists in flattening a superficial artery supported by a bone structure, e.g. radial artery, using a pressure transducer called tonometer. Both methods allow a continuous registration of arterial pressure waveform and a beat-to- beat ABP measurement. Nevertheless, they only enable a distal measurement, particularly vulnerable to motion artifacts. In tonometric measurements, a highly skilled operator is required since the positioning of the tonometer tip, as well as the level of compression on the vessel wall, have to be very precise and repeatedly adjusted and recalibrated. Additionally, in order to obtain a calibrated pressure waveform, applanation tonometry requires an external instrument, usually a cuff technique, to achieve the minimum (diastolic) and the maximum (systolic) blood pressure values.

In measurements based on the Penaz method, there is a high risk to occur venous return occlusion due to the permanent cuff inflation. This method also presents other drawbacks, such as: high cost, inconvenience for the patient and relative inaccuracy for measuring absolute levels of blood pressure .

Based on the preceding, there is a clear urge for a noninvasive pressure sensing method and apparatus for measuring continuously the internal blood pressure and other related parameters of a living subject, based on a simple, convenient, continuous and cuffless way, with reduced susceptibility to noise, operator's and subject's movements and relative insensitivity to the positioning of the apparatus on the subject.

Existing literature reveals several attempts to measure the internal pressure of a fluid in a non-invasive way, using exciter-detector units.

A method and apparatus for determining the internal pressure of a sealed container is disclosed in EP0681168B1. This approach uses a striker to excite, at least, the container' s fundamental radial circumferential mode of vibration and its first harmonic and an accelerometer to detect the vibration resulting from the striking of that container. Nevertheless, this approach differs from the present invention in significant aspects, such as : 1) the referred method is based on a vibratory mode analysis while the present one is based on the attenuation's analysis of a permanent oscillation induced by the vibrator and sensed by the accelerometer; 2) the referred apparatus is applied to a sealed substantially cylindrical container while the present apparatus must only be used in a vessel with an elastic wall or window; and 3) the present apparatus presents an unique configuration, based on the combination of one or several exciters/detectors. Other documents were analysed, namely 1) an exciter- detector unit for measuring physiological parameters

(US5904654), 2) an apparatus and method for measuring an induced perturbation to determine blood pressure

(US55590649) and 3) a method and apparatus for the non^¬ invasive determination of arterial blood pressure

(US6471655) . In the first and second documents, the principle of obtaining hemodynamic parameters is based on the induction of a perturbation on the body followed by the detection and further study of the perturbation after it has travelled a certain distance along the body. In the third one, an ultrasound-based apparatus transmits an acoustic wave into the vessel during its compression, receives its echo and estimates pressure within the vessel using a time-frequency distribution of the echo. All these documents diverge from the present one, where a permanent perturbation is mainly used to excite the accelerometer and not the vessel wall, not being re-introduced in the system.

It also should be stated that the use of accelerometers for the non-invasive assessment of arterial blood pressure is also common in the literature, but usually as an auxiliary element that minimizes the error of a measurement, through the elimination of movement artifacts or positioning adjustment (US20070142730, US20110054329) . The use of accelerometers as the main element of the measurement is in fact the most distinguishable feature of the present invention .

As a final remark, it is also important to say that the idea of using acceleration sensors to assess other aspects of the cardiovascular condition was disclosed in 2007, in the project report entitled "Cardioaccelerometery - The assessment of pulse wave velocity using accelerometers". In this report, different apparatus based on acceleration sensors were presented to assess pulse wave velocity and the results obtained, either on the test bench system or in several subjects, suggested the huge potential of these sensors to assess a significant number of hemodynamic parameters. Nevertheless, the present technology uses a totally different approach, both in terms of device's architecture and on its main application.

Summary

The present application discloses an apparatus and method for the non-invasive and continuous measurement of the internal pressure of a fluid confined in a vessel with elastic or rigid walls fitted with an elastic window.

The apparatus is based on a vibrator-accelerometer unit that is mounted on a common support containing at least one vibrator, as exciter, and at least one acceleration sensor, as detector. The vibrator can be a mechanical or electro^¬ mechanical device such as any type of motor or actuator, an electromagnetic loudspeaker of any type or a piezoelectric transducer. The acceleration sensor is an electric, magnetic, optic or electronic device that delivers at least one electrical voltage proportional to the acceleration that is transmitted to him. The method for pressure measurement relies on the amplitude modulation effect that occurs when the fluid pressure transmitted through the vessel wall or window attenuates the vibrator oscillations sensed by the accelerometer . It comprises the electronic treatment and computation of the accelerometer information and the vibration parameters, when the device is placed in mechanical contact with the elastic wall or window of the vessel. The vessel can be of any shape, including tubular, and of any nature, including the vessels of the vascular system of animals and humans, such as arteries and veins.

This technology further comprises a method of simultaneously estimating the morphology of the pressure wave and other related parameters such as vessel wall compliance, distension and distension dynamics and a method of calibrating the estimated pressure waveform based on the re-establishment of the attenuation to a predefined reference amplitude value whatever the force exerted on the vessel wall may be. Although some other possible applications can be found for the present invention, its main focus is the non-invasive assessment of arterial blood pressure and arterial pressure waveform, in humans.

The present technology is expected to have a particular impact in the fields of physiology and medicine, since it can be used for the non-invasive monitoring of ABP and other hemodynamic parameters.

The technology now disclosed aims at fulfilling all the foregoing needs providing an improved instrument and method for the non-invasive measurement of a fluid pressure confined in a vessel with elastic or rigid walls fitted with an elastic window, by means of a combined vibrator- accelerometer unit. The accelerometer is mounted alongside the vibrator in a common platform and the measurement relies on the amplitude modulation effect that occurs when the fluid pressure, transmitted through the vessel wall or window, attenuates the vibrator oscillations sensed by the accelerometer .

Brief Description of the drawings The following descriptions are based on the attached drawings in which, without any limitative intention, are graphically represented the most important features and advantages of the present technology.

Figure 1 is a block diagram of the present technology.

Figure 2a to 2d are cross-sectional views of a possible general configuration of the apparatus with its main elements, referring some of the preferred embodiments.

Figures 2e to 2f are cross-sectional views of an alternative configuration of the apparatus, referring some of the preferred embodiments.

Figure 3 is a 3-D top view of a possible external configuration of the present apparatus.

Figure 4 depicts the main recording sites for arterial pressure assessment in a living subject and the preferred attachment mechanisms for signal acquisition with the present technology.

Figure 5a is a schematic flowchart explaining the general stages of the first preferred method for fluid pressure and other related parameters estimation.

Figure 5b is a schematic flowchart explaining the general steps of the second preferred method for fluid pressure and other related parameters estimation.

Figure 6 is a graphic representation of the typical signals acquired in a test bench system with the present technology, based on a 3-axis accelerometer, right before the probe is in contact with the vessel wall and when it progressively gets in touch with the vessel wall.

Figures 7a to 7c are graphical representations of the signals obtained in a test bench system, with a 3-axis accelerometer based probe and with a manometer that is inserted through the vessel wall.

Figure 7d is a graphical representation of the manometer reference signal and the upper and lower profiles computed from the signals presented in figures 7a to 7c, using the envelope detector approach.

Figure 7e is a graphical representation of the filtered frequency spectra of the product between the modulated carriers and the original carrier wave, when the product detector approach is used.

Figure If is a graphical representation of the manometer reference signal and the time profiles computed from the signals presented in figures 7a to 7c, using the product detector approach.

Figure 7g is a graphical representation comparing the waveform obtained, invasively, with the manometer and the waveform obtained, non-invasively, using the difference between the lower and the upper profiles of the accelerometric signal that is aligned with the direction of the vessel wall movement.

Figure 7h is a graphical representation comparing the waveform obtained, invasively, with the manometer and the waveform obtained, non-invasively, using the total profile of the profiles presented in figure 7d.

Figure 7i is a graphical representation comparing the waveform obtained, invasively, with the manometer and the waveform obtained, non-invasively, using the upper and lower profiles of the accelerometric signal that is aligned with the direction of the vessel wall movement.

Figure 7j is a graphical representation comparing the waveform obtained, invasively, with the manometer and the waveform obtained, non-invasively, using the recovered time profile of the accelerometric signal that is aligned with the direction of the vessel wall movement presented in figure 7f.

Figure 71 is a graphical representation comparing the waveform obtained, invasively, with the manometer and the waveform obtained, non-invasively, using the total profile of the profiles presented in figure 7f.

Figure 7m is a graphical representation of calibrated pressure waveforms resultant from the computation of the lower profile presented in figure 7i and the calibration equation .

Figure 8 is an example of a calibration curve, obtained experimentally, for use in arterial blood pressure assessment .

Figures 9a to 9c are graphical representations of raw (unprocessed) signals acquired in a human subject with a 3- axis accelerometer based probe, when the sensing apparatus is placed on the carotid artery.

Figure 9d is a graphical representation of the upper and lower profiles computed from the signals presented in figures 9a to 9c, using the envelope detector approach. Figure 9e is a graphical representation of the profiles that result from the four preferred approaches: lower and upper profiles of the accelerometric signal that is aligned with the direction of the vessel wall movement, difference between the previous profiles and total profile of the signals presented in figure 9d.

Figure 9f is a graphical representation showing the sum of the upper and lower profiles of the accelerometric signal that is aligned with the direction of the vessel wall movement .

Figure 9g is a graphical representation of calibrated arterial pressure waveforms resultant from the computation of the difference between the lower and the upper profiles presented in figure 9e and the adequate calibration equation .

Figure 9h is a graphical representation of calibrated arterial pressure waveforms resultant from the computation of the total profile presented in figure 9e and the adequate calibration equation.

Figure 9i is a graphical representation of calibrated arterial pressure waveforms resultant from the computation of the lower profile presented in figure 9e and the adequate calibration equation.

Figure 9j is a graphical representation of calibrated arterial pressure waveforms resultant from the computation of the upper profile presented in figure 9e and the adequate calibration equation.

Figure 91 is a graphical representation of one calibrated arterial pressure waveform presented in figure 9g and its main prominent points.

Figure 10a is an example of a calibration curve, obtained experimentally, for pressure measurement of a fluid confined in an elastic vessel.

Figure 10b is a graphical representation comparing the waveform obtained, invasively, with the manometer and the calibrated waveform obtained, non-invasively, with the apparatus through the computation of the difference between the lower and the upper profiles and the adequate calibration curve (figure 10a) .

Description of embodiments

For illustrative purposes, and without any restrictive character, the present technology will be described in detail with reference to the figures. Throughout the following description specific details are clarified in order to provide a more complete understanding to people skilled in the art. This technology is not by any means limited in its application to the construction details and component organization herein illustrated. In a similar way, the results showed in some figures are not limited to what is exposed here. Likewise, the use of syntax and technical wording should not be interpreted as limitative. The use of included' , Containing' , Existing' , Composed' , involving' , and variations of these is made to include the above mentioned and their equivalent.

In one aspect, the technology is a probe 2 and a system for non-invasive and automatic measurement of the internal pressure of a fluid confined in a vessel with elastic or rigid walls or fitted with an elastic window in the wall. In another aspect, the technology is a set of methods for assessing internal pressure of the fluid and other related parameters that are independent of the operator's action, that is to say, on the force that he exerts on the vessel wall. In a preferred embodiment, the probe 2, system and the set of methods are used to assess physiological function, e.g. arterial blood pressure, of a human subject in order to provide baseline information about his cardiovascular status. The technology may also conceivably be adapted to monitor physiological function on other warmblooded species.

Figure 1 illustrates the components and general process of the preferred embodiment. A probe 2 is placed in mechanical contact with one wall or window of the vessel that contains the confined fluid 1. The probe 2 includes a vibrator 4 and an accelerometer 3, mounted on a common platform, and it is connected to an acquisition system 5 and to a signal generator 9 which is responsible for delivering an oscillating electrical signal to the vibrator. The vibrator generates an equivalent oscillating perturbation that is transmitted to the mechanical unit and is sensed by the accelerometer. At the same time, the fluid pressure that is transmitted through the vessel wall or window is also sensed by the accelerometer in such a way that its permanent oscillatory signal derived from vibrator' s excitation is attenuated and modulated in amplitude.

The output of the probe 2 is fed to a conditioning circuit 6 that amplifies the accelerometer signals and monitors the instantaneous current and power of the vibrator 4 using a sensing resistor. The signals are then deliver to a digitizer 7 that converts the analog signals to a digital representation. The digitized signals are supplied to an electronic control device 8 that performs the various steps that constitute the methodology of the technology, including: peak detection 12, profiles extraction 13, artifact removal 14, calibration 11 and estimation of fluid pressure and other related variables 15. The electronic control device 8 output signal is then converted to a useful form for the user by means of a digital or analog display device that contains numerical and graphical information. The electronic control device 8 can also be responsible for controlling and generating the signals that excite the vibrator, by means, for instance, of a direct digital synthesizer (DDS) .

Figures 2a to 2e show preferred embodiments of the probe 2 that constitutes the present technology. In general, the main elements that constitute the probe 2 are: 1) at least one accelerometer 3 ; 2) at least one vibrator 4; 3) a printed circuit board (PCB) or other support board with the respective electrical connections 16; 4) a mechanical interface 17 that is in contact with the vessel wall 18; 5) a semi-rigid bed such as foam, plastic or some type of metalloid 19 that accommodates the main structure and 6) a frame 20 that encloses the assembly and is adapted to be held against the vessel wall 18 by a fixation device or an operator. The vibrator 4 can be a mechanical or electro- mechanical device such as any type of off-axis motor or actuator, an electromagnetic loudspeaker of any type or a piezoelectric. The acceleration sensor is an electric, magnetic or electronic device that delivers one or more electrical voltages proportional to the acceleration that is transmitted to it. The mechanical interface 17 consists in a flattened-shaped-piece or another element with an ergonomic configuration, such as a polymeric material, that allows the transmission of the moving wall distension to the sensing unit. The centre of the mechanical interface 17 is aligned with the centre of the vibrator. In one embodiment, the accelerometer 3, mounted on the same board than the other two elements, is positioned anywhere either alongside the vibrator 4 or alongside the mechanical interface 17 as illustrated in figures 2a to 2d. In another embodiment, illustrated in figures 2e to 2f, the accelerometer 3 is placed on the opposite side of the board where the vibrator 4 stands and is aligned with the centre of the latter one.

There are several possible implementations of the sensing probe in what concerns its arrangement, elements positioning and encasement. In a straightforward and preferred embodiment, illustrated in figures 2a and 2e, there is an accelerometer 3 of the Microelectromechanical System (MEMS) type and one vibrator, contained and connected by a PCB. In another embodiment, illustrated in figure 2b, the accelerometer 3 is of the MEMS type, and there is at least two vibrators 4, contained and connected by a PCB. In particular, this embodiment is suitable for using either two different frequencies of vibration or very similar frequencies, in order to create beats. In another embodiment, illustrated in figure 2c, there is one vibrator 4 and two accelerometers 3 of the MEMS type contained and connected by a PCB. Both accelerometers 3 are equally oriented and well aligned and the output of both signals can be electronically subtracted, in order to remove all common acceleration components. This configuration is suitable mainly for eliminating artifacts or parasitic mechanical interferences that are introduced, for instance, by an operator. This configuration is also adequate in circumstances in which 2-axis MEMS accelerometers are used to obtain 3-axis acceleration information .

In an additional embodiment, illustrated in figure 2d, the accelerometer 3 of the MEMS type and the vibrator 4 are contained and connected by a PCB that enhances the vibrations of the assembly. The PCB presents a distinctive structure where the rear part is fixed to the case with screws 21 and the front part is free and can vibrate with more amplitude due to the middle beam-shaped structure that joins them.

Regarding the sensing probe's mechanical interface, it is possible to identify various preferred configurations. In one embodiment, illustrated in figures 2a-2d, the interface piece is aligned with the centre of the vibrator 4 that is mounted on the opposite side of the PCB and is pressed against a semi-rigid material. In an additional embodiment, illustrated in figure 2f, the interface piece is directly coupled to a non-flexible bridge 22 that is placed in the middle of the vibrator 4. In another possible embodiment, the interface is a polymeric material that covers and/or encapsulates the vibrator 4 in such a way that the whole vibrator's surface may be used to touch the vessel wall 18. In what concerns the vibrator' s support, two different arrangements can be identified. In one embodiment, illustrated in figures 2a to 2d, the vibrator 4 is in contact with the PCB by means of a foot holder 23 that supports it in the centre. In another embodiment, illustrated in figures 2e to 2f, an equilateral configuration in which three equidistant foot holders 23 sustain the vibrator 4 at its periphery is used. In both cases, the foot holders 23 serve as spacers between the vibrator 4 and the PCB.

In figure 3, a feasible 3-D top view configuration of the external configuration of the probe is presented. In a preferred embodiment, the sensing probe is contained in a sensing frame, such as plastic or metal housing 24, and is in contact with the vessel by means of an interface piece 25 that locally compresses the elastic wall. In another arrangement, the sensing probe can be framed on a patch that includes the vibrator-accelerometer unit and is coupled directly to the vessel wall 18, using an attachment mechanism such as an adhesive layer, a band or belt that holds the patch. In figure 4, the preferred probe 2 implementation is applied to measure the arterial blood pressure of a human subject, using two different approaches. In the first one, the encased probe 2 is held by an operator 26, while in the second it is attached to the subject by means of a Velcro® collar 27, a rubber band or any other fixation device known to one skilled in the art. In both cases, the sensor frame that is being held is adjacent to an anatomical position of the subject, and the mechanical interface is overlying, with a certain pressure, a superficial artery where the pulse is palpable, such as: carotid A, braquial B, radial C or femoral D arteries. In figure 5a, the method for measuring fluid pressure according to the present technology is schematically described in a block diagram. The first step to start of the process consists in setting the parameters of excitation of the vibrator 28. The excitation waveform that drives the vibrator 4 can be sinusoidal, square, triangular or of any suitable shape and its frequency can be comprised of a wide range of frequencies and applied either in a continuous or burst mode. Experiments conducted to determine the range of satisfactory frequencies found that a range superior to 150Hz works well with human tissues. Nevertheless, it was verified that the optimal frequency to be used, though conditioned by the accelerometer bandwidth, should correspond to the frequency where the maximum amplitude of the acceleration oscillations is achieved. As such, in a preferred implementation, the first stage of the method involves the determination of this optimal frequency of vibration, which will allow getting the maximum system' s gain .

The acquisition process starts immediately before the probe 2 is positioned and pressed against the vessel wall 18, in order to establish an acceleration carrier amplitude

(reference value for calibration) and record the unconstrained vibrator oscillations sensed by the accelerometer itself 29. As a result of probe's positioning on the vessel wall 30, the oscillations sensed by the accelerometer 3 are attenuated in amplitude. As such, and for calibration purposes the reference amplitude value initially predefined is restored by adjusting the excitation signal that drives the vibrator 31. Then the new

(constrained) amplitude-modulated oscillations sensed by the accelerometer are recorded 32. In a particular embodiment of the technology, the data may be acquired over a short period of time, for example 5s, 30s or 60 seconds. Other data acquisitions times can also be considered by the present technology. In fact, data can be acquired continuously over a long period of time, such as 24-48h. In a preferred embodiment, recordings should be made over the period of time during which it is interesting to cover fluid pressure variability.

Figure 6 illustrates the typical signals acquired in a test bench system with the present technology, based on a 3-axis accelerometer, in this initial acquisition process. The axis c is aligned with the direction of the vessel wall movement and the other two axis a, b are orthogonally lined up with the axis c, from now on, a, b and c refer to the x, y and z directions depicted in figure 2.. Three different regions can be highlighted in the present signals: region Rl shows the oscillations sensed by the accelerometer, right before the probe is in contact with the vessel wall; region R2 coincides with the positioning of the probe when it progressively gets in touch with the vessel wall and R3 represents the accelerometer oscillations, when the probe is placed on the vessel wall, with a certain pressure. The difference between the amplitude of the oscillations from regions Rl and R3 corresponds to the attenuation At_c that is proportional to the force that is exerted on the tube.

The principle of operation of the sensing apparatus is based on the amplitude modulation of the accelerometric signals that takes place when the probe 2 is set on the vessel wall 18 and the fluid pressure, transmitted through it, attenuates the vibrator 4 oscillations sensed by the accelerometer 3. In a straightforward model and considering that the probe 2 is submitted to a sinusoidal displacement x(t) induced by the vibrator then:

x(t)= L.sin(w.t + 9(t)), (Eq.l)

where L is the amplitude of the oscillation, w is the angular frequency of the oscillation and 6(t) is the phase of the oscillation.

When the probe 2 is unconstrained, the sinusoidal waveform x(t) is sensed by the accelerometer as:

a(t) = - L.w².sin(w.t + Θ (t) ) = -w².x(t) (Eq.2)

This waveform is known as the carrier wave or carrier and corresponds to an unconstrained and therefore unmodulated oscillation .

When the probe 2 is placed against the vessel wall 18, the amplitude L suffers an immediate attenuation At_c (0< At_c <100%) that is proportional to the force that is applied on the vessel wall F and also to the resistance that is offered by the vessel wall 18 to the deformation (stiffness k) . The fluctuations in fluid pressure P are also transmitted through the vessel wall 18, modulating the oscillations sensed by the accelerometer 3 through attenuation. At this stage, the fluid pressure variation is defined as the modulating signal and the accelerometer signal corresponds to the amplitude-modulated carrier.

Thus

a( t ) = -At_c.w .x( t ) (Eq. 3)

At_c a ai.F (Eq. 4)

At_c oc a₂. k (Eq. 5)

where a±, a₂, a₃ are calibration constants. The attenuation At_c of the modulated data can be measured up to 100% in the total amount of amplitude L.

Additionally to the accelerometer signals, the amplitude modulation process is also verified when the electrical parameters of the vibrator 4 are measured. As such: p(t) = Atc . i (t) .v(t) (Eq. 7)

At_c a b₂.k (Eq. 9)

where i (t) and p(t) are the current and power signals of the vibrator 4. v(t) is the voltage delivered to the vibrator 4 by a signal generator 9, matching the carrier wave x(t) and bi, b₂, b₃ are calibration constants.

Referring again to figure 5a, and after the modulation process 33, a reverse process of demodulation 45 is initiated on the electronic control device 8, with peak detection of the respective modulated amplitude oscillations 34. The local maxima and minima are calculated using a threshold based approach and then interpolated, where in the upper and lower envelopes of the modulated carrier are extracted 35. These envelopes, corresponding to the modulating signal, undergo a filtering stage to remove the high frequency noise 36 that is typically present in this type of measurements.

The extracted lower and upper profiles ( lp, up ) are the main sources of information of fluid pressure fluctuations, allowing a direct estimation of the fluid pressure waveform or even the extraction of operator handling artifacts. Different methods, based on the use of at least one profile 38, can be used to assess the pressure wave morphology 40. The simplest method uses either the upper or the lower profiles extracted from the accelerometer 3 signal that matches the direction of the vessel wall 18 movement ( ΙΡ_ηα_γ,ιψτ_ηον) · Another method requires the determination of the difference between the lower and the upper profiles 37, extracted also from the accelerometer 3 signal that matches the direction of the vessel wall 18 movement ( lp_mov - up_mov ) . A third method, applied when the technology is based on a 3- axis accelerometer, involves the computation of the total profile 39, ( total _ profile _a ) extracted from the various orthogonal accelerometer signals, where:

aux_ mov = lp_mov - up_mov (Eq. 10)

aux _ orth_l = lp_ol -up_ol (Eq. 11)

aux _ ort ₂ = lp_o2—up _o2 2) total _ profile _ a = ^aux _ mov ² + aux _ orth_x ² + aux _ orth₂ ² (Eq. 13)

And, lp_0\ -up_oX , lp_o2-upo2 ^are the differences between the lower and the upper profiles, extracted from the accelerometer 3 signals that match the orthogonal directions to the vessel wall 18 movement.

It is also possible to extract relevant information when the upper or the lower profiles extracted from the accelerometer 3 signals are added up { lp + up ) 43. With this approach, it is possible to remove the mean baseline fluctuations of each one of the accelerometer 3 signals. These baseline fluctuations reproduce the artifacts that are typically present in this type of measurements and also the movement induced to the probe by the vessel wall 18 distension 44. In the case of current or power modulated signals sensed by the vibrator 4, the approaches to assess the pressure wave morphology are the same. The only difference is that the method applied when the probe 2 is based on a 3-axis accelerometer ( total _ profile _ a ) 39 is not used, since no vectorial quantities are present. As a result, the methods that employ simply the upper or the lower profiles extracted from the current and power signals as well as the methods that involve the determination of their difference or sum can be applied.

Regarding the process of demodulation 45, it is also feasible to apply a further signal processing method, based on a product detector approach to acceleration, current and/or power modulated signals. This method, illustrated in figure 5b, allows the recovery of the modulating signal through the following steps: a) computation of the product of the modulated carrier (s) and the original carrier wave, i.e. excitation waveform that drives the vibrator 47; b) computation of the frequency spectra using the fast Fourier transform algorithm 48; c) low pass filtering of the frequency spectra 49 and d) computation of the time profiles by means of the inverse Fast Fourier Transform algorithm 50. The recovered time profiles, corresponding to the modulating signal, then undergo a smoothing stage to remove the high frequency noise 51.

Similarly to the previous signal processing method (envelope detector approach - figure 5a) , with this method (product detector approach - figure 5b) , the morphology of the pressure wave can also be assessed using one or several profiles 52. The simplest method uses the profile extracted from the accelerometer/current/power signal that matches the direction of the vessel wall 18 movement ( p_mov ) . The other method, applied when the technology is based on a 3-axis accelerometer, involves the computation of the total profile { total _ profile _b ) 53, extracted from the various orthogonal accelerometer signals ( p_mov p_ol p_{o2 \} , where: aux _ mov = p_mov (Eq . 14)

aw _ orth_l = p_ol (Eq. 15)

a x _ orth₂ = p_o2 (Eq. 16) total _ profile _ b = ^aux _ mov ² + aux _ orth_x ² + anx _ orth₂ ² (Eq. 17) and ^po\ P l ^are the profiles that match the orthogonal directions to the vessel wall movement.

After the stage 45 and in order to quantify the pressure waveform and other related parameters, it is necessary to proceed to the calibration and parameters assessment stage 46. The calibration method is focused on the use of calibration curves (acceleration-pressure type), who's determination and application demand the conservation of the acceleration sensed by the apparatus. In other words, the acceleration carrier amplitude measured when the sensing probe is freely vibrating must be the same as that measured when it is pressed against the vessel wall.

In a preferred embodiment, the electronic control device 8 manually or automatically chooses the adequate calibration curve from the database of curves to be used, according to the acceleration carrier amplitude established as reference. After this step, the implementation of an automatic acceleration-voltage control loop is crucial for the efficiency of the calibration process. The loop constantly measures the attenuation of the acceleration carrier when the sensing tip is placed on the elastic wall, re-establishing the reference amplitude value through the adjustment of the vibrator's input voltage. As such, whatever the force exerted on the vessel wall 18 may be during the acquisition process, it does not interfere with the calibration process.

In addition to fluid pressure and its waveform' s morphology, there are other general related parameters that can be calculated by means of the present technology.

The distension of vessel wall 18, for example, can be inferred when the sensing unit is not being excited, and the accelerometer 3 only senses the displacement changes associated with the pressure wave propagation. It is also possible to assess vessel wall 18 distension, when the sensing unit is being excited, by means of the sum of the upper and lower profiles, extracted from the modulated accelerometer 3 signals (Ip + up) . Independently of the method that is used, it's necessary to end the process with a double integration of the obtained acceleration signal, to quantify the total displacement of the vessel wall 18.

The compliance of the vessel ( Q ₌ ^AV ) defined as the

AP

change of volume (AV ) per unit pressure change {AP) , is another parameter that can be easily estimated. ΔΡ is obtained directly from the calibrated pressure waveform while AV is calculated based on the geometry and distension of the vessel.

Since other significant parameters can also be determined, three different applications of the sensing probe, system and methods will be presented, leveraging the usefulness of the present technology in several fields.

Examples Embodiment 1

This embodiment aimed at non-invasively measuring the morphology of a pressure wave that propagates through a fluid confined in an elastic tube. A special purpose test bench was developed, capable of originating arbitrarily shaped pressure waveforms through a latex tube with diameter and wall thickness of the same order of magnitude as that of a large artery. The pressure waves were generated manually by a piston mechanism placed at one of the extremities and launched in a 70cm long tube (12.7mm inner diameter, 0.8mm wall thickness), filled with water. At the other extremity, a manometer placed longitudinally to the tube monitored the DC pressure level and the pressure fluctuations. Measurements were performed over a short period of time and close to the pressure sensor, using the apparatus and the method according to the present technology. The sensing probe, based on a 3-axis accelerometer, was excited by a continuous 700Hz sinusoidal waveform and held by a fixation device that allowed maintaining a constant force, while the probe 2 was against tube's wall sensing pressure variations. Data acquisition was performed through a dedicated acquisition system 5, based on a 16-bit digitizer. All the signals were sampled at 12.5 kHz and stored for offline analysis. The accelerometric data were then compared with the data taken by the manometer, used as the reference device in pressure assessment .

The acquired data are presented in figures 7a to 7c, where a, b, c are modulated carriers and d is the manometer reference signal, c is the modulated carrier sensed by the accelerometric axis aligned with the direction of the vessel wall 18 movement and the other two modulated carriers (a, b) are sensed by the accelerometric axes orthogonally lined up with the first one. Based on local maxima and minima determination (envelope detector approach) , the upper and lower profiles of the modulated carriers were extracted, as illustrated in figure 7d, and then slightly smoothed, where: al, bl, cl are the lower profiles and a2, b2, c2 are the upper profiles of the previous described modulated carriers. Simultaneously, and based on the product detector approach, it was also possible to recover the time profiles of the modulated carriers (a, b, c) , identified in figure 7f, respectively as a3, b3 and c3. After computing and filtering the frequency spectra of product between the modulated carriers and the original carrier wave, as illustrated in figure 7e, it was applied the inverse fast Fourier Transform algorithm to obtain the referred time profiles.

In figures 7g to 71, graphical representations comparing the waveform obtained, invasively, with the manometer and the waveforms obtained, non-invasively, using the different methods of the present invention are illustrated. The root mean square errors (RMSE) between the reference waveform and the waveforms obtained using different methods were determined and are summarized in table 1.

Table 1 RMSE measurements between the reference waveform (manometer) and the waveforms obtained using different methods of the present technology.

Demodulation

Method RMSE (%) Process

Upper and lower

profiles difference

Envelope Detector 2.62

(e) Total Profile (f) 3.64

Upper Profile (c2) 3.41

Lower Profile (cl) 1.94

Time Profile (c3) 3.88

Product Detector Total Profile (f3) 4.55

The cl, c2 and c3 profile is obtained through the accelerometric axis that is aligned with the direction of the vessel's wall 18 movement.

The present study shows that the different methods used by the present technology have the capability of precisely rendering the original pressure waveform, evidencing a better performance when the lower profile method or the profiles difference approach are used.

In addition to pressure waveform assessment, it was also possible to calibrate the pressure waveform obtained for the different methods and validate the general relationship between fluid pressure and the sensed acceleration. Figure 7m exemplifies a calibrated pressure waveform resultant from the computation of the best method (lower profile cl) and the calibration equation, obtained experimentally, in this study, as illustrated in figure 8. The presented range of values covers the range of the typical normal values of a healthy human artery.

Embodiment 2

This study aimed at non-invasively assessing to the calibrated arterial pressure waveform (APW) at the ascending aorta of a human subject. APW was analysed at a central level, where the true load imposed to the left ventricle and to the large central artery walls is represented. Measurements were performed over a short period of time and in the common carotid artery of a healthy subject, using the apparatus and the method according to the present technology.

The sensing probe 2, based on a 3-axis accelerometer, was excited by a continuous 700Hz sinusoidal waveform and held by an operator that attempted to maintain a constant pressure on the carotid artery. Data acquisition was performed through a dedicated acquisition system 5, based on a 16-bit digitizer. All the signals were sampled at 12.5 kHz and stored for offline analysis.

The acquired data are presented in figures 9a to 9c, where c is the modulated carrier sensed by the accelerometric axis aligned with the direction of the vessel wall 18 movement and a, b are the modulated carriers sensed by the accelerometric axes orthogonally lined up with the first one. With the intention of demodulation it was used the envelope approach, where the upper and lower profiles of the modulated carriers were extracted, as illustrated in figure 7d, and then low-pass filtered, al, bl, cl are the lower profiles and a2, b2, c2 are the upper profiles of the previous described modulated carriers.

In figure 9e, the waveforms obtained, non-invasively, using the different methods of the present technology are illustrated: e corresponds to the upper and lower profiles difference method, f to the total profile method, cl to the lower and c2 to the upper profiles3 approach. An additional method, based on the sum of the upper and lower profiles3 was also applied, in order to obtain the mean baseline fluctuations of the accelerometric data, illustrated in figure 9f. The obtained pattern represents not only the wall acceleration waveform, but also some artifacts that could be induced by the operator, during acquisition. The profile is obtained through the accelerometric axis that is aligned with the direction of the vessel's wall 18 movement .

In order to calibrate the pulsatile APW, it was established a reference acceleration carrier amplitude for the axis aligned with the direction of the vessel wall 18 movement and then, it was chosen the calibration curve from the set of curves that matched the same reference value. In figures 9g to 9f, the resultant calibrated APWs for each one of the methods are depicted, evidencing values for the diastolic (DP) and systolic (SP) pressures very similar. The values that were obtained either for DP, SP and pulse pressure are in the range of normal values of a healthy subject.

In the present application, other data analysis could also be made, in order to assess other type of hemodynamic parameters. In fact, with the present information, it is possible, for example, to determine mean ABP, heart rate or identify the prominent points in the APW, as illustrated in figure 91, such as: wave foot (WF) , systolic peak (SP) , reflection point (RP) , dicrotic notch (DN) or dicrotic peak

(DP) . These fiducial points can then be employed for determination studies of other important indexes, such as: augmentation index (Aix) , left ventricular ejection time

(LVET) , Buckberg index, tension time index, diastolic time index, or other related parameters known to one skilled in the art .

The results of this study show that this technology has the potential to be an important tool for cardiovascular risk assessment, providing significant information on the evaluation of central arterial pressure, for the descriptive analysis of APW and for the assessment of important hemodynamic parameters.

Embodiment 3

This study aimed at non-invasively measuring the water pressure confined in a latex tube for different applanation positions .

Measurements were performed in the test bench system described in embodiment 1, using the apparatus and the method according to the present technology. The sensing probe, rigidly attached to a tri-axial position monitoring system, was placed at the end of the latex tube, close to the manometer and was excited by a continuous sinusoidal waveform whose frequency matched its mechanical resonance value. Ten distinct measurements with successively higher applanation positions were performed over a short period of time, aiming to acquire the pressure wave generated by a piston that was manually maneuvered by an operator. In the applanation process, the linear positioner system travelled to positions ranging from 80500 to 113500 microsteps, guarantying that the PZ's central part was in direct contact and flattening the latex tube. The several acquisitions were carried out following the conservation of the acceleration carrier amplitude, more precisely the value pre-established in the calibration curve to be used, illustrated in figure 10a. As such, for each applanation position, the input voltage of the excitation signal that drives the vibrator was adjusted in order to restore the reference calibration value. Subsequently, the pressure wave was generated and acquired by the apparatus. Data acquisition was performed through a dedicated acquisition system, based on a 16-bit digitizer. All the signals were sampled at 12.5 kHz and stored for offline analysis. For each acquisition, the calibration curve was applied to the acceleration modulating profile recovered using the envelope detector algorithm and the difference between the lower and the upper profiles method. The main characteristic pressure values (maximum, minimum, mean and difference) obtained with the technology after calibration were compared with the reference pressure values obtained with the manometer.

The results obtained are summarized in table 2 and in figure 10b is possible to observe a graphical representation comparing the waveform obtained, invasively, with the manometer and the calibrated waveform obtained, non-invasively, with the apparatus in set Al .

The successive changes in tube's applanation do not impair the accurate estimation of the main pressure values using the present technology. The estimates of maximum, minimum, mean and difference pressure values are on average, 1.53 mmHg, 0.41 mmHg, 0.78 mmHg and 1.33 mmHg higher than those measured respectively with the manometer. The accomplishment of this accuracy criteria in test bench experiments seems to be a very good indicator of the feasibility of this technology in assessing arterial blood pressure, namely when the pressure values generated inside the latex tube are mostly in the range of those found in a healthy subject or with cardiovascular disease.

It is clear from the results obtained that the most important strength of the present technology, namely of the calibration method is its validity for different applanation positions and hence its inherent independence from an operator action. Whatever the force exerted on the vessel wall may be, the procedure foresees the re- establishment of the attenuation to the calibration reference value.

A number of embodiments have already been described. However, several modifications can be made without leaving the scope of the invention, as defined by the following claims .

Table 2 Comparison of the measured (manometer) and the estimated (apparatus) pressure values for different applanation positions in a latex tube filled with water.

¹ Applanation position of the sensing probe on the tube (in microsteps) . The movement of a microstep is exactly 0.09921875 ~ 0.1 pm.

² Difference between the maximum and the minimum pressure values.

³ Difference between the maximum, minimum and mean pressures values obtained with the sensing probe and the manometer (reference) .

Claims

1. An apparatus for non-invasively measuring the pressure of a fluid confined by a vessel with elastic or rigid walls or fitted with an elastic window in the wall, comprising the following components:

a) a sensing probe, held by a sensor frame, including, at least one vibrator that excites the assembly, at least one accelerometer for measuring the amplitude vibration fluctuations, a printed circuit board (PCB) or other support board where both transducers are mounted and electrically connected; a mechanical interface that is in contact with the vessel wall and a semi-rigid bed that accommodates the main structure;

b) an acquisition system including a conditioning circuit that amplifies the accelerometer signals and measures the instantaneous electrical current and power of the vibrator using a sensing resistor;

c) a digitizer that converts the analog signal to a digital representation;

d) an output unit, based on an electronic control device, configured to receive the modulated accelerometer signal and adapted to apply the signal processing methods that extract the modulating signal, calibrate it and therefore obtain parameters that correspond to the fluid pressure, pressure wave morphology and other related parameters.

2. An apparatus according to claim 1, wherein the sensor frame that encloses the assembly is adapted to be held against the vessel wall by a fixation device or an operator .

3. An apparatus according to claim 1, wherein the vibrator is a mechanical or electro-mechanical device such as any type of off-axis motor or actuator, an electromagnetic loudspeaker of any type or a piezoelectric.

4. An apparatus according to claim 1, wherein the accelerometer is an electric, magnetic, optic or electronic device that delivers one or more electrical voltages proportional to the acceleration that is transmitted to it.

5. An apparatus according to claim 1, wherein the mechanical interface presents an ergonomic configuration that allows the transmission of the moving wall distension to the sensing unit.

6. An apparatus according to claim 1, wherein the accelerometer is placed on the opposite side of the board where the vibrator stands and is aligned with the centre of the latter one.

7. An apparatus according to claim 1, wherein the centre of the mechanical interface is aligned with the centre of the vibrator and the accelerometer, mounted on the same board, is positioned anywhere either alongside the vibrator or alongside the mechanical interface.

8. An apparatus according to claim 1, wherein the vibrator is sustained and in contact with the PCB by means of at least one foot holder that supports it in the centre or at the periphery using a geometric configuration.

9. An apparatus according to claim 1, wherein the electronic control device controls and generates the signals that excite the vibrator (s) .

10. An apparatus according to claim 9, where the excitation waveform that drives the vibrator is sinusoidal, square, triangular or of any suitable shape and its frequency can be greater than 150Hz and applied either in a continuous or burst mode.

11. A method for non-invasively measuring the pressure of a fluid confined by a vessel with elastic or rigid walls or fitted with an elastic window in the wall, comprising the following steps:

• adjusting the sweep parameters of the at least one vibrator, including the determination of its optimal vibration frequency;

• starting the acquisition process immediately before the probe is positioned on the vessel wall in order to establish an acceleration carrier amplitude and record the unconstrained vibrator oscillations sensed by the at least one accelerometer itself;

• positioning the probe on the vessel wall, by means of a fixation device or an operator;

• Re-establishing the reference acceleration carrier amplitude by adjusting the excitation signal that drives the at least one vibrator;

• acquiring the new amplitude oscillations sensed by the at least one accelerometer, modulated by the pressure fluid' s fluctuations - the modulated carrier;

• applying signal processing methods that extract the modulating signal, calibrate it and therefore obtain parameters that correspond to the fluid pressure, pressure wave morphology and other related parameters; • displaying the calculated parameters.

12. A method according to claim 11, wherein the attenuation of the modulated data is measured up to 100% in the total amount of the unconstrained vibrations amplitude.

13. A method according to claim 11, wherein the processing methods are applied to the current and power amplitude- modulated carriers sensed by the at least one vibrator.

14. A method according to claim 11, wherein the processing methods are applied to data segments with short duration, e.g. 30s, or long duration, e.g. 24h-48h.

15. A method according to claim 11, wherein the peaks of the modulated carriers data are determined and then interpolated .

16. A method according to claim 15, wherein demodulation is accomplished when the upper and lower profiles of the modulated carriers data, modulating signals, are determined .

17. A method according to claim 16, wherein the upper and lower profiles undergo a filtering stage, with removal of high frequency noise.

18. A method according to claim 11, wherein the frequency spectra of the product of the modulated carrier (s) and the unmodulated carrier is determined.

19. A method according to claim 18, wherein demodulation is accomplished when the frequency spectra of the product is low-pass filtered and then the time profiles, modulating signals, are recovered.

20. A method according to claim 19, wherein the recovered profiles undergo a filtering stage, with removal of high frequency noise.

21. A method according to claim 11, wherein the step of determining the pressure waveform uses the upper profile, the lower profile or other time profiles extracted from the accelerometric signal that matches the direction of vessel wall movement.

22. A method according to claims 11 and 13, wherein the step of determining the pressure waveform uses the upper profile, the lower profile or other time profiles extracted from the electrical signals of the vibrator.

23. A method according to claim 11, wherein the step of determining the pressure waveform uses the difference between the lower and the upper profiles extracted from the accelerometric signal that matches the direction of vessel wall movement.

24. A method according to claims 11 and 13, wherein the step of determining the pressure waveform uses the difference between the lower and the upper profiles extracted from the electrical signals of the vibrator.

25. A method according to claim 11, wherein the step of determining the pressure waveform involves the computation of the total profile extracted from the several orthogonal accelerometric signals, when the said method is based on a 3-axis accelerometer .

26. A method according to claim 11, wherein the step of removing the mean baseline fluctuations uses the sum of the upper and lower profiles of each one of the accelerometric signals or the vibrator's electric signals.

27. A method according to claim 11, wherein the step of calibrating the pressure waveform relies on calibration curves whose determination and application require the conservation of the acceleration carrier amplitude, i.e. the amplitude measured when the sensing probe is freely vibrating is the same as that measured when it compresses the vessel wall.

28. A method according to claims 11 and 27, wherein an automatic acceleration-voltage control loop constantly measures the attenuation of the acceleration carrier when the sensing tip is placed on the elastic wall, re^¬ establishing the reference amplitude value through the adjustment of the vibrator's input voltage.

29. A method according to claim 11, wherein the extracted parameters are at least fluid inner pressure, fluid pressure waveform, vessel distension and vessel various compliance .

30. The method according to claim 11, wherein the vessel is an artery.

31. The method according to claim 30, wherein the recording location is located in a superficial artery, where the pulse is palpable.

32. A method according to claims 27 and 30, wherein the extracted parameters are among others heart rate, arterial pressure waveform (APW) , systolic blood pressure, diastolic blood pressure, mean arterial pressure, pulse pressure, augmentation index, and other related hemodynamic parameters extracted from the prominent points of the calibrated pressure waveform.