Method for determining a pressure offset, correction method, and implantable medical device

Abstract

The invention relates to a method for determining an offset between an estimated fluid pressure and an actual fluid pressure in an inflatable element of an implantable medical device (1), an estimated fluid pressure value being calculated from a measurement performed by a sensor (4), the method comprising the steps of: - S0: obtaining a plurality of points, each point being defined by coordinates corresponding to an injected fluid volume value and to a value representative of the estimated fluid pressure, the plurality of points defining a curve; - S1: determining an inflection point of the curve from the plurality of points; - S2: determining the offset between the estimated fluid pressure and the actual fluid pressure, the offset being the estimated fluid pressure value at the determined inflection point.

Description

[0001] METHOD FOR DETERMINING A PRESSURE DISPLACEMENT, METHOD FOR CORRECTION AND IMPLANTABLE MEDICAL DEVICE

[0002] TECHNICAL FIELD

[0003] This disclosure relates to the general field of sensor measurement correction processes, and more specifically to pressure measurement offset estimation for sensors for implantable medical devices, including occlusive systems of an anatomical conduit of a living being's body.

[0004] STATE OF THE ART

[0005] It is known to occlude an anatomical duct, for example the urethra, using an implantable occlusive system. Typically, in artificial urinary sphincters, occlusion of the urethra or bladder neck is achieved by an inflatable cuff filled with fluid that exerts varying degrees of pressure on the external walls of the urethra or bladder neck around which it is positioned, depending on the volume of fluid injected into the inflatable cuff.

[0006] The fluid volume can be regulated by an actuator coupling a variable-volume fluid reservoir via a fluidic linkage to the inflatable cuff. The inflatable cuff, fluid reservoir, and the fluidic linkage together form a fluidic circuit. For example, the actuator can transfer fluid between the reservoir and the cuff by injecting fluid from the reservoir into the cuff to increase the pressure exerted on the anatomical conduit when the conduit's cross-section needs to be reduced, or conversely, by transferring fluid from the cuff into the reservoir to reduce the pressure exerted on the anatomical conduit when the conduit's cross-section needs to be increased, particularly to facilitate urination.

[0007] To enable effective regulation, it is necessary to have access to a pressure measurement in the inflatable sleeve or, more generally, at some point in the fluid circuit. Using a sensor implanted in the inflatable sleeve presents several challenges. The sensor must be biocompatible, stable over time, and compact. Furthermore, the sensor must be completely sealed and resistant to the humid environment in which it is integrated.

[0008] An example of an implantable occlusive system incorporating a sensor to address these difficulties was proposed in WO 2016 / 083428 A1. However, it has been observed that the pressure induced by the volume of fluid injected into the inflatable cuff, after implantation of the medical device with the occlusive system positioned around the patient's anatomical canal, does not correspond to the actual pressure. This can lead to malfunction of the implantable occlusive system, specifically poor occlusion of the anatomical canal if the estimated pressure in the occlusive cuff is higher than the actual pressure, resulting in leakage, or conversely, a risk of damage or injury to the anatomical canal if the estimated pressure in the occlusive cuff is lower than the actual pressure, as the high actual pressure excessively stresses the external walls of the anatomical canal.

[0009] The measurement discrepancy can be due to various factors, including mechanical aspects of the implantable occlusive system, such as the stiffness constant of the variable-volume reservoir. For example, the sensor may drift over time, and the mechanical properties of the implantable occlusive system components, which are considered in estimating the pressure variation, may vary. Furthermore, the fluid pressure measurement may be offset depending on the patient's position in which the occlusive system is implanted. For instance, when using the implantable occlusive system for a human urethra, a column of water between the device's reservoir, in which the sensor is placed, and the occlusive cuff can induce a pressure difference when the patient is standing.Finally, the measurement discrepancy may be related to the pressure and temperature conditions of the environment inside the body of a living being in which the medical device is implanted.

[0010] To apply the correct pressure to the anatomical conduit to be occluded, it is therefore necessary to first estimate the fluid pressure measurement discrepancy in the occlusive cuff, induced by all the different factors. Similar issues arise more generally in the context of implantable medical devices containing an inflatable element.

[0011] DESCRIPTION OF THE INVENTION

[0012] One objective of this disclosure is to estimate the discrepancy between a fluid pressure estimated from a sensor measurement and an actual fluid pressure, so that the estimated fluid pressure can then be corrected so that the inflatable element exhibits the desired fluid pressure to perform its function. For example, in the case of an occlusive system, one objective of the invention is to ensure that the inflatable element exerts the correct pressure on the anatomical duct to be occluded.

[0013] This goal is achieved by the method of determining the difference between an estimated fluid pressure and an actual fluid pressure in an inflatable element of an implantable medical device, an estimated fluid pressure value being calculated from a sensor measurement, the sensor being configured to measure a physical quantity dependent on the actual fluid pressure, the determination method comprising the steps of:

[0014] obtaining a plurality of points, each point being defined by coordinates corresponding to a value of injected fluid volume and a representative value of the estimated fluid pressure, the plurality of points defining a curve;

[0015] determination of an inflection point of the curve, from the plurality of points; determination of the offset between the estimated fluid pressure and the actual fluid pressure, the offset being the estimated fluid pressure value of the determined inflection point.

[0016] The resulting curve represents the estimated dynamics of the implantable medical device. This curve allows observation of the difference in behavior of the implantable medical device between a fluid injection phase (overpressure), corresponding to the points on the curve where the actual pressure is positive, and a fluid withdrawal phase (underpressure), corresponding to the points on the curve where the actual pressure is negative. The curve exhibits an inflection point between these two phases. The shift to be determined corresponds to the fluid pressure at this inflection point.

[0017] By definition, the inflection point is the point on the curve where the slope is minimal. However, due to noise or measurement errors, calculating slopes alone is insufficient to determine the offset without error. Therefore, it is proposed to filter the calculated slope values ​​by taking into account the slope value calculated for a nearby point on the curve.

[0018] This helps to avoid "false positives" in identifying the inflection point of the curve, and therefore to obtain a more reliable determination of the offset between the estimated pressure values ​​and the actual fluid pressure in the occlusive cuff.

[0019] The invention is advantageously complemented by the following features, taken individually or in any of their technically possible combinations:

[0020] - Determining the inflection point includes:

[0021] calculation of a plurality of slope values, each slope value being calculated for a point on the curve, from the coordinates of the point and the coordinates of at least one neighboring point on the curve, selection of a point from the plurality of points for which the slope value is minimal.

[0022] - The at least one neighboring point includes a next point and a previous point, so that each slope value is calculated for a point on the curve of index i, from the coordinates of the point of index i, the coordinates of the previous point on the curve of index i-1, and the coordinates of the next point on the curve of index i+1.

[0023] - Determining the inflection point also includes:

[0024] Calculation of a plurality of filtered slope values, each filtered slope value being calculated for a point on the curve, based on the slope value calculated for the point and at least one slope value calculated for a neighboring point on the curve.

[0025] selection of a point from the plurality of points, for which the filtered slope value is minimal.

[0026] - Each filtered slope value is calculated for a point on the curve, from the slope value calculated for the point and a distance between the slope value calculated for the point and a slope value calculated for a neighboring point on the curve.

[0027] - The at least one neighboring point includes a following point and a preceding point, so that each filtered slope value is calculated for a point on the curve with index i, by the formula:

[0028] >

[0029]

[0030] with Si the slope value calculated for the point with index i, S i-1 the slope value calculated for the previous point on the curve with index i-1, and S i+1 the slope value calculated for the next point on the curve with index i+1.

[0031] - The process also includes the following steps:

[0032] obtaining a plurality of approximated points from the plurality of points by a polynomial regression, the plurality of approximated points being defined by coordinates corresponding to the value of the volume of injected fluid and to an approximate value of pressure, the approximate value of pressure satisfying Si" = J'Cyj, with P a polynomial of order n > 2;

[0033] calculation, for each approximated point, of a value of an indicator, from the coordinates of the approximated point, and the coordinates of a neighboring point; calculation of a plurality of average values, each average value being calculated for a point of the plurality of points as the average between the filtered slope value calculated for the point and the indicator value calculated for the approximated point of the same index;

[0034] and the determination of the inflection point further includes a selection of a point for which the average value is minimal.

[0035] - Each value of indicator E t is calculated for an approximate point of index i as a distance

[0036] between the approximate pressure value S"t of the point with index i and the approximate pressure value S" i-1 from a previous point of index i-1,

[0037] between the value of the injected fluid volume V[ of the point with index i and the value of the injected fluid volume V i-1 from the previous point of index i-1,

[0038] between the approximate pressure value S"t of the point with index i and the approximate pressure value S" i+1 from a subsequent point with index i+1, and

[0039] between the value of the injected fluid volume V[ of the point with index i and the value of the injected fluid volume V i+1 of the next point with index i+1, according to the formula

[0040]

[0041] " "

[0042] - The process also includes the following steps:

[0043] normalization of the plurality of filtered slope values, so as to obtain a plurality of normalized slope values;

[0044] normalization of the plurality of values ​​of the indicator, so as to obtain a plurality of normalized indicator values;

[0045] and each average value is calculated for a point on the curve as the average between the normalized filtered slope value calculated for the point and the normalized indicator value calculated for the point.

[0046] - The normalized slope value NS- is calculated for a point with index i, where index i is an integer between 1 and N, and N is the number of points defining the curve, as a ratio between

[0047] a difference between the slope value S- calculated for the point with index i and a minimum slope value among the plurality of slope values ​​calculated for the points, and a difference between a maximum slope value among the plurality of slope values ​​calculated for the points and the minimum slope value according to the formula: >

[0048]

[0049] minimum slope calculated for points with index j including the maximum slope value calculated for points

[0050]

[0051] index j between 3 and N-2.

[0052] - The normalized indicator value NE t is calculated for a point of index i, the index i being an integer between 1 and N, N being the number of points on the curve, as a ratio between

[0053] a difference between the indicator value E t calculated for the index point i and a minimum indicator value among the plurality of indicator values ​​calculated for the approximated points, and a difference between a maximum indicator value among the plurality of indicator values ​​calculated for the approximated points and the minimum indicator value, according to the formula:

[0054]

[0055] with min(Ej). ejlw j is the minimum indicator value calculated for points on the index curve j between 1 and the maximum indicator value calculated for

[0056]

[0057] the points on the curve with index j between 1 and N.

[0058] According to another aspect, a method is proposed for correcting a fluid pressure value estimated from a sensor measurement, the sensor being configured to measure a physical quantity dependent on the actual fluid pressure, in an inflatable element of an implantable medical device, the correction method comprising the steps of:

[0059] estimation of a fluid pressure value from the measurement of the physical quantity by the sensor;

[0060] determination of a discrepancy between the estimated fluid pressure value and an actual fluid pressure, by implementing the discrepancy determination method as described above;

[0061] obtaining a corrected fluid pressure value, the corrected pressure value being equal to a difference between the estimated fluid pressure value and the determined offset. According to another aspect, a computer program product is proposed comprising instructions which, when the program is executed by a computer, cause the computer to implement the correction process described above.

[0062] According to another aspect, a computer-readable recording medium is proposed, comprising instructions which, when executed by a computer, lead the computer to implement the correction process described above.

[0063] In another aspect, a medical device is proposed that can be implanted in a human or animal body, comprising:

[0064] - a fluidic circuit comprising:

[0065] - an inflatable element into which a volume of a fluid is intended to be injected,

[0066] - a suitable reservoir to contain the fluid intended to be injected into the inflatable element and fluidically connected to the inflatable element,

[0067] - an actuator configured to transfer a volume of fluid between the tank and the inflatable element;

[0068] - a sensor configured to measure a physical quantity dependent on the actual fluid pressure in the inflatable element;

[0069] - a processing unit, configured for:

[0070] - obtain a corrected fluid pressure value by implementing the correction process described above;

[0071] - determine a volume of fluid to be transferred between the tank and the inflatable element, based on the corrected fluid pressure value;

[0072] - to control the actuator to transfer the specified volume of fluid. The implantable medical device is advantageously complemented by the following features, taken individually or in any technically feasible combination thereof:

[0073] - The tank is a variable volume tank comprising a fixed part and a moving part, the actuator is mechanically coupled to the moving part of the tank, and configured to linearly move the moving part relative to the fixed part in a direction of movement to adjust the variable volume of the tank.

[0074] - The sensor is a force sensor mechanically linked to the actuator and / or the moving part of the tank, the force sensor being configured to measure a tensile and / or compressive force in the direction of travel, the processing unit being further configured to

[0075] to estimate fluid pressure from the force measured by the force sensor, and

[0076] determine a volume of fluid injected from a position of the actuator.

[0077] - The implantable medical device further comprises a sealed housing in which the actuator, reservoir and sensor are arranged, the sealed housing containing a gas.

[0078] - The implantable medical device is an occlusive system in which the inflatable element is an occlusive cuff configured to selectively occlude at least one anatomical duct of said human or animal body taken from among: a urethra, a gastric duct, a colon and a rectum.

[0079] - The inflatable element is elongated in shape and configured for use as a penile implant.

[0080] Finally, a human-machine interface is proposed, configured to communicate with an implantable medical device as described above, and configured to:

[0081] - implement a calibration process for the implantable medical device, so as to obtain a curve comprising a plurality of points, each point being defined by coordinates corresponding to a value of volume of injected fluid and a pressure value estimated by the processing unit of the implantable medical device;

[0082] - display the resulting curve;

[0083] - implement a method for determining the offset as described above, the determined offset being stored in a memory accessible by the processing unit of the implantable medical device.

[0084] DESCRIPTION OF THE FIGURES

[0085] Other features, purposes and advantages of the invention will become apparent from the following description, which is purely illustrative and not limiting, and which should be read in conjunction with the accompanying drawings on which:

[0086] Figure 1 schematically illustrates an implantable occlusive system according to one aspect of the invention. Figure 2 represents curves illustrating the evolution of fluid pressure and injected fluid volume as a function of time, during a calibration process of the implantable occlusive system.

[0087] Figure 3 shows a pressure-volume curve obtained by the calibration process. Figure 4 is a flowchart of the steps of a method for determining the offset according to the invention.

[0088] Figure 5 is a flowchart of steps in a process for determining offset according to a first alternative.

[0089] Figure 6 is a flowchart of steps in a process for determining offset according to a second alternative.

[0090] Figure 7 is a flowchart of steps in a process for determining offset according to a third alternative.

[0091] Figure 8 is a flowchart of steps in a process for determining offset according to another alternative.

[0092] Figure 9 is a flowchart of the steps in a method for correcting a fluid pressure estimated from a sensor measurement, according to another aspect of the invention. Throughout the figures, similar elements are identified by identical reference numerals.

[0093] DETAILED DESCRIPTION OF THE INVENTION

[0094] Implantable medical device

[0095] With reference to Figure 1, the implantable medical device includes an inflatable element 3, configured to be inflated via a fluidic link 2.

[0096] The following describes an implantable medical device, specifically an occlusive system 1, where the inflatable element 3 is an occlusive cuff. However, the method described below can be implemented with other types of implantable medical devices. For example, the inflatable element 3 of the implantable medical device can be elongated and configured for use as a penile implant to treat erectile dysfunction.

[0097] In the illustrated example, the implantable medical device is an occlusive system 1 configured to be implanted in a human or animal body to selectively occlude the anatomical duct via the occlusive cuff 3. The anatomical duct can be one or more ducts, including a urethra, a gastric duct, a colon, and a rectum. Thus, the occlusive system 1 could be, for example, a urinary sphincter, an anal sphincter, or a gastric band.

[0098] The occlusive cuff 3 includes a deformable portion 31 containing a variable volume of fluid. The deformable portion 31 is intended to surround a part of an anatomical duct to be occluded, and to exert pressure on the external walls of the anatomical duct, depending, among other things, on the volume of fluid contained in the deformable portion 31.

[0099] The occlusive system 1 includes a fluid reservoir 5, in fluidic communication with the occlusive sleeve 3 via the fluidic link 2. The fluid reservoir 5 is filled with the fluid, intended to be injected into the occlusive sleeve 3.

[0100] Preferably, the reservoir 5 has a variable volume. Typically, the reservoir 5 has a fixed part and a moving part whose movement varies the volume of the fluid reservoir 5, and therefore the volume of fluid contained in the fluid reservoir 5 and in the occlusive cuff 3. The moving part is, for example, a metal bellows, a piston sliding in a cylinder, or a rolling diaphragm. The bellows has the advantage of ensuring a complete seal of the implant while allowing movement of the movable wall.

[0101] Preferably, for each volume of the reservoir 5, the moving part has a known effective pressure surface, which can be constant or variable depending on the embodiment. The volume of injected fluid can be deduced from the volume of the reservoir 5.

[0102] The occlusive system includes an actuator 6 configured to vary the volume of fluid injected into the occlusive sleeve 3. In an embodiment where the reservoir 5 has a fixed volume, the actuator 6 can be a pump configured to inject a predetermined volume of fluid into the occlusive sleeve 3, or transfer a volume of fluid from the occlusive sleeve 3 to the reservoir 5.

[0103] In the illustrated example, the actuator 6 is mechanically coupled to the moving part of the fluid reservoir 5 so as to move the moving part relative to the fixed part to adjust the volume of the fluid reservoir 5. Typically, the actuator 6 is configured to linearly move the moving part relative to the fixed part in a direction of movement to adjust the volume of the reservoir 5. A change in the volume of the reservoir 5 results in the addition or removal of fluid in the occlusive sleeve 3, thus increasing or decreasing the compression exerted on the anatomical conduit surrounded by the occlusive sleeve 3. The actuator 6 can be chosen from any electromechanical system capable of converting electrical energy into mechanical motion with the power required to allow the movement of the moving part of the fluid reservoir 5 at a required force and speed.Actuator 6 includes, for example, a piezoelectric actuator, an electromagnetic motor, coupled or not to a reducer, an electro-active polymer or a shape memory alloy.

[0104] The occlusive system 1 includes a processing unit 7 configured to control the actuator 6 in order to regulate the fluid pressure in the fluidic circuit, and in particular the pressure exerted by the occlusive cuff 3 on the anatomical conduit. The processing unit 7 typically includes a microprocessor. The processing unit 7 is configured to control the actuator 5 based on a previously calculated fluid pressure in the fluidic circuit.

[0105] In the preferred embodiment described, the actuator 6 can be controlled by a position control of the moving part of the variable-volume reservoir 5. Knowing the effective pressure surface of the moving part at each position of the actuator 6, this therefore corresponds to a volume control of the occlusion system.

[0106] To do this, the occlusive system 1 includes a sensor 4 configured to measure a physical quantity dependent on the actual fluid pressure in the occlusive sleeve 3. As explained previously, this actual fluid pressure varies in particular as a function of a volume V of fluid injected into the occlusive sleeve 3.

[0107] Sensor 4 can be a pressure sensor configured to directly measure the pressure in the occlusive cuff 3. However, the use of such a pressure sensor may pose problems of integration, size, sealing and / or biocompatibility of sensor 4.

[0108] Preferably, the sensor 4 is a force sensor configured to measure a force dependent on the actual fluid pressure in the occlusive sleeve 3. The processing unit 7 can estimate the fluid pressure P from the force measurement of the sensor 4. An example of such a sensor 4 is proposed in WO 2016 / 083428 A1. The sensor 4 is mechanically linked to the actuator 6 and / or to the moving part of the variable-volume fluid reservoir 5. The sensor 4 is configured to measure a compressive and / or tensile force in the direction of movement of the moving part of the fluid reservoir 5. In this embodiment, the sensor 4 may include, for example, one or more strain gauges, one or more FSR™ (Force Sensing Resistor) type sensors, or one or more pressure sensors coupled to a mechanism for measuring a force.For example, one can cite a hydraulic pressure sensor combined with a fluid-filled pouch arranged to measure pressure on a predetermined surface, thus allowing the force applied to the measurement surface to be deduced.

[0109] The processing unit 7 takes into account the force(s) measured by the sensor 4 to indirectly determine the fluid pressure in the fluid circuit. From the determined fluid pressure and a relationship linking the volume of fluid injected into the fluid circuit and the fluid pressure, the processing unit 7 can control the movement of the actuator 6 to inject the appropriate volume of fluid to reach a target pressure.

[0110] The relationship between the injected fluid volume and the fluid pressure in the occlusive sleeve 3 is stored in memory. This memory can also store a relationship between the position of the actuator 6 and the injected fluid volume V.

[0111] The occlusal system 1 may include memory accessible by the processing unit 7. The processing unit 7 may also incorporate its own memory. Alternatively, the processing unit 7 communicates wirelessly with the memory of a remote device.

[0112] Preferably, the occlusive system 1 comprises a housing 10. The actuator 6, the reservoir 5 and the sensor 4 are arranged inside the housing 10. The housing 10 is sealed and contains a gas.

[0113] In the illustrated embodiment, the fixed part of the reservoir 5 is attached to the housing 10. Thus, a change in the volume of the reservoir 5 causes the internal volume of the housing 10 to change, and consequently the pressure of the gas contained in the housing 10. Preferably, the housing 10 includes a gas pressure sensor (not shown) configured to measure the pressure of the gas contained in the housing 10.

[0114] Preferably, the processing unit 7 is configured to determine the fluid pressure P from other mechanical, physical, and dimensional parameters of the occlusive system 1. For example, the mechanical parameters may include the stiffness of the moving part of the tank 5, or an effective pressure area of ​​the moving part of the tank 5. The processing unit 7 can also be configured to determine the fluid pressure P using the gas pressure measurement from the gas pressure sensor.

[0115] The mechanical parameters can be provided by the manufacturer or estimated during a preliminary calibration step. If the effective pressure surface varies with the position of the actuator 6, the memory can store a table representing the relative variations of these parameters.

[0116] Preferably, the occlusive system 1 may include an external pressure sensor. The external pressure sensor is configured to measure the atmospheric pressure in the user's environment. The external pressure sensor may include, for example, a barometric sensor capable of measuring the current atmospheric pressure exerted on the user's body and may be arranged in an external device worn by the user. The external pressure sensor may communicate wirelessly with a communication module of the processing unit 7.

[0117] Calibration process

[0118] In order to obtain an accurate estimation of the fluid pressure in the fluidic circuit, and therefore correct regulation of the occlusive system 1, the sensor 4 must be calibrated beforehand. During calibration, the processing unit 7 determines a relationship between the pressure in the fluidic circuit and the volume of fluid injected into the occlusive sleeve 3, under specific user conditions, i.e., when the user is stationary and in a specific position (lying down or standing, for example).

[0119] For example, in the case of an artificial urinary sphincter, calibration can be performed a few minutes after urination, when the user is standing and relatively still. Calibration can be performed at a defined frequency, for example, once a week, or during a doctor's appointment, via the wireless communication module of the occlusive system 1.

[0120] During calibration, the processing unit 7 commands the actuator 6 to gradually increase the pressure for predetermined injection volumes, and receives the force values ​​from the sensor 4.

[0121] Figure 2 illustrates the variation of the volume injected into the occlusive sleeve 3 and the fluid pressure P in the fluidic circuit as a function of time t, during a pressure increase in the fluidic circuit performed during calibration. Typically, the processing unit 7 controls the actuator 6 to vary the injected volume between a minimum injected volume V m in and a maximum injected volume V max, incrementally. The volume difference dV between each increment corresponds, for example, to the resolution of actuator 6 or is set by the operator performing the calibration. The volume range Vmax-Vmin depends on the physical parameters of the reservoir 5, in particular the dimensions of the moving part and the volume of the reservoir 5. With each increment of injected volume, the pressure in the fluidic circuit increases. In the illustrated embodiment, sensor 4 measures the force in newtons. Processing unit 7 converts the force into a pressure expressed in cmH2O. Generally, sensor 4 provides a pressure estimate with an accuracy of + / - 15 cmH2O, preferably + / - 10 cmH2O. The accuracy of the pressure estimate by processing unit 7 can vary depending on the accuracy of sensor 4. The accuracy of sensor 4 depends, for example, on the technology used and the battery level of sensor 4.

[0122] The accuracy of the pressure estimation by the processing unit 7 can vary depending on the accuracy of the parameter values ​​used for the conversion. For example, the stiffness constant of the bellows forming the moving part of the sensor 4, used for the conversion, can vary depending on the position of the sensor 4 and the patient's posture.

[0123] The processing unit 7 stores the determined fluid pressure values ​​in a table located in memory. It should be noted that the relationship between the fluid pressure P in the fluid circuit and the volume of fluid V injected or withdrawn may exhibit hysteresis, i.e., an asymmetry in behavior between fluid injection and withdrawal. Consequently, the curve defined by the plurality of points representing the evolution of the fluid pressure P associated with a pressure increase, when the injected fluid volume V changes from a first volume Vi to a second volume V?, greater than Vi, may differ from the curve defined by the plurality of points representing the evolution of the pressure associated with a pressure decrease, when the fluid volume changes from the second volume V? to the first volume Vi. It should be noted that the injected fluid volume V varies inversely proportionally to the variable volume of the reservoir 5.

[0124] Calibration can take into account any hysteresis that might exist in the fluid circuit. Preferably, the table includes values ​​for the volume to be injected to reach a given pressure and values ​​for the volume to be withdrawn to reach a given pressure. The values ​​in the table can be represented graphically as a plurality of points defining a P / V curve. Typically, the resulting curve has an S-shape, as illustrated in Figure 3; that is, it includes an inflection point for a volume V* of injected fluid. By "inflection point," we mean a point in the set of points corresponding to a region of minimum or zero slope.This zone corresponds to the transition between a pressurization phase of the occlusive cuff 3, in which the actual pressure inside the cuff is positive, and a depressurization phase of the occlusive cuff 3, in which the actual pressure inside the cuff is negative. This type of curve is characteristic of curves representing the evolution of stress as a function of strain for an elastically deformable material such as the material used for the occlusive cuff 3. The occlusive cuff 3 is made, for example, of rubber or silicone. Materials exhibiting an S-shaped stress-strain curve are particularly susceptible to elastic instabilities.

[0125] The inflection point of the curve indicates the transition of the occlusive sleeve 3 from a depressurized to a pressurized state, that is, when the pressure changes from negative to positive values. Therefore, the inflection point on the P / V curve obtained during the calibration process should correspond to a true zero pressure, Po = 0 cmH2O.

[0126] Human-machine interface

[0127] The calibration procedure can be implemented by an operator using a human-machine interface configured to communicate with an implantable occlusal system 1 described previously. The human-machine interface typically includes

[0128] a processing module, for example a processor, configured to perform calculations,

[0129] a communication module with the processing unit 7 of the occlusive system, and / or directly with the sensor 4;

[0130] a screen, to display data received by the processing unit 7 and / or the sensor 4; and

[0131] a memory, configured to save data received by the communication module or calculated by the processing module.

[0132] The human-machine interface enables the implementation of the calibration process for the implantable occlusive system described previously, in order to obtain the P / V curve. The plurality of points on the P / V curve are defined by their coordinates corresponding to a value of the injected fluid volume, typically determined by the processing unit 7 from the position of the actuator 6, and to a pressure value estimated by the processing unit 7, from the measurement of a physical quantity representative of the actual pressure, by the sensor 4.

[0133] The human-machine interface can be integrated into a remote control available to a practitioner for secure communication with the occlusal system during implantation and / or during subsequent patient consultations. This remote control can be used, in particular, to retrieve operating data from the occlusal system, stored in implanted memory. The remote control can also be used to modify the occlusal system's configuration.

[0134] Observation of a pressure measurement discrepancy

[0135] It has been observed that the relationship between the estimated fluid pressure P in the fluidic circuit and the volume V of fluid injected into the occlusive sleeve 3 exhibits a drift or lag that varies over time. This lag may be related to variations in the mechanical parameters of the components of the occlusive system 1, particularly the stiffness of the moving part of the reservoir 5, or the pressure of the gas contained in the housing 10, in the described embodiment. The sensor 4 may also exhibit drift due to degradation of its components.

[0136] Furthermore, the conversion formula used by the processing unit 7 to estimate the fluid pressure P from the measurements of sensor 4 is derived for specific environmental conditions. In particular, it is suitable for a reference temperature specific to the use of the occlusive system, typically a reference temperature of 37 °C. Thus, the fluid pressure P determined by the processing unit 7 from the force measurement of sensor 4 includes an additional error related to the operating temperature of sensor 4. This temperature can vary by approximately + / - 3 °C, depending on the ambient temperature of the subject in whom the occlusive system 1 is implanted, and the subject's health, particularly in the case of fever.It is estimated that such a difference with the reference temperature induces a shift of 5 cmH20 / °C, between the fluid pressure P estimated by the processing unit 7 via the sensor 4, and the actual fluid pressure in the occlusive sleeve 3.

[0137] Since occlusive system 1 does not include a temperature sensor, it is not possible to directly correct this additional error by taking into account a correlation between the temperature of sensor 4 and the pressure variation.

[0138] Finally, a discrepancy between the fluid pressure P estimated from the measurements of sensor 4 and the actual fluid pressure in the occlusive cuff 3 may be due to a difference in height between sensor 4 and the processing unit 7 arranged in the housing 10, which is typically implanted in the subject's abdomen, and the anatomical duct around which the occlusive cuff 3 is placed, for example, the urethra. This is because a column of water forms between the reservoir of the medical device, in which sensor 4 is placed, and the occlusive cuff 3, particularly if the subject is an adult. The height of this water column is approximately 20 cm when the subject is standing. The human-machine interface screen can display the resulting P / V curve, allowing the operator to visualize it.A doctor can thus visually estimate an apparent shift from the displayed P / V curve, and quickly identify a dysfunction of the implantable occlusal system, especially if the apparent shift is high.

[0139] The purpose of the process described below is to determine the offset induced by all these factors in order to obtain a better estimate of the actual fluid pressure in the fluidic circuit, and thus to allow better control of actuator 6.

[0140] Based on the above, it is necessary to determine the pressure value corresponding to the inflection point of the P / V curve obtained in order to estimate the desired offset between the estimated fluid pressure and the actual fluid pressure in the occlusive sleeve 3. For example, if the inflection point is at 20 cmH20, whereas in theory the inflection point should be 0 cmH20, then the offset of the sensor 4 is +20 cmH20.

[0141] Method for determining the difference between the estimated fluid pressure and the actual fluid pressure

[0142] As explained previously, during the calibration process, data representing the evolution of the estimated fluid pressure in the fluidic circuit as a function of the injected fluid volume are collected.

[0143] An embodiment of a method for determining the difference between the estimated fluid pressure and the actual fluid pressure is described with reference to Figure 4. During an initial step S0, a plurality of points Xi representing the evolution of the estimated fluid pressure P, in cmH2O, as a function of the injected fluid volume V, in ml, are obtained. The points Xi can be ordered by increasing volume, so that a P / V curve is formed by the plurality of points Xi. The P / V curve is hereafter considered to comprise a number of points N, and the index i is an integer between 1 and N. Each point Xi = (Vi, Pi) corresponds to a value Vi of the injected fluid volume, calculated from the actuator displacement setpoint 6, associated with a corresponding value Pi. The value Pi represents the actual fluid pressure in the occlusive sleeve 3.It can be the fluid pressure value estimated from measurements by sensor 4, or more generally a quantity that changes like pressure. As explained previously, this means that the fluid pressure Pi can be measured directly by sensor 4, or estimated by the processing unit 7 from a measurement of a physical quantity dependent on the actual fluid pressure, typically a force exerted on the moving part of the tank, by sensor 4. The coordinates of the plurality of points Xi forming the P / V curve can be stored as a table in the memory of the processing unit 7.

[0144] The P / V curve can represent the evolution of fluid pressure and injected fluid volume during a pressure increase or decrease. As explained previously, the corresponding P / V curves may differ due to hysteresis.

[0145] As explained previously, the desired shift corresponds to the shift in the representative value of the fluid pressure at the inflection point of the P / V curve. During a step S1, the processing unit 7 determines the inflection point of the P / V curve, from the plurality of points obtained.

[0146] During an S2 step, the processing unit 7 determines the offset between the estimated fluid pressure and the actual fluid pressure, the offset being the estimated fluid pressure value P* of the determined inflection point.

[0147] Several methods can be implemented by processing unit 7 to determine the inflection point of the P / V curve. The choice of method for determining the inflection point of the P / V curve, from among the methods described below, may depend on the offset determined. Typically, if the offset determined during step S2 using a first method has an outlier value, for example, greater than 50 cmH2O, processing unit 7 can be configured to implement an alternative method to determine the offset using a different procedure.

[0148] Polynomial approximation of the P / V curve

[0149] In certain situations where the P / V curve formed by the plurality of points does not clearly show an inflection point separating the overpressure and depression phases, it is preferable to determine the inflection point using an approximate P / V curve as described with reference to Figure 5. Preferably, the processing unit 7 obtains, during an approximation step S1'-A, a plurality of approximated points forming an approximate curve of the P / V curve by a polynomial regression.

[0150] The plurality of approximated points are defined by coordinates corresponding to the injected fluid volume and an approximate pressure value, the approximate pressure value satisfying Si" = J'G' with P a polynomial of order n. In other words, the processing unit 7 determines an nth-order polynomial modeling the P / V curve generated from the points obtained in step S0. Typically, n > 2 is chosen. For example, the processing unit 7 implements a polynomial regression of order n = 4 to determine a 4th-order polynomial modeling the evolution of the representative fluid pressure value P as a function of the injected fluid volume V. The processing unit 7 thus obtains the coefficients of the polynomial P such that, for a point of index i, the approximate pressure value satisfies:

[0151]

[0152] "The coefficients a0, a4, a2, a3, a4 can be determined by the least squares method.

[0153] Advantageously, the approximate pressure value S"; can be obtained for all points of the P / V curve, including the first point with index 1 and the last point with index N.

[0154] During a step S1 '-B, the processing unit7 calculates, for each point of the plurality of points (and therefore for each approximated point, the abscissas being identical), a distance di between the approximate pressure value of the point and the approximate pressure value of a neighboring point.

[0155] The distance di can be a Euclidean distance calculated from the approximate pressure values ​​associated with the previous and next points on the curve. For example, each distance is calculated for an approximate point of index i as a distance between the approximate pressure value S" of the point of index i and the approximate pressure value S". i-1from a previous point of index i-1, and between the approximate pressure value S"i of the point of index i and the approximate pressure value S" i+1 of a subsequent point with index i+1, according to the formula

[0156]

[0157] " "

[0158] During step S1'-C, the processing unit 7 selects a point for which the calculated distance di is minimal. The determined inflection point can be this selected point. It corresponds to the point where the pressure changes from low to high. The offset determined in step S2 can be the approximate pressure value for this point, obtained in step S1'-A, or the representative fluid pressure value for the point obtained in step S0.

[0159] Calculation of Slope Values ​​By definition, the inflection point corresponds to a point of minimum (or zero) slope on a curve. Alternatively or complementaryly, the determination of the inflection point during step S1 may include the following steps, with reference to Figure 6. During a first step S1-A, the processing unit 7 calculates a plurality of slope values ​​Si. Each slope value is calculated for a point on the P / V curve with index i, that is, at two coordinates associating a representative pressure value Pi and a volume value Vi of injected fluid.

[0160] The slope value Si is calculated from the coordinates of the point with index i and the coordinates of at least one neighboring point on the P / V curve. Generally, a neighboring point is understood to be the previous point with index i-1, or the following point with index i+1.

[0161] In an implementation where the resulting P / V curve comprises a very large number of points N, for example N > 100, with very small volume differences between two successive points on the P / V curve, the definition of a "neighbor" point can be extended to a point whose injected fluid value Vj is sufficiently close to the injected fluid value Vi of the point under consideration. For example, a neighborhood threshold e is defined. v and we consider that a point Xj is close to the point Xi if the relation

[0162]

[0163] < e v The proximity threshold is preferably less than 0.1 ml, for example equal to 0.05 ml.

[0164] Several methods can be used to calculate the slope value associated with a point on the P / V curve.

[0165] For example, the slope value associated with the point Xi = (Vi, Pi) can be calculated from the coordinates of the point with index i and the coordinates of the next point with index i+1 using the following formula:

[0166]

[0167] According to another example, the slope value associated with the point Xi = (Vi, Pi) can be calculated from the coordinates of the point with index i and the coordinates of the previous point with index i-1 using the following formula: St = Vl Vl ~ 1 .

[0168] Pi-Pi-l

[0169] The slope value associated with the point Xi = (Vi,Pi) can be equal to the average between the slope values ​​of the two previous examples, according to the following formula:

[0170]

[0171] Preferably, each slope value is calculated from the injected pressure value and volume value, the pressure and volume values ​​associated with the previous point on the P / V curve, and the pressure and volume values ​​associated with the next point on the P / V curve.

[0172] Typically, processing unit 7 performs a linear regression to determine the slope of the line passing closest to the three points formed by the previous point with index i-1, the point with index i, and the next point with index i+1.

[0173] The slope value associated with point Xi = (Vi, Pi) can be calculated using the following formula:

[0174]

[0175] with V the average of the values ​​of injected fluid volume of the points considered, and P the average of the fluid pressure values ​​of the points considered.

[0176] In this embodiment, the slope value St cannot be calculated for the first point of the P / V curve (index 1) and for the last point of the P / V curve (index N). More generally, the slope value St can be calculated using the set of neighboring points according to the aforementioned proximity criterion:

[0177]

[0178] <

[0179] Preferably, during a preliminary step, points on the P / V curve where the corresponding pressure value Pi is below a threshold value are removed. Typically, the threshold value is -200 cmH2O. This ensures that points on the P / V curve that are too far from the desired inflection point are not considered.

[0180] To avoid identifying an injection point at each break in slope due to a pressure variation caused by a measurement error, it is preferable not to use slope values ​​directly.

[0181]

[0182] calculated in step S1 to identify the injection point.

[0183] During step S1-B, the processing unit 7 can select a point from the plurality of points for which the slope value is minimal. The determined inflection point can be the selected point, or chosen from the point selected in step S1-B and the point selected in step S1'-C described previously. This improves the determination of the inflection point, particularly in the case where the inflection point of the P / V curve does not correspond to a point in the plurality of points, but to an intermediate value (the "overshoot" phenomenon).

[0184] Preferably, processing unit 7 can also calculate an approximation of the derivative of the slope values ​​S L that is, a second derivative of the P / V curve for the points obtained in step S0.

[0185] In an alternative or complementary step to S1-B, the processing unit 7 can select the point for which the second derivative value is closest to zero, that is, a value for which the second derivative is closest to zero. The shift can be obtained by averaging the pressure values ​​associated with the two points between which the second derivative is closest to zero.

[0186] Slope filtering

[0187] Preferably, the inflection point is determined during step S1 from filtered slope values.

[0188] More specifically, during a filtering step S1-C, the processing unit 7 calculates a plurality of filtered slope values ​​S-. Each filtered slope value S- is calculated for a point on the P / V curve, from the slope value calculated for the point and at least one slope value calculated for a neighboring point on the P / V curve.

[0189] For example, in the case where the slope value associated with the point Xi = (Vi, Pi) is calculated from the coordinates of the point with index i and the coordinates of the next point with index i+1 (example 1), the filtered slope value S- can be calculated from the slope value calculated for the point with index i and the slope value calculated for the next point with index i+1, according to the following formula:

[0190]

[0191] Similarly, in the case where the slope value associated with the point Xi = (Vi,Pi) is calculated from the coordinates of the point with index i and the coordinates of the previous point with index i-1 (example 2), the filtered slope value S- can be calculated from the slope value calculated for the point with index i and the slope value calculated for the previous point with index i-1, according to the following formula: S-

[0192]

[0193] Preferably, the filtered slope value S- is calculated from the slope value and the slope values ​​associated with the previous and next point of the curve.

[0194] In one embodiment, the filtered slope value is calculated as an average between the slope value associated with the point on the curve (Pi, Vi), the distance between the slope value S i-1of the previous point and the slope value of the point in question, and the distance between the slope value of the following point S i+1 and the slope value of the point considered. Thus, the filtered slope value S- is calculated using the following formula:

[0195]

[0196] Note that the filtered slope value

[0197]

[0198] can be obtained as the sum of the three terms above, without it being necessary to divide the sum by a constant (above equal to 3).

[0199] These different methods of calculating the filtered slope value allow for penalizing a slope value S t at a point when the differences in slope value compared to the preceding and following points are significant. Indeed, this may indicate an outlier, for example a local minimum.

[0200] More generally, the two differences can be weighted by positive coefficients a, B, so as to penalize more or less the difference in slope with the next point or with the previous point:

[0201]

[0202] In this embodiment, the filtered slope value S- cannot be calculated by the previous formula for the first two points of the P / V curve with indices 1 and 2, and for the last two points of the P / V curve with indices N and N-1.

[0203] As before, the previous examples can be adapted to the neighborhood under consideration, the filtered slope value being able to be calculated from the slope values ​​of all points neighboring the point with index i.

[0204] Other filtering methods are possible. For example, the filtered slope value calculated for point i can be calculated from the filtered slope value calculated for the previous point i-1, and the slope value of point i can be calculated using the formula:

[0205]

[0206] with a = —, a smoothing constant, with n an integer chosen according to the desired precision.

[0207] According to another example, the filtered slope value calculated for the point with index i can be calculated from the filtered slope value calculated for the previous point with index i-1, and the slope value calculated for the previous point with index i-1, according to the formula:

[0208]

[0209] This example corresponds to a first-order low-pass filtering, with a cutoff frequency of |.

[0210] By convention, the filtered slope values ​​SJ or SQ that cannot be calculated according to the above formulas can be chosen arbitrarily, for example equal to the average of the plurality of slope values ​​calculated in step S1.

[0211] During an S1-D step, the processing unit 7 can select a point from the plurality of points for which the filtered slope value is minimal. The determined inflection point can be the selected point, or chosen from the point selected in step S1-D and the point selected in step S1'-C described previously.

[0212] Preferably, the processing unit 7 thus determines the shift of the fluid pressure measurement sought in step S2, by first determining the minimum filtered slope value S r* For example, processing unit 7 implements a minimum search algorithm by iteratively bounding the plurality of calculated filtered slope values. Alternatively, processing unit 7 implements a sequential search algorithm or a binary search algorithm.

[0213] Preferably, in a case where the plurality of filtered slope values ​​includes two local minima, the processing unit 7 selects the minimum filtered slope value associated with the point whose injected volume value is closest to the average of the injected volumes across the plurality of points. Indeed, since the P / V curve exhibits a certain symmetry around the inflection point, it can reasonably be assumed that the volume V* of fluid injected at the inflection point is close to the average.

[0214] Thus, the difference between the estimated fluid pressure and the actual fluid pressure corresponds to the pressure value P* at point X* for which the filtered slope value is the minimum filtered slope value S r *. The same minimum search algorithms among the plurality of points can be applied for the selection at steps S1'-C and S1-B.

[0215] This method thus allows for a more precise determination of the offset, compared to simply finding the inflection point by determining the minimum slope value. Combining filtering and approximation: As explained previously, it may be preferable in some cases to refine the choice of the inflection point from among the points on the P / V curve using a polynomial approximation. The resulting determination method is illustrated in Figure 7.

[0216] Preferably, the processing unit 7 obtains, during the S1'-A approximation step, the curve approximated from the P / V curve by a polynomial regression.

[0217] Processing unit 7 then calculates, during step S3, a plurality of values ​​for an indicator. Each indicator value is calculated for an approximate point, based on the coordinates of the approximate point and the coordinates of a neighboring point. It should be noted that the approximate curve also includes a number N of points whose abscissas, corresponding to the volume values ​​Vi of injected fluid, are identical to the abscissas of the P / V curve determined from the points obtained in step S0.

[0218] The choice of the relevant indicator may vary. Here, the indicator does not represent the evolution of slope values, but rather the quality of the polynomial approximation performed, and therefore the similarity between the approximate curve obtained in step S1'-A and the initial P / V curve. Typically, processing unit 7 calculates a distance from the approximate pressure values ​​obtained by the polynomial regression, as in step S1'-B.

[0219] Preferably, the E indicator t can be calculated as a distance between the approximate pressure value S"t of the point and the approximate pressure value of a neighboring point, and between the volume value V[ of injected fluid from the point and the volume value of injected fluid from the neighboring point.

[0220] For example, in the case where the slope value associated with the point Xi = (Vi, Pi) is calculated from the coordinates of the point with index i and the coordinates of the next point with index i+1 (example 1), the neighboring point can be chosen as the next point with index i+1, and the indicator E t calculated according to the following formula:

[0221]

[0222] Similarly, in the case where the slope value associated with the point Xi = (Vi,Pi) is calculated from the coordinates of the point with index i and the coordinates of the previous point with index i-1 (example 2), the neighboring point can be the previous point with index i-1, and the indicator E t calculated according to the following formula:

[0223]

[0224] As before, several points neighboring the point with index i can be taken into account in the calculation of the value of the indicator E tfrom the index point i. Preferably, the distance is a Euclidean distance calculated from the approximate pressure values ​​associated with the previous and next points on the curve. For example, each value of the indicator E t is calculated for a point on the curve of index i as a distance between the approximate pressure value S"t of the point of index i and the approximate pressure value S" i-1 from a previous point with index i-1, between the value of the injected fluid volume V[ of the point with index i and the value of the injected fluid volume V i-1 from the previous point with index i-1, between the approximate pressure value S"t of the point with index i and the approximate pressure value S" i+1 from a subsequent point with index i+1, and between the value of the injected fluid volume V[ of the point with index i and the value of the injected fluid volume V i+1 of the next point with index i+1, according to the formula

[0225]

[0226] " "

[0227] Similarly, the slope value S t , the value of indicator E t cannot be calculated for the first point of the P / V curve (index 1) and for the last point of the P / V curve (index N), according to the previous formula.

[0228] The processing unit 7 then calculates, during a step S4, a plurality of average values ​​M; . Each average value

[0229]

[0230] is calculated for a point on the P / V curve as the average between the filtered slope value calculated for the point and the indicator value calculated for the approximate point of the same index. In other words, the average value Mi is calculated for any point of index i as

[0231]

[0232] In this embodiment, the processing unit 7 then searches for a minimum M* of the plurality of average values

[0233]

[0234] Determining the difference between the estimated and actual pressure involves selecting the point X* on the P / V curve where the average value is the minimum M*. The inflection point can be chosen as the selected point X*. The resulting difference can then be updated and set as the pressure value at the selected inflection point X*.

[0235] As before, the average value

[0236]

[0237] is not necessarily calculated for all points on the P / V curve; typically, it is not calculated for the first point with index 1 and the last point with index N, for which the value of the indicator E t cannot be calculated.

[0238] Preferably, with reference to Figure 8, the determination procedure further includes normalization steps for the filtered slope values ​​S- and the indicator values ​​E t This allows, during the calculation of average values ​​in step S4, the averaging of two quantities between 0 and 1.

[0239] During a first normalization step S1-E, the processing unit 7 normalizes the plurality of filtered slope values ​​S-, so as to obtain a plurality of normalized slope values ​​NS-.

[0240] Typically, processing unit 7 calculates the normalized slope value NS- at index point i as the ratio between the difference between the filtered slope value S- at index point i and the minimum filtered slope value among the plurality of filtered slope values ​​calculated in step S1-C, and the difference between the maximum filtered slope value among the plurality of filtered slope values ​​calculated in step S1-C, and the minimum filtered slope value of the P / V curve. Thus, the following calculation yields the normalized slope value NE t :

[0241] >

[0242]

[0243] Note that the indices of the points among which the minimum and maximum of the filtered slope values ​​are sought depend on the calculation method chosen.

[0244] During a second normalization step S3', the processing unit 7 normalizes the plurality of indicator values ​​Et, so as to obtain a plurality of normalized indicator values ​​NS-.

[0245] Typically, the processing unit calculates the normalized indicator value at index point i, using a calculation similar to that performed during the first normalization step S1-E. In other words, the normalized indicator value NE t is calculated for a point of index i as the ratio between the difference or gap between the indicator value E t calculated for the index point i and a minimum indicator value among the plurality of indicator values ​​calculated in step S3, and a difference between a maximum indicator value among the plurality of indicator values ​​calculated in step S3 and the minimum indicator value, according to the formula:

[0246]

[0247] >

[0248] The normalization steps S1-E and S3' can be implemented independently, simultaneously, or sequentially. In this embodiment, the normalization steps S1-E and S3' take place before the averaging step S4.

[0249] Thus, during a step S4', the processing unit 7 calculates, for the point with index i, a normalized average value from the normalized indicator value calculated for the point during step S3', and the normalized (filtered) slope value calculated for the point with index i during step S1-E, according to the formula:

[0250]

[0251] Step S4' replaces step S4 in this embodiment. Finally, during step S2 for determining the offset, the processing unit 7 then searches for a minimum NM* of the plurality of normalized mean values ​​NM tDetermining the offset involves selecting the point X* on the P / V curve for which the average value is the minimum NM*. The offset can then be updated and chosen as the pressure value of the selected point X*.

[0252] In the preferred embodiment described above, at the end of the shift determination process, each point Xi on the P / V curve is associated with:

[0253] to a fluid pressure value Pi, to a volume value Vi of injected fluid, and to an approximate pressure value Si”, for i between 1 and N;

[0254] to a slope value Si, to an indicator value Ei, and to a normalized indicator value NEi, for i between 2 and N-1; and

[0255] to a filtered slope value Si', to a normalized slope value NSi', and to an average value Mi, for i between 3 and N-2.

[0256] The values ​​can be stored in the memory of processing unit 7, so that they can be reused later. For example, during a new calibration procedure, new data can be collected, enriching the set of stored data.

[0257] In the preceding description, the offset determination process is implemented by the processing unit 7 of the occlusal system 1. Preferably, the determination process is also implemented by an external computer, using data collected during calibration, typically by a surgeon during a follow-up appointment for the patient in whom the device is implanted. The external computer preferably communicates wirelessly with the processing unit and / or directly with the sensor 4. The external computer can transmit the offset determined at the end of the determination process to the processing unit 7. The processing unit 7 can then store the determined offset in its memory and use it to subsequently correct the fluid pressure estimation.

[0258] The operator can, via the human-machine interface processing module, implement the offset determination process described above. The determined offset can be stored in a memory accessible by the processing unit 7 of the implantable occlusal system.

[0259] Method for correcting an estimated fluid pressure

[0260] Indeed, the implementation of the shift determination process described above can advantageously allow the processing unit 7 of the occlusive system 1 to subsequently correct the estimated fluid pressure value P.

[0261] More specifically, when the occlusive system 1 is implanted in the subject's body, the sensor 4 regularly takes measurements, in order to allow the processing unit 7 to control the actuator 6 in order to regulate the pressure exerted by the occlusive cuff 3 on the anatomical conduit.

[0262] A method for correcting a fluid pressure value estimated from the measurement of sensor 4 includes the following steps, with reference to Figure 9.

[0263] In the first step C1, the processing unit 7 estimates a fluid pressure value from a measurement of a physical quantity by the sensor. As explained previously, the processing unit 7 can, for example, use the conversion formula stored in memory.

[0264] In a second step C2, the processing unit 7 determines the difference between an estimated fluid pressure and an actual fluid pressure by implementing the offset determination process described previously. Preferably, the offset is stored in the memory of the occlusive system 1, so that the processing unit 7 does not have to repeat all the steps of the offset determination process for each new measurement from the sensor 4.

[0265] The determined offset can be updated at each new calibration of the occlusive system 1 by an operator, for example on a weekly basis.

[0266] In a third step, C3, processing unit 7 obtains a corrected fluid pressure value. To do this, processing unit 7 calculates the difference between the estimated fluid pressure value and the determined offset. For example, if the estimated fluid pressure is 10 cmH2O and the determined offset is 5 cmH2O, this means the estimate is offset by 5 cmH2O. The corrected value is therefore 10 - 5 = 5 cmH2O.

[0267] The corrected fluid pressure value can then be used to control actuator 6 according to a volume-injected regulation as described in WO 2016 / 083428 A1. In other words, the processing unit 7 is configured to determine the volume of fluid to be transferred between the reservoir and the occlusive cuff 3, based on the corrected fluid pressure value, and to control actuator 6 to transfer the determined volume of fluid. This allows the correct pressure to be applied to the anatomical conduit, and thus the anatomical conduit to be properly occluded without damaging the tissues.

[0268] The correction process can be implemented automatically by a computer. Alternatively, a computer program is proposed, comprising instructions which, when executed by a computer—that is, a dedicated computer—cause the computer to implement the correction process described above. The instructions can be stored on a computer-readable storage medium.

Claims

DEMANDS 1. Method for determining a difference between an estimated fluid pressure and an actual fluid pressure in an inflatable element (3) of an implantable medical device (1), an estimated fluid pressure value (P) being calculated from a measurement of a sensor (4), the sensor (4) being configured to measure a physical quantity dependent on the actual fluid pressure, the determination method comprising steps of: obtaining a plurality of points, each point being defined by coordinates corresponding to a value of injected fluid volume (V) and a representative value of the estimated fluid pressure (P), the plurality of points defining a curve (P / V); determination of an inflection point of the curve (P / V), from the plurality of points; determination of the offset between the estimated fluid pressure and the actual fluid pressure, the offset being the estimated fluid pressure value (P*) of the determined inflection point.

2. A method for determining the inflection point according to claim 1, wherein the determination of the inflection point comprises: obtaining a plurality of approximated points from the plurality of points by a polynomial regression, the plurality of approximated points being defined by coordinates corresponding to the value of the volume of injected fluid and to an approximate value of pressure, the approximate value of pressure satisfying Si" = J' 'i), with P a polynomial of order n > 2; Calculation, for each point in the plurality of points, of a distance between the approximate pressure value of the point and the approximate pressure value of a neighboring point selection of a point from the plurality of points for which the calculated distance is minimal.

3. A method for determining the inflection point according to claim 1 or 2, wherein the determination of the inflection point comprises: calculation of a plurality of slope values, each slope value being calculated for a point on the curve (P / V), from the coordinates of the point and the coordinates of at least one neighboring point on the curve, selection of a point from the plurality of points for which the slope value is minimal.

4. Method according to claim 3, wherein the at least one neighboring point comprises a next point and a previous point, such that each slope value is calculated for a point of the curve (P / V) of index i, from the coordinates of the point of index i, the coordinates of the previous point of the curve of index i-1, and the coordinates of the next point of the curve of index i+1.

5. A method for determining the inflection point according to claim 3 or 4, wherein the determination of the inflection point further comprises: calculation of a plurality of filtered slope values, each filtered slope value being calculated for a point on the curve (P / V), from the slope value calculated for the point and at least one slope value calculated for a neighboring point on the curve (P / V), selection of a point from the plurality of points, for which the filtered slope value is minimal.

6. Method according to claim 5, wherein each filtered slope value is calculated for a point on the curve (P / V), from the slope value calculated for the point and a distance between the slope value calculated for the point and a slope value calculated for a nearby point on the curve.

7. A method according to claim 6, wherein at least one neighboring point comprises a following point and a preceding point, such that each filtered slope value (S-) is calculated for a point on the (P / V) curve of index i, by the formula: S- = with Si the slope value calculated for the point with index i, S i-1 there slope value calculated for the previous point on the curve with index i-1, etS i+1 the slope value calculated for the next point on the curve with index i+1.

8. A method according to any one of claims 5 to 7, further comprising the steps of: obtaining a plurality of approximated points from the plurality of points by a polynomial regression, the plurality of approximated points being defined by coordinates corresponding to the value of the volume of injected fluid and to an approximate value of pressure, the approximate value of pressure satisfying Si" = P(VÎ), with P a polynomial of order n > 2; calculation, for each approximated point, of a value of an indicator (E^), from the coordinates of the approximated point, and the coordinates of a neighboring point; calculation of a plurality of average values, each average value being calculated for a point of the plurality of points as the average between the filtered slope value calculated for the point and the indicator value calculated for the approximated point of the same index; in which the determination of the inflection point further includes the selection of a point for which the average value is minimal.

9. A method according to claim 8, wherein each value of the indicator E t is calculated for an approximate point of index i as a distance between the approximate pressure value S"t of the point with index i and the approximate pressure value S" i-1 from a previous point of index i-1, between the value of the injected fluid volume V[ of the point with index i and the value of the injected fluid volume V i-1 from the previous point of index i-1, between the approximate pressure value S"t of the point with index i and the approximate pressure value S" i+1 from a subsequent point with index i+1, and between the value of the injected fluid volume V[ of the point with index i and the value of the injected fluid volume V i+1 of the next point with index i+1, according to the formula " " 10. A method according to any one of claims 8 and 9, further comprising steps of: normalizing the plurality of filtered slope values, so as to obtain a plurality of normalized slope values normalization of the plurality of values ​​of the indicator (EJ), so as to obtain a plurality of normalized indicator values ​​NEi); and in which each average value is calculated for a point on the curve (P / V) as the average between the normalized filtered slope value calculated for the point and the normalized indicator value calculated for the point.

11. A method according to claim 10, wherein the normalized slope value NS- is calculated for a point of index i, the index i being an integer between 1 and N, N being the number of points defining the curve (P / V), as a ratio between a difference between the slope value S- calculated for the point with index i and a minimum slope value among the plurality of slope values ​​calculated for the points, and a difference between a maximum slope value among the plurality of slope values ​​calculated for the points and the minimum slope value according to the formula: > minimum slope calculated for points with index j including the maximum slope value calculated for points index j between 3 and N-2.

12. A method according to any one of claims 10 and 11, wherein the normalized indicator value NE t is calculated for a point of index i, the index i being an integer between 1 and N, N being the number of points on the (P / V) curve, as a ratio between a difference between the indicator value E t calculated for the index point i and a minimum indicator value among the plurality of indicator values ​​calculated for the approximated points, and a difference between a maximum indicator value among the plurality of indicator values ​​calculated for the approximated points and the minimum indicator value, according to the formula: with min(Ej). ejlw j is the minimum indicator value calculated for points on the (P / V) curve with index j between 1 and N, the maximum indicator value calculated for points on the (P / V) curve with index j between 1 and N.

13. A method for correcting a fluid pressure value estimated from a measurement by a sensor (4), the sensor (4) being configured to measure a physical quantity dependent on the actual fluid pressure, in an inflatable element (3) of an implantable medical device (1), the correction method comprising the steps of: estimation of a fluid pressure value from the measurement of the physical quantity by the sensor (4); determination of a discrepancy between the estimated fluid pressure value and an actual fluid pressure, by implementing the method according to any one of claims 1 to 12; obtaining a corrected fluid pressure value, the corrected pressure value being equal to a difference between the estimated fluid pressure value and the determined offset.

14. Product computer program comprising instructions which, when the program is executed by a computer, cause the computer to implement the correction method according to claim 13.

15. Computer-readable recording medium comprising instructions which, when executed by a computer, cause the computer to implement the correction method according to claim 13.

16. Implantable medical device (1) in a human or animal body, comprising: - a fluidic circuit comprising: - an inflatable element (3) into which a volume of a fluid is intended to be injected, - a reservoir (5) suitable for containing the fluid intended to be injected into the inflatable element (3) and fluidically connected to the inflatable element (3), - an actuator configured to transfer a volume of fluid between the reservoir (5) and the inflatable element (3); - a sensor (4) configured to measure a physical quantity dependent on the actual fluid pressure in the inflatable element (3); - a processing unit (7), configured for: - obtain a corrected fluid pressure value by implementing the correction process according to claim 13; - determine a volume of fluid to be transferred between the reservoir (5) and the inflatable element (3), from the corrected fluid pressure value; - control the actuator (6) to transfer the determined volume of fluid.

17. Implantable medical device in a human or animal body according to claim 16, wherein the reservoir (5) is a variable volume reservoir comprising a fixed part and a moving part, the actuator (6) is mechanically coupled to the moving part of the reservoir (5), and configured to linearly move the moving part relative to the fixed part in a direction of movement to adjust the variable volume of the reservoir (5).

18. A medical device implantable in a human or animal body according to claim 17, in which the sensor (4) is a force sensor (4) mechanically linked to the actuator (6) and / or the moving part of the tank (5), the force sensor (4) being configured to measure a tensile and / or compressive force in the direction of travel, the processing unit (7) being further configured to estimate a fluid pressure (P) from the force measured by the force sensor (4), and determine a volume (V) of injected fluid from a position of the actuator (6).

19. Implantable medical device in a human or animal body according to any one of claims 16 to 18, further comprising a sealed housing (10) in which the actuator (6), the reservoir (5) and the sensor (4) are arranged, the sealed housing (10) containing a gas.

20. A medical device implantable in a human or animal body according to any one of claims 16 to 19, the implantable medical device being an occlusive system (1) in which the inflatable element (3) is an occlusive cuff configured to selectively occlude at least one anatomical duct of said human or animal body taken from among: a urethra, a gastric duct, a colon and a rectum.

21. Implantable medical device (1) in a human or animal body according to any one of claims 16 to 20, wherein the inflatable element (3) is elongated in shape and configured for use as a penile implant.

22. Human-machine interface configured to communicate with an implantable medical device (1) according to any one of claims 16 to 21, configured to: - implement a calibration process for the implantable medical device, so as to obtain a curve (P / V) comprising a plurality of points, each point being defined by coordinates corresponding to a value of volume of injected fluid and a pressure value estimated by the processing unit (7) of the implantable medical device (1); - display the (P / V) curve obtained; - implement a method according to any one of claims 1 to 12, the determined offset being stored in a memory accessible by the processing unit (7) of the implantable medical device (1).

Images