Method for simulating deformation of an implantable medical device after implantation

Abstract

The present invention relates to a method for simulating the deformation of an IMD after implantation in a natural cavity from a three-dimensional model of the wall of the cavity, comprising the following steps: determining an intermediate deformation state of a numerical IMD based on the shape of the wall model while remaining contained in the shape; calculating a mechanical equilibrium state of the numerical IMD from the intermediate deformation state, the calculation comprising calculating the mechanical stress of the numerical IMD in the intermediate deformation state, the mechanical stress being a function of the mechanical behavior of the numerical IMD and the mechanical behavior of the wall model; and relaxation of the stress, wherein when calculating the mechanical equilibrium state, the behavior of the wall model is regarded as rigidly non-deformable, and the mechanical behavior of the numerical IMD and / or the static state of the numerical IMD are different between the step of determining the intermediate deformation state and the step of calculating the mechanical equilibrium.

Description

Technical Field

[0001] The present invention relates to the field of numerical simulation of implants in natural cavities before or during an implantation operation. Background Art

[0002] Arteries affected by, for example, aneurysms are often treated with expandable implantable medical devices (IMD)-type implants, such as "stents," "intra-sac cages," or "flow diverters." The goal is to prevent the aneurysm from expanding and rupturing. It is also sought to prevent blood clots formed in the aneurysm sac from migrating and partially blocking the artery.

[0003] For health practitioners who are self-assessing a three-dimensional image of an artery experiencing a localized pathology, such as an aneurysm, it is advantageous to predict the final shape and position that an implant will assume after deployment within the artery. Prediction of implant deformation is preferably performed prior to inserting the implant via an intravascular approach. Thus, the practitioner can select an appropriate implantable medical device (hereinafter referred to as IMD) reference with optimal device size and initial positioning.

[0004] Planning the placement of implants via an intravascular approach is typically done using 2D or 3D images of the artery. Until now, the selection of an IMD reference for the implant has been based at least on planar local measurements (2D) taken by the physician from images taken of the patient. These measurements cannot reliably predict the IMD's alignment, shape, or final position after deployment.

[0005] Furthermore, planar local measurements are based on the physician's experience, and the physician does not benefit from computer assistance in these measurements, which results in a large variability in the selection of the IMD to be implanted.

[0006] Several manufacturers of implantable medical devices have proposed using nomograms, or predetermined mathematical relationships based on planar local measurements, to guide practitioners in selecting an IMD for a given patient. These nomograms can simulate the final shape of the IMD and characterize a reference IMD that fits the morphology of the natural cavity.

[0007] However, the use of nomograms has limitations. The design of nomograms (eg, in the form of a relationship between the diameter of a patient's natural lumen and the predicted length of an IMD after deployment in the lumen) is primarily based on empirical observations.

[0008] International patent application WO 2019 / 122665 A1 in the applicant's name describes a method for simulating the final length of an IMD (especially a "drainer" type implant) after its deployment in a natural cavity. The IMD is discretized into a set of longitudinal three-dimensional segments, which are simulated as cylindrical shapes. The diameter of each segment is then iteratively modified while taking into account the geometric stresses exerted on the IMD by the cavity walls.

[0009] Although the document indicates that mechanical modeling of each segment can be employed before predicting the final length, this prior art approach is limited to predicting the longitudinal deformation and alignment of the IMD, based on the calculation of the geometric stresses exerted by the walls of the cavity.

[0010] This method of the prior art is unable to calculate the deformation field on the mechanical components of the IMD after the IMD is implanted in the cavity, and does not calculate the mechanical balance of the IMD after deployment.

[0011] Furthermore, this simulation method has good accuracy for “drainer” type implants, but is less applicable to the typically non-cylindrical shapes of intracapsular cages or “laser-cut stent” type implants.

[0012] Furthermore, the method described in this document works best if the vessel lumen is considered as a series of straight cylindrical segments with varying radii. As a first approximation, the structural variation of the series of segments of the IMD model is considered to preserve the cylindrical shape of these segments, even though alternatives to the method described in this document propose parameters that take into account the longitudinal compression exerted on the IMD during implantation in the vessel lumen.

[0013] The publication Patient-specific numerical simulation of stent-graft deployment: Validation on three clinical cases by Perrin et al. (HAL Id: inserm-01201545, 17 / 09 / 2015) describes a simulation method that can predict the final deployed shape of a roughly cylindrical arterial endoprosthesis. The three-dimensional geometry of the arterial surface is obtained by interpolation using a shell element model, and the arterial wall is modeled with orthotropic linear elastic behavior. The behavior of the implant is also simulated as orthotropic elastic. The document describes the radial compression of the implant model, which is then interpolated into the arterial wall model using boundary conditions that avoid simulating a complete deployment of the implant model.

[0014] However, calculating the static mechanical equilibrium between the implant model and the arterial wall model remains difficult and can easily lead to unstable modes depending on the shape of the IMD and artery. In particular, the method described in this paper seems less applicable to other types of IMDs, such as intracapsular cages.

[0015] Therefore, the accuracy of predicting the final shape of an IMD after deployment can be further improved, and satisfactory prediction methods are not yet available for non-"flow diverter" IMDs. In the case of an intrasaccular or intra-aneurysmal cage with a variable geometry, the implant can expand longitudinally and radially within the aneurysm sac about its axis of rotation after implantation. After expansion, the implant conforms to the shape of the aneurysm sac until mechanical equilibrium is achieved with the aneurysm sac wall. Prior art simulation methods do not account for these deployment characteristics of the implant. Summary of the Invention

[0016] In view of the foregoing, a method for simulating the final position and deformation of an implantable medical device (hereinafter referred to as IMD) in a natural cavity is needed, which can be performed in a patient before any intervention involving the implantation of the IMD. The method sought must provide post-implantation deformation results that closely resemble the observed final position in order to achieve sufficient clinical accuracy while being applicable to a wide variety of patients and interventions.

[0017] The method sought must also be fast and have no computational time requirements, so as to quickly converge to the simulation results (the mechanical equilibrium state of the IMD), which would be very useful for practitioners to select IMDs and implant conditions, especially in emergency situations. Once the input data are selected and for a given IMD reference, the required computational time is, for example, of the order of 5 to 60 seconds.

[0018] Furthermore, the sought-after method must be robust to apply to the widest possible range of IMDs, for example, non-tubular shapes such as braided cages, while providing stable numerical solutions. None of the prior art methods can satisfy this type of IMD shapes.

[0019] There is also a need for a method of simulating the final position and shape of an IMD that allows prediction of the local approximation of the implant to the walls of the natural cavity.

[0020] In this respect, according to a first aspect, the present invention relates to a method for simulating deformations following implantation of an implantable medical device (referred to as IMD) in a natural cavity from a three-dimensional numerical model of the walls of the cavity, said method comprising the following steps implemented by a processing unit:

[0021] i. determining an intermediate deformation state of a numerical value IMD representing an IMD, the numerical value IMD in the intermediate deformation state deforming according to the shape of the wall model while remaining included in said shape,

[0022] ii. calculating a mechanical equilibrium state of the numerical IMD from the intermediate deformation state, including calculating the mechanical stress to which the numerical IMD is subjected in the intermediate deformation state, wherein the mechanical stress is a function of the mechanical behavior of the numerical IMD and the mechanical behavior of the wall model, and includes relaxation of the stress, wherein the calculated mechanical equilibrium state corresponds to the deformation of the simulated IMD after implantation.

[0023] The simulation method of the present invention enables health practitioners to obtain a prediction of the final state of an IMD implanted in a patient's natural cavity. This prediction takes into account the predicted mechanical behavior of the IMD and the behavior of the walls of the natural cavity, thereby enabling accurate simulation.

[0024] Determining intermediate deformation states before calculating the mechanical equilibrium state simplifies the calculation of mechanical equilibrium because it provides an efficient initialization of the calculation. Numerous methods for simulating IMD deformation in the existing scientific literature reproduce nearly the entire actual deployment of an IMD initially compressed in a microcatheter. During their actual deployment, an IMD employs a "deformation history" that is very complex to simulate. Consequently, these methods have proven to be computationally expensive and are sometimes unstable, making them incompatible with clinical practice.

[0025] The simulation method of the present invention does not require the reproduction of the entire actual history of deformation and proposes to define intermediate deformation states (theoretically) in order to speed up and simplify the simulation of the unfolding of the IMD.

[0026] During the method of the present invention, the simulated numerical IMD uses a simplified "non-physical" deformation history, which is easier to simulate (faster and more stable than simulating a complete deformation history). This simplified deformation history preferably starts from a rest state of the numerical IMD. The IMD then undergoes intermediate deformation states, then considers the mechanical stresses experienced by the IMD in the intermediate deformation states and simulates the relaxation of these stresses to reach the final deformed state of the IMD.

[0027] The final deformation state of the IMD obtained in this way remains close to reality. The intermediate deformation states are carefully selected so that the entire deformation history can be effectively simulated. In particular, during the determination of the intermediate deformation states, a specific mechanical behavior of the numerical IMD can be selected, which can optionally be different from the mechanical behavior used during the calculation of the mechanical equilibrium, to facilitate and accelerate the calculation (or alternatively, the static state of the numerical IMD is different from the calculation of the intermediate deformation states and the mechanical equilibrium state). This selection can also prevent the determination of intermediate deformation states and / or the calculation of the mechanical equilibrium state from leading to localized instability modes, thereby adversely affecting the robustness of the simulation in the face of the variability of the possible IMD.

[0028] It should be noted that the intermediate deformation states do not have to be calculated in a complex manner by the processing unit using an equation solver. The calculation of intermediate deformation states of an IMD completely contained within a natural cavity is simpler and faster than the complete deformation history of the IMD. Furthermore, the fact that a three-dimensional wall model of the natural cavity is used makes the results more precise, especially compared to using a generic nomogram. For example, a customized simulation specific to the patient is obtained. Preferably, during the calculation of the mechanical equilibrium state, the mechanical behavior adopted by the natural cavity wall model is a non-deformable, rigid behavior, which facilitates the calculation of the mechanical equilibrium state while constituting an acceptable approximation from a physical perspective.

[0029] The rapid, robust, and accurate simulation results of the present method enable health practitioners to quickly infer the predictive validity of a given IMD reference for treating a patient, especially in emergency situations. The computational time from the moment the numerical IMD model is obtained can be less than one minute, unlike the average computational time of prior art methods that require simulating a realistic deformation history. Therefore, the present method is compatible with clinical practice.

[0030] Other possible and non-limiting features of the method of the invention, alone or in any technically possible combination thereof, are as follows:

[0031] The mechanical behavior of the numerical value IMD considered during the determination of the intermediate deformation state differs from the mechanical behavior of the numerical value IMD considered during the calculation of the mechanical equilibrium.

[0032] The state of rest of the value IMD considered during the determination of the intermediate deformation state differs from the state of rest of the value IMD considered during the calculation of the mechanical equilibrium.

[0033] -The mechanical behavior of the wall model used to calculate the mechanical equilibrium state is non-deformable rigid behavior.

[0034] - The intermediate deformation states are determined based on the contact interactions calculated between the 3D vertices of the IMD and the 3D vertices of the wall model.

[0035] During the determination of the intermediate deformation states, the wall model is geometrically deformed from the initial state so as to completely contain the numerical IMD in its rest state, and then the wall model is restored to the initial state to obtain the intermediate deformation states of the numerical IMD.

[0036] - The determination of the intermediate deformation state comprises obtaining numerical IMD constrained in a tool surface associated with the implant tool model, and integrating the constrained numerical IMD in the wall model in order to obtain the intermediate deformation state.

[0037] - The method further comprises the step of determining a centre line of the natural cavity from the wall model, and wherein the numerical IMD is deformed during its integration to follow the centre line.

[0038] The numerical IMD includes a plurality of segments, and further includes a plurality of nodes, each node connecting ends of two consecutive segments.

[0039] The mechanical behavior of at least one segment corresponds to that of a beam, preferably of cylindrical shape.

[0040] - The segments have a predetermined Young's modulus and / or density and / or Poisson's coefficient.

[0041] at least one segment of the numerical IMD has a beam mechanical behavior and is modeled with a first diameter, or with a first thickness, and / or with a first elastic modulus (such as Young's modulus and / or Poisson's coefficient), and / or with a first slenderness ratio coefficient, and / or with a first radius of gyration, and / or with a first set of critical buckling loads during the determination of the intermediate deformation state,

[0042] And the segments are modeled with different second diameters, and / or different second thicknesses, and / or different second elastic moduli, and / or different second slenderness ratios, and / or different second radii of gyration, and / or different second sets of critical instability loads during the calculation of the mechanical equilibrium state.

[0043] The mechanical behavior of at least one node corresponds to the behavior of a rotating body.

[0044] The calculation of the mechanical equilibrium state of the numerical IMD consists in calculating for each node i of the numerical IMD, in a three-dimensional reference system linked to the wall model, the displacement fields Dxi, Dyi, Dzi and the rotation fields Rxi, Ryi, Rzi, the two fields being calculated by applying the basic dynamics principles at said node.

[0045] - Calculation of the mechanical equilibrium state of the numerical IMD includes, for at least one node of the numerical IMD, calculation of the normal force and / or friction force exerted by the wall model on said node, modeling the penetration resistance of the wall and the friction between the IMD and the wall, respectively.

[0046] The segments and nodes of the numerical IMD have at least one end point, preferably at least two end points.

[0047] The overall shape of the IMD is flattened at an end point, preferably at a plurality of end points.

[0048] During the determination of the intermediate deformation state, the end extreme points are modeled with a first concavity, and during the calculation of the mechanical equilibrium state, the end extreme points are modeled with a second concavity different from the first concavity.

[0049] - Numerical IMD is a model of an intracapsular cage.

[0050] - Numerical IMD is for the model of the laser-cut stent.

[0051] - The method comprises the subsequent step of calculating the predicted parity of at least a portion of the three-dimensional vertices of the numerical IMD on the wall model, preferably calculating the parity of a plurality of nodes of the numerical IMD on the wall model.

[0052] - the numerical IMD corresponds to an IMD reference derived from a set of IMD references recorded in a database,

[0053] Steps i., ii. and iii. as defined above are repeated for each reference in a set of references.

[0054] -The method comprises a subsequent step of determining the actual IMD best suited for implantation in the natural cavity from a set of references, determined based on the deformation state of each reference IMD from the set of references after implantation and / or based on the alignment of said IMD on the wall of the natural cavity after implantation.

[0055] According to a second aspect, the invention relates to a computer program product comprising code instructions for implementing the simulation method defined above, when said code instructions are executed by a processing unit.

[0056] According to another aspect, the present invention relates to a processing unit comprising:

[0057] Device for obtaining a three-dimensional wall model of a natural cavity,

[0058] The means for obtaining a numerical IMD is preferably configured to generate said numerical IMD based on an IMD reference derived from a database,

[0059] a computing device configured to determine an intermediate deformed state in which the numerical value IMD deforms according to the shape of the wall model while remaining contained within said shape,

[0060] The computing device is further configured to compute a mechanical equilibrium state of the numerical IMD based on the mechanical behavior of the numerical IMD and the mechanical behavior of the wall model,

[0061] The processing unit is configured to implement the simulation method as defined above. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Other features, objects and advantages of the present invention will become apparent from the following description, which is intended to be purely illustrative and not restrictive, and should be read in conjunction with the accompanying drawings, in which:

[0063] Figure 1 Components for simulating deformation after IMD implantation according to an example are schematically illustrated.

[0064] Figure 2 1 and 2 show the steps of a method for determining deformation after IMD implantation according to a first embodiment.

[0065] Figure 3a and Figure 3b Represents a series of steps to generate a numerical vascular tree wall model.

[0066] Figure 4 Numerical model representing an intracapsular cage-type IMD.

[0067] Figure 5 is a schematic diagram of an exemplary modeling possible for IMD, showing nodes modeled by rotation.

[0068] Figures 6a to 6d Indicates that according to Figure 2 The proposed method simulates the deformation process of the cage after implantation, and the series of states of the cage model and the wall model in the capsule.

[0069] Figure 7 1 and 2 show the steps of a method for determining deformation after IMD implantation according to a second embodiment.

[0070] Figure 8 Numerical model representing the IMD of the “laser-cut stent” type.

[0071] Figures 9a to 9f Indicates that according to Figure 7 The method simulates the series of states of the stent model during the deformation process after stent implantation. Figure 9a 、 Figure 9e and Figure 9f In the , the stent model is included in the wall model of the patient's natural cavity. Figures 9b to 9d In this study, the stent model was confined within a microcatheter outside the natural lumen.

[0072] Figure 10Numerical IMDs simulating the final shape and isotopy of a "laser-cut" type stent are shown superimposed on an image of a real stent deployed inside a real vascular tree. DETAILED DESCRIPTION

[0073] In the following, reference will be made indiscriminately to an "implant" or "implantable medical device" (or IMD) to an expandable implant that is capable of adopting a final position within a natural cavity (after implantation and before deployment) that is different from its initial position (after implantation and before deployment) and also different from its resting position (deployed in the open air).

[0074] Such implants generally have a structure consisting of a material that is biocompatible with human tissue.At the beginning of implantation, the implant is generally held in a compressed position by an implantation tool (eg by a catheter).

[0075] Hereinafter, an example will be considered in which the natural lumen to be treated is an artery of a human or animal patient. However, it will be understood that the present invention can be applied with equal advantage to any other body conduit capable of receiving an IMD.

[0076] Implantation of the IMD simulated by the method for determining IMD deformation described below is then performed via an intravascular approach. For example, the implantation is performed under interventional radioscopic guidance using an implantation tool such as a microcatheter.

[0077] "Geometrical characteristics" or "morphological characteristics" of a natural cavity refer to the shape characteristics of the cavity that locally influence the final position of the implant, especially but not limited to the minimum diameter, perimeter, radius of curvature and spatial derivatives thereof.

[0078] Furthermore, a region of interest (including the artery to be treated) may be indicated by the abbreviation “ROI.” An aneurysm region constitutes an example of a ROI.

[0079] Throughout the drawings and the following description, similar elements are designated by the same letter numerical designations.

[0080] System for simulating deformation after IMD implantation

[0081] exist Figure 1 2 shows a system for determining the position of an IMD according to the present invention, the system comprising a processing unit 20. The processing unit is, for example, a processor configured to implement the method for determining the deformation of an IMD according to any exemplary embodiment described below.

[0082] Advantageously, the processing unit 20 is configured to communicate with an acquisition unit 22 capable of acquiring views such that a three-dimensional image of a region of interest of the patient can be reconstructed. The region of interest comprises an artery.

[0083] The acquisition unit 22 can be, for example, an X-ray imaging system, and the views can be acquired, for example, within the scope of a neuroradiological procedure, such as a three-dimensional angiography acquisition.

[0084] The processing unit 20 communicates with the acquisition unit 22 and / or with the database DB1 to receive the image Im via a wired link and / or via a wireless link via any suitable network (e.g. the Internet). Alternatively, the processing unit may extract the image from a hard disk or receive the image from a peripheral storage support reading device (e.g. a CD reader or a USB port).

[0085] In an alternative or in combination with the acquisition unit 22 , the processing unit 20 can communicate with a database DB1 in which three-dimensional images and / or views of the patient's natural cavity to be treated are recorded, so that such a three-dimensional image can be reconstructed.

[0086] The processing unit further includes a database DB2 containing data regarding reference implantable medical devices (IMDs). This data may be provided by the IMD manufacturer or determined through analysis or experimentation. As will be seen below, the processing unit 20 includes a numerical model generation device capable of generating a numerical IMD based on the reference IMDs derived from the database DB2.

[0087] The data associated with the IMD reference in database DB2 includes physical IMD characteristics (such as maximum diameter and / or maximum length), and / or a pre-recorded mechanical IMD model (eg, a model in the form of a network of segments connected together by nodes).

[0088] According to a possible alternative, the database DB2 contains a set of IMD references, of which a specific reference can be selected to initiate a simulation according to the method described below. This alternative is advantageous because it enables the user to obtain simulation results for a plurality of different IMD references and then select the reference that gives the most satisfactory results (e.g., the best isotope on the cavity wall) for intervention.

[0089] In an alternative, the database DB2 is remote from the processing unit 20 and the link between the processing unit 20 and the database DB2 is established by any suitable means, for example wirelessly via a communication network.

[0090] The processing unit 20 is further connected to a display device 21 which provides a graphical interface to the user for displaying a three-dimensional image modeling the region of interest (usually comprising the natural cavity to be treated).

[0091] The display device 21 can be further configured to display simulation results resulting from implementing the methods described below. For example, after simulating the deployment of an IMD in a region of interest, it can display a three-dimensional view of the numerical IMD. The display device 21 displays a user interface for inputting commands and, optionally, for selecting an IMD reference. The user of the processing unit 20 and the associated display device 21 is, for example, a healthcare practitioner.

[0092] Method for simulating deformation after IMD implantation

[0093] Figure 1 The system shown, in particular the system including the processing unit 20, can be used to implement a method for simulating deformation of an IMD after implantation in an artery of a patient.The natural lumen is then a region of interest within the patient's vascular tree.

[0094] exist Figure 2 An example of a simulation method that can be implemented by the processing unit 20 is shown in , where the IMD to be simulated is an intrasaccular cage that must be used to treat an aneurysm at a region of interest. It should be noted that the simulation method according to this example can be used for other types of self-expanding IMDs.

[0095] From the three-dimensional model of the arterial wall at the ROI, and if necessary, from the three-dimensional IMD model (hereinafter referred to as "numerical IMD", which for example corresponds to a static IMD, such as can be found "off the shelf"), the simulation can obtain a three-dimensional model of the IMD in mechanical equilibrium within the artery after implantation and deployment in the artery.

[0096] Here, the simulation results include the relative positions of each point of the numerical value IMD with respect to each point of the wall model representing the ROI of the artery in the mechanical equilibrium state.

[0097] This simulation method is very useful in situations where health practitioners need to make very quick decisions, such as selecting an IMD to implant in a patient who has just suffered or is currently experiencing a cerebrovascular accident (CVA). The simulation also offers advantages in treating stenosis, thrombus removal, heart valve replacement, or abdominal aortic aneurysm.

[0098] It should be noted that simulation of deformation after IMD implantation can be performed upstream of the implantation or during the implantation.

[0099] It is crucial to have a simulation with very high safety and performance levels to ensure the patient's physical integrity and the effectiveness of the treatment, while also being fast enough to enable real-time decisions. From the moment a 3D model of the arterial wall, where the ROI is located, is available, and with a processing unit of standard computing power in the medical computing field, a full simulation for a given IMD reference can be performed in a short time, for example, between 5 and 60 seconds. The simulation must also be robust and take into account the anatomical characteristics of the natural lumen being treated within the patient's body.

[0100] Numerical wall model of the cavity to be treated

[0101] In optional step 100 , a model 1 of the patient's arterial wall including the ROI to be treated is generated by the processing unit 20 or by a separate computing device capable of exchanging data with the processing unit.

[0102] Alternatively, the model 1 of the arterial wall has been generated before the simulation and is obtained by the processing unit 20 from a medical database.

[0103] The wall model 1 is typically obtained by image processing from a three-dimensional image of a patient's artery, obtained, for example, by rotational 3D angiography.The three-dimensional image of the artery can be segmented by the "marching cubes" method known in the art of image reconstruction.

[0104] Here, the three-dimensional image is directly extracted from the acquisition unit 22. In an alternative solution, the three-dimensional image can be obtained from the database DB1.

[0105] Figure 3a An exemplary arterial wall model 1 is shown. An entry point I and an exit point S are defined, for example, manually by a practitioner. In an alternative embodiment, the entry and exit points are automatically detected. Point I is the point through which an implantation tool (e.g., a microcatheter) including a compressed IMD can be inserted into the ROI during an intervention.

[0106] The wall model 1 preferably comprises a discrete three-dimensional surface that approximates the actual wall of the artery.For example, the model 1 comprises a surface formed by mutually adjacent triangles, said surface being flat within a given triangle.

[0107] Preferably, the processing unit further generates or obtains the center line C of the artery at the ROI. Figure 3b It is shown in Figure 3a Calculate the center line C of the wall model 1.

[0108] The center line C is advantageously oriented. It then comprises a set of spatial points. At a plurality of such points, optionally at each such point, a preferably directly orthogonal local basis R is obtained. In an alternative, the basis obtained may be non-directly orthogonal.

[0109] For example, the centerline C can be calculated by minimizing the travel time of the fluid particles along the wall model 1. The centerline then corresponds to the fastest path for the particles to migrate from the entry point I to the exit point S. For example, the fastest path can be calculated by considering the assumption that the velocity of the fluid particles is proportional to their distance from the vessel wall.

[0110] The calculation of the centerline C facilitates the initial positioning of the numerical IMD. The numerical IMD can be positioned at any point on the centerline inside the wall model.

[0111] In the case where the IMD is simulated inside an implantation tool (such as a microcatheter), the calculation of the centerline C also helps determine the intermediate deformation states of the IMD. This situation is described below.

[0112] For the subsequent calculation of intermediate deformation states and the calculation of the mechanical equilibrium, the processing unit 20 does not have to itself dispose of a physical model of the mechanical behavior of the elements of the wall model 1. A geometrical representation of the surface of the wall model 1 may be sufficient.

[0113] Numerical IMD Model - Example of an Intracapsular Cage

[0114] In optional step 200, an IMD model 2 or "numerical IMD" is generated by the processing unit 20 or by a separate computing device capable of communicating with the processing unit and recorded in a memory of the processing unit.

[0115] Alternatively, the processing unit 20 extracts the numerical value IMD from a database.

[0116] The numerical IMD 2 constitutes a physical and geometric modeling of the IMD, which allows simulation of its interaction with the wall of the artery to be treated. At this stage, the numerical IMD preferably corresponds to a static implant, such as one that might be found "off the shelf." The numerical IMD 2 is recorded as a series of points whose three-dimensional coordinates are stored in memory. Preferably, the connections between the nodes are also recorded in memory, which allows reconstruction of the segments that discretize the mechanical structure of the IMD.

[0117] Advantageously, the point set of the numerical IMD 2 comprises a plurality of segments 10 connected together by nodes 11. Each node connects the two ends of two consecutive segments. A given node may optionally connect more than two segments. Thus, the nodes 11 of the numerical IMD are interconnected by segments 10. The assembly formed by the nodes and segments forms a mesh that constitutes a discrete model of the shape of the IMD.

[0118] This model is particularly suitable for modeling low-thickness IMD.

[0119] Here, the simulated IMD is an intracapsular cage, which is made of metal wires made of biocompatible materials and interwoven to form a grid. Figure 4 As shown, the numerical IMD 2a applicable to such an implant is a set of segments describing an overall spherical shape that flattens at the poles of the numerical IMD.

[0120] The numerical IMD obtained or generated by the processing unit may correspond to a reference derived from a set of IMD references recorded in the database DB 2. For example, the practitioner may select, via the user interface, a reference corresponding to a shape and / or a specific size of a stationary implant, and / or a specific material, and / or a type of implant, such as an intracapsular cage, a laser-cut stent, a flow diverter, an overall tapered implant, etc.

[0121] For intra-aneurysm cage-type IMDs, a mechanical model of the braided structure of the IMD can be employed using a set of equivalent shells. However, the surface of the equivalent shells varies only slightly in area, making it difficult to adequately model the braided IMD structure. Consequently, the solution obtained by simulating mechanical equilibrium is not very stable.

[0122] At this stage, predetermined mechanical behavior can be attributed to the elements of the numerical IMD, here to the segments 10 and nodes 11 of the numerical IMD 2a.

[0123] As described below, the mechanical behavior associated with the various elements of the numerical IMD when the IMD is in intermediate deformation states is particularly useful for determining the mechanical stresses exerted on the IMD.

[0124] It should be noted that during determination of the numerical IMD and intermediate deformation states of the wall model, the mechanical behavior of the numerical IMD may be unknown.

[0125] On the other hand, the mechanical behavior of the IMD is known from the post-analysis of mechanical equilibrium.

[0126] As an example, each segment 10 is considered here as a beam element, which makes it possible to discretize the neutral fiber of the IMD.

[0127] The "neutral fiber" refers to the curved line connecting the centers of gravity of the linear segments forming the structure of the IMD.

[0128] Therefore, the structure of the IMD is equivalent to a tubular volume described by segments placed end-to-end along the neutral fiber. The tubular volume is generated by a set of straight line segments. The straight line segments are circles here, but other types of segments (triangles, rectangles, etc.) can also be used for modeling.

[0129] This model allows the forces exerted on the tubular volume of the segment to be differentiated at the neutral fiber during subsequent numerical simulations of the mechanical interaction between the IMD and the wall.

[0130] A set of predetermined parameters may be associated with each beam element (each segment), including a predetermined Young's modulus E, Poisson's coefficient ν, and density p. To simplify modeling, all beam elements may be associated with the same parameters.

[0131] The material constituting the beam elements is preferably considered to be elastic, homogeneous and isotropic.

[0132] In the specific example of a braided stent, such as the intracapsular cage modeled by numerical IMD 2a, the relative translational motion between two superimposed wires of the mesh, as well as the friction caused by the motion of one wire relative to the other, are preferably ignored in order to simplify and speed up the subsequent mechanical balance calculations.

[0133] In this respect, it is advantageous to model the nodes located at the intersections of several wires of the mesh of the intracapsular cage as simple rotational mechanical connections. Figure 5 is a schematic diagram of such modeling of a node 11 for a value of IMD 2a. Here, four segments intersect at the node 11, where two segments 101 simulate the first wire of the cage and two segments 102 simulate the second wire of the cage.

[0134] The node has a rotational mechanical behavior, assuming that each segment is free to rotate in space relative to the other segments (neglecting friction). On the other hand, the segments are not free to translate relative to each other (neglecting the relative translation of the two wires).

[0135] This modeling approach is suitable for stents with densely woven wires, such as intracapsular cages, and enhances the speed, stability, and robustness of deformation calculations applied to numerical IMD 2a.

[0136] It should be noted that modeling the IMD as a set of nodes and segments, and modeling the nodes as solids of revolution, can even be used to simulate the deformation of the IMD without determining intermediate deformation states prior to the solution of the mechanical equilibrium.

[0137] Determination of intermediate deformation state

[0138] Back to Figure 2 The simulation then comprises determining an intermediate deformation state 300a of the system formed by the wall model 1 and the numerical IMD 2a.

[0139] This intermediate deformation state is the theoretical state of the IMD relative to the arterial wall. In this intermediate deformation state, the numerical IMD is completely contained within the wall model.

[0140] In intermediate deformation states of numerical IMD, the wall model preferably has the same shape as at rest.The "shape of the wall model" refers to the spatial positions of the points of the wall model relative to each other.

[0141] Similarly, the shape of the numerical IMD depends on the positions of the three-dimensional vertices of the numerical IMD. The shape of the numerical IMD is drawn by the surface connecting the nodes.

[0142] The walls of the natural cavity are considered here to be rigid and non-deformable, especially during the calculation of the mechanical balance between the numerical IMD and said walls. Therefore, the calculation of the mechanical balance is robust and fast while maintaining an acceptable approximation to reality.

[0143] On the other hand, numerical IMD is not considered to be non-deformable. It will deform according to the shape of the wall model.

[0144] However, according to Figure 2 In the method of the illustrated example, the wall model is allowed to deform temporarily to help calculate the intermediate deformation state of the numerical IMD. During the calculation of the intermediate deformation, this transient deformation of the wall model is not mandatory.

[0145] More specifically, during the calculation of the intermediate deformation state, in substep 301, the value IMD 2a is first placed in a reference system linked to the wall model 1. The center of the value IMD 2a is located on the center line C if it has been calculated.

[0146] Advantageously, the numerical IMD 2a may be placed at a location point (eg an entry point, not shown) in a reference system linked to the wall model.

[0147] The numerical IMD 2a simulates an intrasaccular cage at rest, in which it is not subjected to mechanical stresses that would tend to cause it to retract. To properly align the cage with the aneurysm wall, the cage must extend further at rest than within the aneurysm. Therefore, the numerical IMD 2a is preferably selected to be of sufficient size to intersect the surface of the wall model 1 at rest when the IMD is positioned on the centerline.

[0148] Figure 6a The numerical IMD is shown in a first extended state 2a-1 and the wall model in a rest state 1-1 at the end of sub-step 301. Reference numeral 12 denotes two opposite end poles of the IMD, which are initially retracted inwards.

[0149] Next, in sub-step 302, the wall model is enlarged so that it covers the entire surface of the numerical IMD. This enlargement corresponds to the instantaneous deformation of the wall model described above. The deformation of the wall model is a sub-step for calculating the intermediate deformation state of the IMD within the wall, but this deformation of the wall model is not retained for subsequent calculation of the mechanical balance of the IMD.

[0150] At the end of sub-step 302 , each node 11 of the numerical IMD is surrounded by the surface of the wall model.

[0151] In order to enlarge the wall model, a cylindrical deformation, for example, is applied to the surface of the wall model 1 according to the maximum diameter and / or the maximum height of the value IMD 2a at rest.

[0152] Figure 6b The numerical IMD is shown in a second extended state 2a-2 and the wall model in its enlarged state 1-2 at the end of sub-step 302 so as to include the extended IMD.

[0153] In this example, in the second extended state 2a-2, the end points 12 of the numerical IMD are reoriented outward. Thus, the numerical IMD in state 2a-2 is generally convex.

[0154] Before considering contact interactions with the wall model, the rest state of the numerical IMD can be selected as the starting position for the numerical IMD to simplify the calculation of intermediate deformation states. In particular, the geometry of the numerical IMD (drawn by segments and nodes) at rest can be modified between determining the intermediate deformation states and calculating the mechanical equilibrium state.

[0155] In this example, the geometry of the numerical IMD at rest is modified by reversing the concavity of the end poles of the intracapsular cage to obtain a second extended state 2a-2, as shown in FIG. Figure 6b The advantage of this geometry is that during the subsequent mechanical balance calculation, a deformation history that is closer to the true deformation history of the IMD can be obtained.

[0156] Back to Figure 2 , next in sub-step 303 the wall model is gradually deformed to return to its rest state 1-1 (the wall is considered rigid and non-deformable), which results in a deformation of the numerical IMD until the sought intermediate deformation state is reached.

[0157] As the wall model gradually returns to rest, the mechanical calculation of the contact interaction between the IMD and the wall is preferably performed iteratively. During these iterative processes, a series of deformations of the numerical IMD are calculated based on the obtained contact interactions. In each iteration, the positions of the three-dimensional vertices of the numerical IMD are recalculated. During these series of deformations, the numerical IMD remains contained in the wall model.

[0158] Preferably, for the calculation of the contact interactions, the mechanical behavior of the mechanical elements of the numerical IMD (here the nodes and segments of the intracapsular cage) is simplified compared to the mechanical behavior considered below for the calculation of the mechanical balance.

[0159] In this example, the thickness of the segments is multiplied by 10 to avoid buckling phenomena during the step 300a of calculating the intermediate deformation state.

[0160] In an alternative or in combination, one or more of the following parameters of the segments (e.g. each of which is modelled as a beam) may be modified only for step 300a of calculating the intermediate deformation state, in particular so that the calculation can be facilitated: diameter and / or thickness and / or elastic modulus (Poisson's modulus and / or Young's modulus) and / or slenderness ratio and / or radius of gyration and / or one or more critical buckling loads.

[0161] It should be noted that these options for modeling numerical IMD to determine intermediate deformation states can also be used for laser cut stent types (step 300b described below) or other types of numerical IMD.

[0162] As an illustrative example, for a laser-cut stent type IMD, the average diameter of the stent in an intermediate deformed state (e.g., the stent inserted into the tool surface) can be selected to be strictly smaller than the "true" diameter of the stent calculated during determination of the mechanical equilibrium state of the IMD.

[0163] At the end of step 303 of returning the wall to the rest state, the value IMD has a deformation state selected as the intermediate deformation state E2. Figure 6c The intermediate deformation state E2 of this example is shown in FIG.

[0164] The advantage of determining this intermediate deformation state is that it neatly initializes the subsequent calculation of the mechanical equilibrium between the numerical IMD and the wall model.

[0165] Starting from an intermediate deformation state, the numerical IMD will gradually relax during the calculation of the mechanical equilibrium of the IMD, taking into account the more complex mechanical behavior of the IMD and optionally the wall.

[0166] By using the intermediate deformation states of the IMD (in theory) during subsequent calculations of mechanical equilibrium, there is no need to re-trace the entire deformation history of the IMD between the rest state and the final state. This reduces the overall computational time for simulating the unfolding of the IMD. However, the calculation of mechanical equilibrium remains reliable and robust, as the intermediate deformation states of the numerical IMD are determined based on the geometry of the IMD at rest and the shape of the walls of the natural cavity.

[0167] Modeling of mechanical interaction between IMD and walls

[0168] Considering the calculation of mechanical equilibrium, it is recommended to model the mechanical interactions between the surface of the IMD and the walls of the lumen. These are primarily the contact interactions that determine how the IMD deforms to adapt to the anatomy of the artery after implantation.

[0169] Advantageously, the walls of the cavity are considered rigid and non-deformable, so that expansion of the walls is negligible under the influence of spontaneous expansion following implantation of the IMD.

[0170] Therefore, the numerical analysis of the mechanical equilibrium can be performed by using the displacement and / or rotation equations for each point in the numerical IMD of the reference wall model.

[0171] In this example, a co-rotational formulation is used to interpret the mechanical equations for the displacement and rotation of the nodes 11 of the numerical IMD. By applying fundamental dynamics principles to each node, indexed by index i, the displacement fields (Dxi, Dyi, Dzi) and rotation fields (Rxi, Ryi, Rzi) for each node i of the numerical IMD are computed in a reference frame linked to the wall model.

[0172] Preferably, in applying the basic dynamics principles, the inertia of the IMD is neglected and the acceleration is taken to be zero, thus applying the basic statics principles.

[0173] In order to establish the equations for the displacement and rotation fields at each node, it is recommended to model the forces exerted by the walls of the cavity at each node i.

[0174] In this regard, penalty methods are advantageously used.

[0175] For each node i, starting from a given state of the numerical value IMD (eg, the second extended state 2a-2 of the numerical value IMD), the processing unit 20 first determines whether the node i penetrates the wall of the cavity.

[0176] During the process of gradually narrowing the wall model 1 to determine the intermediate deformation state of the IMD, some nodes of the numerical IMD re-contact the surface of the wall model.

[0177] Similarly, during the relaxation of the mechanical stress of the IMD, until the mechanical equilibrium between the IMD and the wall is determined, the nodes of the numerical IMD re-contact the surface of the wall model.

[0178] If penetration is detected at node i, the force applied to node i is the normal force F exerted by the wall model on said node normal and friction force F friction Modeling is performed to simulate the resistance of the wall to node penetration and the friction between the IMD and the wall.

[0179] Force F normal The norm of is equal to the product k×p, where k is the stiffness of the spring, which simulates the contact stiffness of the wall, and p is the penetration distance of node i in the wall model according to the direction perpendicular to the wall. When the penetration and stiffness of the spring are high, the force F normal The bigger it is.

[0180] Force F friction is modeled in the direction tangential to the wall and its norm is equal to the force F normal The norm of is multiplied by the friction coefficient μ.

[0181] Alternatively, the model of the mechanical interaction can be made by integrating only the force F normal , which corresponds to a frictionless contact. However, it is preferred to integrate the friction and normal forces to ensure good accuracy of the mechanical balance simulation.

[0182] Furthermore, taking friction into account accelerates the convergence of the mechanical equilibrium state calculation, thus reducing simulation time.

[0183] The above-mentioned penalty method for obtaining the mechanical equations for displacement and rotation provides a good compromise between computational speed, robustness of the mechanical model, and accuracy of the simulation results. It should be noted that this penalty method can also be used for numerical IMDs modeled by means other than nodes and segments.

[0184] It will be appreciated that other numerical methods may be envisaged to calculate the contact interaction between the IMD and the wall in combination with or in lieu of the penalty method.

[0185] Preferably, boundary conditions can be imposed on certain vertices of the numerical IMD and taken into account in the mechanical behavior of the IMD.

[0186] During the calculation of the series of deformations of the IMD, the translational and / or rotational displacements of certain nodes of the numerical IMD (e.g., nodes located on the lower edge of the numerical IMD) may be constrained. An advantage of using boundary conditions at the vertices of the inner edge of the IMD is that it improves the contact modeling at the attachment of the IMD to an implant tool (such as a microcatheter).

[0187] These boundary conditions are used during the calculation of intermediate deformation states and / or during the calculation of the mechanical equilibrium of the numerical IMD.

[0188] The first advantage of using boundary conditions is that the solution of the system of equations can be better adapted to the conditions and the obtained solution can be more stable.

[0189] A second advantage, particularly with respect to the determination of the mechanical balance of the numerical IMD, is that the numerical IMD is guided to gradually deform to a final deformed state that is closer to clinical reality.

[0190] Calculation of mechanical balance

[0191] According to the intermediate deformation state E2 obtained previously, the mechanical equilibrium state E3 of the numerical IMD is calculated by obtaining the mechanical stresses at the vertices of the numerical IMD and simulating the relaxation of these stresses through calculation.

[0192] The relaxation of the mechanical stresses applied on the vertices of the IMD corresponds to an iterative calculation of a series of deformed states of the numerical IMD until a position is reached that is considered to be in mechanical equilibrium with the wall model.

[0193] During these series of states, the numerical IMD 2 gradually relaxes to conform to the shape of the wall model 1 , as would be the actual behavior of an IMD as it moves toward its rest position as it deploys inside an artery.

[0194] The calculated stresses depend on the mechanical behavior of the numerical IMD and the wall model. To calculate the mechanical equilibrium, the mechanical stresses exerted on the IMD by the wall are included. The mechanical stresses in this context include the contact interactions between the IMD and the wall, such as those calculated using the modeling defined above.

[0195] It should be remembered that during the calculation of the mechanical equilibrium, the behavior of the wall model 1 is preferably chosen to be rigid and non-deformable. Thus, the wall has the same shape during the mechanical equilibrium calculation as at the beginning of the simulation, except that, as described above, the wall model may undergo transient deformations during the determination of intermediate deformation states.

[0196] Back to Figure 2 , at step 400, the deformation of the IMD to achieve mechanical equilibrium is calculated by solving equations representing the mechanical interaction between the numerical IMD and the mechanical elements of the wall model.

[0197] The mechanical equilibrium state is optionally the last state of several progressive deformation states calculated from an intermediate deformation state.

[0198] The formulation of the equation for the mechanical interaction between the IMD and the wall preferably corresponds to a co-rotational formulation of the displacements of the nodes of the numerical IMD 2 and the rotational field, such as defined above.

[0199] Figure 6d shows the mechanical balance calculation at the end of the Figures 6a to 6c Numerical IMD and wall models. Figure 6c The mechanical equilibrium state E3 of the numerical IMD is calculated by considering the real geometric shape of the IMD in the static state.

[0200] It should be recalled that in the extended state 2a-1, the end pole 12 of the IMD retracts inward. During the mechanical equilibrium calculation, the mechanical stresses applied to the IMD relax, and due to this relaxation, the end pole 12 spontaneously retracts inward again. Therefore, the concavity of the end pole 12 is selected to be different between the intermediate deformation state of the numerical IMD (here, the intracapsular cage) and its mechanical equilibrium state.

[0201] The mechanical equilibrium calculations performed here are of a nonlinear type. This allows for a very close approximation of the actual mechanical interactions. This results in a very reliable and precise prediction of the implant's shape and final placement in the natural cavity.

[0202] In this example, the mechanical equilibrium is calculated for each node with the numerical IMD 2a. Alternatively, the solution can be performed only for certain points, and the positions of the other nodes at equilibrium can be inferred from the positions of these points.

[0203] The calculation of the mechanical equilibrium is considered complete when the convergence criterion of the deformation state recorded in the processing unit is reached.

[0204] Computing the mechanical equilibrium from intermediate deformation states improves the efficiency of the simulation due to the rapidity and stability of the solutions obtained.

[0205] Taking into account all the interactions between the IMD, the walls of the natural lumen, and the microcatheter, it is necessary to model the longitudinal or "push / pull" compression applied by the practitioner via the implanted device during IMD placement. This model is very slow and not very robust in practice. Therefore, this modeling cannot be used in certain situations where rapid decision-making is required, such as in the sub-minute time it takes to obtain a 3D image of the vessel wall.

[0206] Alternative Example - Simulation of Deformation of a Laser Cut Stent

[0207] exist Figure 7 The following table shows the Figure 2 A different second example of a simulation method that can be performed by processing unit 20 .

[0208] In this example, the IMD to be simulated is of the "laser-cut stent" type. This example of an IMD does not have an overall spherical shape. It should be noted that the simulation method according to this second example can be used for all types of scalable IMDs, even though this article describes its use with a laser-cut stent type IMD.

[0209] Figure 8 A 3D view of a numerical IMD 2b is shown, consisting of segments 10 and nodes 11, which are generated to simulate the shape of a laser-cut stent. Preferably, the behavior of the mechanical elements attributed to this model is similar to that of the IMD 2a described above, except that rotation is not required to simulate the interactions between the segments (wires) of the numerical IMD. In reality, laser-cut stents are not braided implants, so the rotation model is not particularly relevant in this context.

[0210] The compression state in the implant tool and the mechanical equilibrium state in the artery make the longitudinal deformation greater than that described above. Figure 2 The first exemplary simulation method is more important.

[0211] Back to Figure 7According to the method, after obtaining the wall model 1 of the natural cavity (here, the artery) and the stent model 2b of the simulated deformation, the processing unit performs determination 300b of the intermediate deformation state of the system formed by the wall model 1 and the numerical IMD 2a.

[0212] In the same manner as in the previous example, the intermediate deformation state sought is a theoretical state of the IMD relative to the arterial wall, where the numerical IMD is completely contained within the wall model.

[0213] This intermediate deformation state is theoretical and does not require calculation of the mechanical interaction between the IMD and the wall.

[0214] The intermediate deformation states are obtained here via the following substeps:

[0215] - Position the numerical IMD at rest along the centerline inside the wall model,

[0216] - obtaining 311 a model of the implantation tool 3 (here a microcatheter), the tool model comprising in particular the tool surface 30,

[0217] - generation 312 of the value IMD 2b limited in the tool surface 30,

[0218] - Positioning 313 of the value IMD 2b in the constrained state in the wall model 1 in order to obtain the intermediate deformation state E2. The tool model itself is not located inside the wall model 1.

[0219] Next, similar to Figure 2 The simulation continues in the manner of the first exemplary simulation method, by deforming the numerical IMD to an intermediate deformation state, followed by relaxation of the mechanical stress experienced by the numerical IMD, until a mechanical equilibrium state is reached.

[0220] Figures 9a to 9e The acquisition of intermediate deformation states is shown.

[0221] exist Figure 9a In the embodiment, after the center line C of the artery is obtained in advance, the value IMD in state 2b-1 is positioned near the area to be treated, for example, at a positioning point along the center line C.

[0222] State 2b-1 corresponds to the IMD at rest, with no stress. The numerical IMD then intersects the wall model in multiple regions without considering the mechanical interaction between the IMD and the wall.

[0223] In sub-step 311, a microcatheter is generated in an initial state, and the length of the microcatheter is preferably substantially greater than the length of the value IMD in state 2b-1. It is preferred that the length of the catheter be greater than the value IMD because the IMD becomes longer when compressed in a microcatheter with a small diameter.

[0224] The simulated microcatheter is, for example, cylindrical in shape, and its radius is preferably smaller than the minimum radius of the natural cavity in the ROI.

[0225] It should be noted that the implant tool model is not necessarily generated by the processing unit 20, but can be retrieved in a database.

[0226] During sub-step 312, the numerical IMD is inserted into the model of the implantation tool (here, the microcatheter). First, the surface 30 of the microcatheter is expanded so that the microcatheter contains the numerical IMD in the rest state 2b-1. Figure 9b A microcatheter containing a numerical IMD is shown in FIG (separated from a wall model of the patient's natural lumen).

[0227] Next, the surface 30 is gradually retracted to return the microcatheter to its initial state, with the value IMD remaining contained within the microcatheter.

[0228] During the gradual retraction of the microcatheter surface, the contact interaction between the IMD and the microcatheter surface is resolved. Figure 9c The intermediate position shown is gradually compressed until the deformed state 2b-2 is reached. At the end of the series of calculations, the restricted state 2b-2 of the IMD is obtained, as shown in Figure 9d shown.

[0229] The numerical IMD in the restricted state 2b-2 is compressed in the tool surface 30 of the microcatheter. The numerical IMD in the restricted state 2b-2 is ready to be included in the wall model. On the other hand, the previously generated microcatheter is not included in the wall model.

[0230] In an advantageous alternative, the sub-step of calculating the restricted states of the IMD is performed only once for each IMD reference, "offline" upstream of the simulation. The restricted states of the IMD are stored in a database so that they can be reused later in the simulation to determine intermediate deformation states of the numerical IMD in the natural cavity.

[0231] The advantage of performing the calculation of the constrained state of the IMD "offline" is that the simulation time is greatly reduced, which increases the reactivity of the simulation and speeds up the potential selection of the implanted IMD reference.

[0232] In the case of an “offline” calculation, the sub-step 312 implemented for the simulation consists only in recovering in the database the numerical values ​​IMD that are in a restricted state in the tool surface.

[0233] Next, in sub-step 313 , the numerical IMD is integrated inside the wall model 1 in the reference frame of the wall model.

[0234] Advantageously, the value IMD 2b is deformed from its compressed state 2b-2 in the tool surface 30 during its positioning in step 313 so as to make it follow the centre line C of the artery with a curvilinear abscissa and its local reference system.

[0235] Therefore, the value IMD reaches the deformation state 2b-3.

[0236] By interpolating the numerical IMD in its state 2b-3 along the center line, an intermediate deformation state E2 of the numerical IMD is obtained, in which the numerical IMD is completely contained in the wall. This intermediate deformation state E2 is Figure 9e Shown in.

[0237] Advantageously, in order to align the numerical IMD along the centre line and obtain intermediate deformation states, the mechanical behaviour of the elements of the numerical IMD is not taken into account.At this stage, the transformation of the numerical IMD is only a geometric transformation.

[0238] Finally, in step 400 , starting from the intermediate deformation state E2 , the state of mechanical equilibrium E3 between the numerical value IMD 2 b and the wall model 1 can be calculated by the processing unit 20 in a robust and fast manner.

[0239] Taking into account the mechanical interaction between the numerical IMD and the wall model, a series of deformation states are calculated in an iterative manner until convergence to mechanical equilibrium. Here, multiple parts of the numerical IMD are deformed in a local orthogonal reference system R of the centerline C. The parts of the IMD are, for example, a series of longitudinal parts along the centerline.

[0240] The above and Figure 2 Methods related to the method of are advantageously used for this purpose: penalty methods for modeling the mechanical interaction between the IMD and the wall and / or co-rotational formulations of the deformation and rotation fields at the nodes of the numerical IMD, etc. However, other methods can also be used to numerically solve the contact interaction between the IMD and the wall.

[0241] exist Figure 9f The numerical IMD in state 2b-4 at which mechanical equilibrium is achieved is shown in FIG. The mechanical equilibrium state corresponds to the simulated deformation of the IMD after implantation in a natural cavity in a patient.

[0242] Calculation of the predicted isotopy of IMD on the wall

[0243] From the mechanical equilibrium state E3 obtained from the wall model 1 and the numerical IMD 2, it is advantageous to calculate the distance between each point of the numerical IMD and the surface of the wall. In particular, if a node and segment modeling is used for the numerical IMD, this distance can be obtained for a plurality of nodes of the numerical IMD, or even for all of the nodes.

[0244] In this regard, respectively Figure 2 and Figure 7 The simulation method shown in includes the step of calculating the predicted local isotopy of the IMD, including calculating the distance between each node and the wall model of the numerical IMD in the equilibrium state E3 obtained above.

[0245] The distance data thus calculated are advantageously illustrated by a graph representing the predicted local affinity of the implant to the walls of the natural cavity. One speaks of "local" affinity because this affinity is specific to each vertex of the numerical IMD.

[0246] For example, from a three-dimensional image of a wall model 1 and numerical IMD 2 in mechanical equilibrium state E3, different colors are associated with regions of the numerical IMD, and a parity map is obtained based on the parity of the nodes contained in these regions. Green is associated with IMD regions that are considered to be correctly parity, while red is used for regions with incorrect parity.

[0247] The threshold distance can be pre-recorded in the memory of the processing unit. Points (e.g., nodes) of the numerical IMD whose distance to the surface of the wall model is below the threshold distance are considered to be in contact with the wall model, corresponding to correct colocation. It should be understood that the distinction between correct colocation and incorrect colocation, and thus the coloring of the IMD region, depends on the pre-recorded or selected threshold distance.

[0248] Advantageously, the predicted local isotopy map of the IMD is displayed on a graphical interface provided by the display device 21. The practitioner can select a reference IMD to be implanted or confirm the selection of the IMD by noting the quality of the isotopy between the IMD and the natural cavity at the ROI.

[0249] Therefore, in Figure 10 An exemplary alignment map generated by simulating the deformation of a "laser-cut stent" type IMD is shown in . Good alignment is observed in the central region of the IMD. Regions near the ends of the IMD exhibit poor alignment; for this type of stent, it is undesirable that all surfaces of the IMD be aligned with the artery wall.

[0250] According to the above Figure 7 Related methods, the shape of the numerical IMD 2b-4 at mechanical equilibrium is obtained.

[0251] Figure 10 Also included is a view of the actual IMD 5 after dilation within the artery 4, superimposed with an isotopic map associated with the numerical IMD. This view is derived from a 3DRA image of the artery.

[0252] It should be noted that Figure 10 The numerical value of IMD 2b-4 corresponding to the simulation result is obtained in only 6 seconds. Therefore, the simulation method is very fast and robust.

[0253] Furthermore, the results of the simulated deformation of the IMD after implantation are very close to the actual clinical situation, with the ends of the model close to the points of the actual IMD 5 visible in the 3DRA image. This simulation method is very accurate.

[0254] Based on the predicted parity data for the deformed IMD at multiple points in the numerical IMD, an average predicted parity can be calculated. When multiple IMD references are derived from a set of references using simulations, the simulations allow the determination of the IMD reference with the highest average predicted parity. Practitioners can use this information to ultimately determine their most appropriate IMD reference.

[0255] However, the practitioner may make a selection based on other information generated by simulating deformation after implantation of the IMD. For example, the practitioner may disregard an IMD reference that predicts undesirable occlusion of an artery adjacent to the area to be treated.

Claims

1. A method for simulating deformation of an implantable medical device, referred to as an IMD, after implantation in a natural cavity from a three-dimensional numerical model of the wall of the natural cavity, the method comprising the following steps performed by a processing unit: Step i determines an intermediate deformation state (E2) of a numerical value IMD representing said IMD, the numerical value IMD in the intermediate deformation state deforming according to the shape of the wall model (1) while remaining contained in said shape, wherein The intermediate deformation state is determined based on contact interactions calculated between the three-dimensional vertices of the IMD and the three-dimensional vertices of the wall model, and wherein, during the determination of the intermediate deformation state (E2), the wall model (1) is geometrically deformed from an initial state so as to completely contain the numerical IMD in its rest state, and subsequently the wall model is restored to the initial state to obtain the intermediate deformation state of the numerical IMD, Step ii. calculating (400) a mechanical equilibrium state (E3) of the numerical value IMD from said intermediate deformation state (E2), comprising calculating the mechanical stress to which the numerical value IMD is subjected in the intermediate deformation state (E2), said mechanical stress being a function of the mechanical behavior of the numerical value IMD and of the mechanical behavior of the wall model (1), and including relaxation of said stress, wherein the mechanical behavior of the numerical IMD during determination of the intermediate deformation state is different from the mechanical behavior of the numerical IMD during calculation of the mechanical equilibrium, and / or the static state of the numerical IMD during determination of the intermediate deformation state is different from the static state of the numerical IMD during calculation of the mechanical equilibrium, wherein the calculated mechanical equilibrium state corresponds to the simulated deformation after implantation of the IMD.

2. The method according to claim 1, wherein The mechanical behavior of the wall model (1) used for calculating (400) the mechanical equilibrium state is a non-deformable rigid behavior.

3. The method according to claim 1, wherein Determining the intermediate deformation state includes: - obtaining (312) a numerical IMD constrained in a tool surface associated with a model of the implant tool, - Integrating (313) the constrained numerical IMD in the wall model (1) in order to obtain said intermediate deformation state.

4. The method according to claim 3, further comprising the step of determining the center line (C) of the natural cavity from the wall model (1), and wherein, The numerical IMD is deformed during its integration (313) to follow the center line.

5. The method according to any one of claims 1 to 4, wherein The numerical IMD comprises a plurality of segments (10), and further comprises a plurality of nodes (11), each node (11) connecting ends of two consecutive segments (10).

6. The method according to claim 5, wherein: The mechanical behavior of at least one segment (10) corresponds to that of a beam of cylindrical shape.

7. The method according to claim 6, wherein: During the determination of the intermediate deformation state, at least one segment (10) having a numerical IMD of the beam mechanical behavior is modeled with a first diameter, and / or with a first thickness, and / or with a first elastic modulus, and / or with a first slenderness ratio coefficient, and / or with a first radius of gyration, and / or with a first set of critical buckling loads, And wherein the segment (10) is modeled with a different second diameter, and / or a different second thickness, and / or a different second elastic modulus, and / or a different second slenderness ratio, and / or a different second radii of gyration, and / or a different second set of critical instability loads during calculation (400) of the mechanical equilibrium state.

8. The method according to claim 6, wherein: The mechanical behavior of at least one node (11) corresponds to the behavior of a rotating body.

9. The method according to claim 6, wherein: The calculation (400) of the mechanical equilibrium state (E3) of the numerical IMD comprises calculating, in a three-dimensional reference system linked to the wall model, for each node i of the numerical IMD the displacement fields Dxi, Dyi, Dzi and the rotation fields Rxi, Ryi, Rzi, the two fields being calculated by applying basic dynamic principles at the nodes.

10. The method according to claim 6, wherein: The calculation (400) of the mechanical equilibrium state (E3) of the numerical IMD includes, for at least one node of the numerical IMD, calculating the normal force and / or friction force exerted by the wall model on the node, and modeling the penetration resistance of the wall model and the friction force between the numerical IMD and the wall model, respectively.

11. The method according to any one of claims 6 to 10, wherein The segments (10) and nodes (11) of the numerical IMD have at least one end point where the overall shape of the numerical IMD flattens.

12. The method according to claim 11, wherein The end extreme points are modeled with a first concavity during determination of the intermediate deformation state, and the end extreme points are modeled with a second concavity different from the first concavity during calculation (400) of the mechanical equilibrium state.

13. The method according to any one of claims 1 to 4, wherein The numerical IMD is a model of the intracapsular cage.

14. The method according to any one of claims 1 to 4, wherein The numerical IMD is a model of a laser-cut stent.

15. Method according to any one of claims 1 to 4, comprising a subsequent step iii of calculating the predicted local isotopy of at least a portion of the three-dimensional vertices of the numerical IMD on the wall model (1).

16. The method according to claim 15, comprising calculating the local co-location of a plurality of nodes (11) of the numerical IMD on the wall model.

17. The method according to claim 15, wherein: The numerical value IMD corresponds to an IMD reference derived from a set of IMD references recorded in a database (DB2), steps i, ii and iii being repeated for each reference in the set of IMD references.

18. A computer program product comprising code instructions for implementing the method according to any one of claims 1 to 17 when the code instructions are executed by a processing unit.

19. A processing unit (20), comprising: - a device for obtaining a three-dimensional wall model of a natural cavity, - means for obtaining a numerical IMD, configured to generate said numerical IMD from an IMD reference derived from a database (DB2), - calculation means configured to determine an intermediate deformation state (E2) in which said numerical value IMD is deformed according to the shape of the wall model while remaining contained in said shape, The calculation device is further configured to calculate a mechanical equilibrium state (E3) of the numerical IMD based on the mechanical behavior of the numerical IMD and the mechanical behavior of the wall model, The processing unit is configured to implement the method according to any one of claims 1 to 17.

Images