System for assisting in localizing the source of an arrhythmia

Abstract

Title: System for assisting in locating a source of an arrhythmia System (S) configured to: determine (DV) a value of at least one parameter of a parametric model modeling a propagation of an electrical signal of cardiac activity, the value of the parameter being determined from a position (Pi) of at least one stimulation (Si) of the individual's heart and from an electrical response signal (SRi) to the stimulation (Si) and from an anatomical model (MA), determine (DPE) an estimated position (PE) of the source of the arrhythmia from the parametric model, the anatomical model (MA) and a first electrical signal of the arrhythmia (SA) representative of the cardiac activity of the individual and generated spontaneously by the source of the arrhythmia, calculate (CE) an error (EP) affecting the estimated position (PE) of the source of the arrhythmia from the position (Pi) and the electrical response signal (SRi). Figure for abstract: Fig. 3

Description

Title of the invention: System for assisting in locating the source of an arrhythmia Field of the invention

[0001] The field of the invention is that of assistance in locating sources of arrhythmias.

[0002] The invention applies to arrhythmias having a focal source such as extrasystoles and ectopic foci in the atria in general.

[0003] A goal of the practitioner is to precisely locate an arrhythmia, for example an extrasystole, within the heart of an individual in order, for example, to perform an ablation.

[0004] Invasive solutions are known for aiding in the localization of arrhythmias. European patent EP2848191B1 discloses an invasive method for locating a source of an arrhythmia based on mapping by stimulation comprising cardiac stimulation and correlation, called "pace mapped", of an electrical signal acquired during stimulation with an electrical signal recorded during the arrhythmia. The estimated location of the source of the arrhythmia is that which presents the strongest correlation with the signal corresponding to the arrhythmia. However, this type of solution has the disadvantage of requiring a large number of stimulations to precisely determine the source of the arrhythmia. Estimating the position of the source of the arrhythmia by this type of solution is therefore relatively long, expensive and risky for the patient.

[0005] Non-invasive solutions for estimating the position of an arrhythmia source are also known using an inverse problem solving method to locate an arrhythmia from the measurement of an electrical signal measured during an arrhythmia, for example an electrocardiogram measured on the surface of an individual's torso. These solutions have the disadvantage of being inaccurate. Indeed, small imperfections in the values of the electrical signal can significantly affect the result of the inverse problem solving.

[0006] An aim of the invention is to limit at least one of the aforementioned drawbacks.

[0007] To this end, the invention relates to a system for assisting in locating a source of an arrhythmia affecting the heart of an individual, the system comprising a processing unit configured to implement, by computer, a method for assisting in locating a source of an arrhythmia comprising: - generate a cardiac and thoracic anatomical model of the individual from at least one image of the individual, - determining a value of at least one parameter of a parametric model modeling a propagation of an electrical signal of cardiac activity, the value of the parameter being determined from a position of at least one stimulation of the individual's heart and from an electrical response signal representative of the individual's cardiac activity and measured in response to the stimulation and from the anatomical model, - determine an estimated position of the source of the arrhythmia from the parametric model, the anatomical model and a first electrical signal of the arrhythmia representative of the cardiac activity of the individual and generated spontaneously by the source of the arrhythmia, - calculate a positioning error affecting the estimated position of the arrhythmia source from the and the response electrical signal, - generate information representative of the estimated position and information representative of the positioning error affecting the estimated position so that the system returns this information to a practitioner.

[0008] Advantageously, the determination of the value of the parameter comprises: - generate a first estimate of a value of a first characteristic of the cardiac activity of the individual using an initial parametric model, distinct from the parametric model in that it presents an initial value of the parameter distinct from the value, the anatomical model, and an initial data taken from a position or an electrical signal representative of the cardiac activity of the individual, - Determine the value of the parameter from the first estimate and a reference value of a second characteristic of the cardiac activity resulting from a measurement taken from the position of at least one stimulation and the electrical signal of response to the stimulation.

[0009] In a first embodiment, the initial data is a position.

[0010] Advantageously, the determination of the estimated position of the source of the arrhythmia comprises: - calculate J simulated electrical response signals using the parametric model, the anatomical model and J so-called predetermined simulated positions of the arrhythmia source, - select a simulated position from the J simulated positions, from the electrical signal of the arrhythmia.

[0011] Advantageously, the calculation of the positioning error affecting the estimated position of the source of the arrhythmia comprises the selection, for at least one stimulation, of a simulated position taken from among the J simulated positions from the electrical signal of response to the stimulation.

[0012] Advantageously, the determination of the value of the parameter comprises, for at least one stimulation: - select a candidate value of the parameter, associated with a value of a third characteristic of the cardiac activity closest to a first reference value of the third characteristic from the electrical signal responding to the stimulation in a correspondence table, - calculate for each of J so-called simulated positions, a candidate estimate of a simulated electrical response signal associated with a source located at the simulated position from a candidate parametric model in which the parameter has the candidate value, - select a candidate estimate from among the J candidate estimates.

[0013] Advantageously, the method for assisting in locating the source of the arrhythmia comprises: - calculate J initial signals being simulated electrical signals associated with sources occupying the respective simulated positions using a starting parametric model and the anatomical model, - modifying a value of at least a first parameter of the initial parametric model so as to generate the initial parametric model, from a second electrical signal of the arrhythmia affecting the individual.

[0014] Advantageously, the parameter comprises a propagation speed of an eikonal model or a characteristic time constant of an ionic model.

[0015] In a second embodiment, the initial data is an electrical signal representative of the cardiac activity of the individual.

[0016] Advantageously, at least one parameter is a regularization parameter of the inverse problem.

[0017] Advantageously, the method for assisting in locating the source of the arrhythmia comprises, for at least one stimulation: - generating a plurality of estimated positions of the stimulation from the response signal and the anatomical model by separate so-called inverse methods differing in that they implement separate methods for solving the inverse problem and / or separate methods for calculating an activation map from electrograms, - select a reverse process from among these processes.

[0018] Advantageously, the system comprises: - an imager configured to generate the image, - a stimulator to carry out the stimulations, - a recorder configured to record electrical signals responding to stimulation, - a locator configured to measure the positions of the stimuli so as to measure the positions.

[0019] Advantageously, the at least one image of the individual is acquired during a predetermined phase of the individual's respiratory cycle and the stimulation is carried out during said phase of another respiratory cycle of the individual.

[0020] Advantageously, the system comprises a respiration recorder intended to deliver measurements making it possible to monitor the patient's respiration and a stimulator for carrying out stimulation, in which the processing unit is configured to impose, from a measurement delivered by the respiration recorder, that the stimulator carries out the stimulation only during a predetermined phase of a respiratory cycle of the patient being the phase during which the image was acquired.

[0021] Advantageously, the system comprises a human-machine interface comprising a display, the system being configured to simultaneously display on the display: - a graphic representation of at least part of the individual's heart or myocardium, - a visual indicator occupying a position, on the graphic representation of at least one part of the heart or myocardium of the individual, corresponding to the estimated position, - a graphical representation of an area, around the visual indicator, containing the points of the heart or myocardium located at a distance from the estimated position less than or equal to the positioning error.

[0022] The invention also relates to a computer program product comprising instructions which, when the program is executed by a computer, cause the latter to implement the method for assisting in the localization of arrhythmias.

[0023] The invention also relates to a computer-readable medium comprising instructions which, when executed by a computer, cause the latter to implement the method for assisting in locating arrhythmias. Brief description of the figures

[0024] Other characteristics and advantages of the invention will emerge on reading the detailed description which follows, with reference to the appended figures, which illustrate:

[0025] [Fig-1]: an example of an embodiment of a system according to the invention,

[0026] [Fig.2]: a representation of stimulations,

[0027] [Fig.3]: a flowchart of the steps implemented by the system according to the invention,

[0028] [Fig.4]: an example of displaying position information and error information,

[0029] [Fig.5]: a flowchart of the steps implemented by the system according to an example of a first embodiment,

[0030] [Fig.6]: a flowchart of the steps implemented by the system according to an example of a second embodiment. Description of the invention

[0031] The present invention relates to a system S for assisting in locating a source of a predetermined arrhythmia affecting the heart of an individual, an example of which is shown in [Fig.l]. This system S is configured to assist a practitioner in locating a source of an arrhythmia, in particular during a surgical intervention, for example with a view to an ablation.

[0032] The arrhythmia is predetermined.

[0033] Arrhythmia is an arrhythmia having a focal source.

[0034] The focal source is the position from which electrical activity is triggered.

[0035] This electrical activity triggers or modifies a cardiac cycle (including depolarization and repolarization) or modifies the cardiac cycle.

[0036] For example, when the source is located in the atria it modifies the called cardiac cycle and when it is located in the ventricle it triggers a cardiac cycle.

[0037] This information can be recognized on an electrocardiogram.

[0038] An example of this type of arrhythmia is, for example, an extrasystole. The source of the extrasystole is then an ectopic focus, that is to say located outside the sinus node.

[0039] Another non-limiting example of this type of arrhythmia is focal atrial tachycardia. System

[0040] As visible in [Fig.l], the system S comprises a processing system T comprising a processing unit UT and a human-machine interface INT.

[0041] The system S also advantageously comprises a STIM stimulator for stimulating the patient's heart. This STIM stimulator comprises, for example, a catheter provided with a stimulation electrode capable of stimulating the patient's heart in a stimulation position P; defined by the practitioner and an exciter for generating an electrical excitation pulse at the electrode so as to carry out the stimulation Si at the position P;.

[0042] The stimulation Si is performed so as to trigger a cardiac cycle or to modify the cardiac cycle as well as the source of the arrhythmia.

[0043] In other words, the stimulation S; is carried out to trigger the arrhythmia from a source of predetermined position P;.

[0044] The system advantageously comprises an ENRS recorder for recording the cardiac activity of the individual's heart.

[0045] This recorder is an electrocardiograph. It comprises electrodes capable of being positioned at fixed measurable positions on the surface of the individual's torso.

[0046] The ENRS recorder comprises, for example, a vest incorporating these electrodes. Alternatively, the electrodes may be placed directly on the patient's skin and attached to the patient's torso by fastening means.

[0047] The recorder is capable of measuring one or more derivations.

[0048] The recorder is conventionally capable of measuring 12, 56, 128 or 256 leads, but these numbers are not limiting.

[0049] The ENRS recorder delivers an electrical signal representative of the individual's cardiac activity which may be a response to stimulation or an electrical signal generated spontaneously by the source of the arrhythmia affecting the patient.

[0050] This signal is composed of one or more derivations.

[0051] The electrical signal representative of cardiac activity which is of interest in the patent application corresponds here to a cardiac cycle, that is to say a portion of each derivation comprising a QRS complex.

[0052] The electrical signal preferably extends from a P wave to a T or U wave.

[0053] The processing unit UT is advantageously configured to extract a cardiac cycle from each electrical signal delivered by the recorder so as to generate the electrical signal representative of the cardiac activity of the individual that the processing unit UT uses to implement the location assistance method that we describe in the remainder of the document.

[0054] The system S advantageously comprises an imager IM intended to deliver a set of at least one image I, of the patient such as to allow a processing unit UT of the system to generate a cardiac and thoracic anatomical model of the heart.

[0055] The set of images comprises for example at least one image obtained by tomodensimeter also called CT-scan in English terminology, CT being the acronym for “Computed tomography” and / or an image obtained by magnetic resonance.

[0056] The IM imager therefore comprises a scanner and / or a magnetic resonance imaging device.

[0057] The system S also comprises a locator LOC configured to deliver a position measurement which will be called position P; of a stimulation Si.

[0058] This is conventionally achieved by X-ray imaging of the pacemaker catheter and processing the X-ray image to determine the position P of the stimulation Si on the individual's heart.

[0059] Advantageously, the system comprises a scanner for this purpose, for example the scanner of the IM imager.

[0060] The position P; of the stimulation is a three-dimensional position on the patient's heart.

[0061] The processing system ST comprises a human-machine interface INT comprising an output interface INTS.

[0062] Advantageously, the system S comprises an ENRR recorder of the patient's breathing delivering measurements making it possible to monitor the patient's breathing from which it is possible to deduce the phase of the respiratory cycle in which the patient is located, as we will see later.

[0063] This recorder may comprise means for measuring impedance between electrodes of the ENRS recorder and calculation means generating measurements representative of the phase of the respiratory cycle in which the patient is located as a function of time.

[0064] This allows the processing unit to identify the phase of the respiratory cycle in which the patient is at a given time.

[0065] The invention relates to a global method for assisting in locating the source of the arrhythmia comprising the following preliminary steps: - acquire at least one image I by the imager, - possibly acquire measurements from which the phase of the respiratory cycle in which the patient is located can be deduced, - carry out one or more stimulations S,cn of the positions P; of the heart, i = 1 to P with P an integer greater than or equal to 1, as represented in [Fig.2] for P = 4, - acquire the response signal SR; at each stimulation Si, as represented in [Fig.2], during a surgical phase, - acquire a first electrical signal called “arrhythmia” SA representative of the cardiac activity of the individual generated spontaneously in response to an excitation generated by the source of the arrhythmia affecting the individual, during the surgical phase, - optionally, acquire a second electrical signal of the SA2 arrhythmia representative of the cardiac activity of the individual generated in response to an excitation generated by the source of the arrhythmia affecting the individual.

[0066] These last three steps advantageously include the acquisition of signals by the ENRS recorder and the extraction of the electrical response signals from these signals by the UT processing unit.

[0067] Thus the method advantageously comprises the reception of the other data acquired or measured during this prior step, by the processing unit T.

[0068] Advantageously, the first electrical signal of the SA arrhythmia is measured in the surgical phase. Alternatively, this electrical signal is acquired prior to the surgical phase.

[0069] Advantageously, the second signal SA2 is acquired prior to the surgical phase.

[0070] An aim of the invention is to help a practitioner identify the position of the source of the arrhythmia in the individual.

[0071] The processing system comprises a processing unit UT configured to implement a method for assisting in locating a source of the arrhythmia. The invention also relates to the method for assisting in locating the source of the arrhythmia and to the overall method comprising the method for assisting in locating the source of the arrhythmia.

[0072] The processing unit UT uses the following input data to implement the method: - the position P; and the electrical response signal SR; of each stimulation Si, - the set of at least one image I of the individual, - the first SA and possibly the second SA2 signal of the arrhythmia, - possibly measures allowing the phase of the respiratory cycle in which the patient is located to be deduced.

[0073] A flowchart of the process is shown in [Fig.2].

[0074] The method for assisting in locating a source of the arrhythmia comprises: - generate GA an anatomical model MA cardiac and thoracic of the individual from at least one image, - determine DV a value of at least one parameter of a parametric model modeling a propagation of an electrical signal of a cardiac activity, the value of the parameter being determined from a position P; of at least one stimulation Si of the heart of the individual and of an electrical response signal SR; representative of the cardiac activity of the individual and measured in response to the stimulation Si and from the anatomical model MA, - determine DPE an estimated position PE of the source of the arrhythmia from the parametric model, the anatomical model MA and a first signal electrical signal of SA arrhythmia representative of the individual's cardiac activity and generated spontaneously by the source of the arrhythmia, - calculate CE a positioning error EP affecting the estimated position PE of the arrhythmia source from the position P; and the electrical response signal SR;, - generate information representative of the estimated position PE and information representative of the positioning error EP affecting the estimated position.

[0075] The generation of information by the processing unit UT is carried out so that the system S returns, during a restitution step, this information to a practitioner.

[0076] Advantageously, the system S is configured to return to the practitioner the information representative of the estimated position PE and of the positioning error EP on the output interface INTS.

[0077] This step is advantageously a step of displaying information representative of the estimated position PE and of the positioning error EP on a set of at least one screen of the output interface INTS.

[0078] This step is a step of the overall process.

[0079] Advantageously, this information is generated so that the estimated position PE and the positioning error PE are represented at least graphically on the screen.

[0080] In an exemplary embodiment, the information is displayed as shown in [Fig.3].

[0081] Advantageously, the generation step is such that during the display step, the output interface displays, in a superimposed manner: - a graphic representation of a part of the heart or part of the individual's myocardium MC, - a visual indicator INDP occupying a position, on the graphic representation of T at least a part of the heart or myocardium of the individual, corresponding to the estimated position PE, - a graphical representation of a zone Z, around the visual indicator, containing the points of the heart or myocardium MC located at a distance from the estimated position PE less than or equal to the positioning error EP.

[0082] Advantageously, this information is displayed simultaneously.

[0083] The graphical representation of the zone Z comprises for example a graphical representation of the limit L of this positioning zone Z.

[0084] Alternatively or in addition, the points located in zone Z are colored differently from the other points of the myocardium.

[0085] The graphical representation of the part of the heart or myocardium is, advantageously, a two-dimensional or three-dimensional geometric representation

[0086] The invention uses an invasive technique to estimate a position of a source of the arrhythmia and an error zone around this position, which helps a surgeon to locate this source on the heart of a patient, during surgery. The idea is to stimulate the patient's heart at several measured positions in order to interactively construct a parametric calculation method best suited to the patient (one that limits an error between the estimated position of the source and an actual position of a source in the heart of the patient being treated) which makes it possible to generate an estimate of the position of the source of an arrhythmia accurately and to determine a reduced error zone around this position. The display of this information allows the practitioner to refine the determination of the source of the arrhythmia during surgery, which ultimately makes it possible to precisely target a treatment such as an ablation.

[0087] Furthermore, this method makes it possible to inform the surgeon of an error or a personalized degree of confidence, i.e. corresponding to the patient and his condition in the operating room (respiratory rate, position of the patient) and therefore to give a region of confidence or error making it possible to alert the practitioner in the event of a significant error on the estimated position of the localization of the arrhythmia, which is of great help to the surgeon when searching for the source of the arrhythmia in surgery. Indeed, this allows him not to focus unnecessarily on an erroneous estimated position and, ultimately, not to waste time.

[0088] The MA thoracic and cardiac anatomical model includes a geometric model of the individual's heart.

[0089] This geometric model of the individual's heart comprises, for example, a three-dimensional cloud or mesh of points of the heart.

[0090] This model advantageously includes points of the epicardium of the endocardium and advantageously points of the myocardium located between these two surfaces.

[0091] The mesh advantageously comprises points forming the vertices of the mesh connected by cells.

[0092] The thoracic anatomical model is defined so as to allow the electrodes of the recorder of electrical signals representative of cardiac activity in response to stimulation to be positioned relative to the heart.

[0093] The anatomical model advantageously comprises a geometric model of the surface of the torso and / or the lungs and / or the volume of the torso.

[0094] The processing unit UT is, for example, configured to segment the whole of at least one image I to extract the heart of the image and the positions of the electrodes.

[0095] Advantageously, the anatomical model MA comprises labels assigned to areas of the geometric model(s) representative of the nature of the tissues in these areas.

[0096] These areas include, for example, the cells of the mesh.

[0097] Information on the nature of the tissues is, for example, taken from: damaged tissue, healthy tissue, degree of fibrosis of the damaged tissue, adipose tissue, fluid in the pericardium, orientation of the fibers.

[0098] The processing unit UT is advantageously configured to extract information on the nature of the tissues from at least one image of the patient obtained by magnetic resonance and to assign labels to areas of at least one geometric model representative of the information extracted by the processing unit for these areas.

[0099] Advantageously, the number P of arrhythmia stimulations is between 1 and 10. The stimulations are used to construct the model interactively, but they must not be too numerous to limit the risks for the patient as well as the time of the intervention.

[0100] Advantageously, the step DV of determining the value of the parameter advantageously comprises at least one elementary determination step comprising: - generate GE a first estimate EG1 of a value of a first characteristic of the cardiac activity of the individual using an initial parametric model MOP, distinct from the parametric model MOPA in that it presents an initial value of the parameter distinct from the value, the anatomical model MA, and an initial data DI taken from a position or an electrical signal representative of the cardiac activity of the individual, - determine GMA the value of the parameter from the first estimate and a reference value of a second characteristic (identical or distinct from the first characteristic) of the cardiac activity taken, this reference value coming from a measurement taken from the position P; from at least one stimulation Si and the electrical signal SR; from the response to the stimulation Si.

[0101] This implementation makes it possible to interactively adapt a previously parameterized parametric model based on the measurements made during the stimulations. Modifying the parameter values makes it possible to improve the quality of the model.

[0102] The parametric model is a given model which comprises a certain number of parameters to which values are assigned.

[0103] When a new value of a parameter of a model is determined, the resulting model differs from the initial parametric model only by the value of the parameter.

[0104] Advantageously, the value of the parameter is defined so as to reduce a deviation function, that is to say so as to reduce an error.

[0105] In other words, the value of the parameter is defined so that a difference between a second deviation defined for at least one stimulation between a reference value of the second characteristic resulting from the measurement and a second estimate of the value of the second characteristic of the cardiac activity obtained from the MOPA parametric model is smaller than a first deviation obtained for this stimulation between a first estimate of the second characteristic determined from the parametric model and the reference value.

[0106] The step DV of determining the value of the parameter may comprise an elementary step of parameter determination as defined previously or a series of elementary steps of parameter determination, such that the initial parametric model MOP of a second elementary step following a first elementary step is the parametric model MOPA generated during the first elementary step.

[0107] The initial parametric model MOP of the first elementary step implemented is an input to the process or is generated during the process from another model as we will see later.

[0108] The first and second characteristics may be identical or different in the different elementary steps.

[0109] The parameter can be a single parameter or be composed of several elementary parameters.

[0110] We will detail the different steps implemented by the calculation unit by describing two embodiments of the invention. The characteristics previously described are applicable to these two embodiments. First embodiment#: direct model

[0111] According to a first embodiment, a direct global parametric model is used modeling the propagation of an electrical signal from a source of an arrhythmia to the individual's torso.

[0112] In other words, the processing unit UT is configured to implement a step during which it will solve the direct problem of electrocardiography during the method of assisting in locating the arrhythmia.

[0113] The result of this step is used to determine the positioning error and / or the estimated position.

[0114] This embodiment has the particular advantage of taking into account spatiotemporal coherence.

[0115] With reference to [Fig.5], we will describe more precisely an example of the steps implemented by the processing unit in the first embodiment.

[0116] In this embodiment, the initial data used during the GE generation step is a position.

[0117] In this first embodiment, the processing unit UT is configured to implement, by computer, the following steps: - generate GA the thoracic anatomical model MA from image I, - determine DV1 a value of at least one parameter of a parametric model MOP1 modeling a propagation of an electrical signal of cardiac activity, - determine DPE1 an estimated position PE1 of the source of the arrhythmia from the parametric model, the anatomical model MA and a first electrical signal of the arrhythmia SA representative of the cardiac activity of the individual generated spontaneously in response to an excitation generated by the source of the arrhythmia affecting the individual, - calculate CEI a positioning error EP affecting the estimated position PE1 of the arrhythmia source from the position P; and the electrical response signal SR;, - generate RES information representative of the estimated position PE1 and information representative of the positioning error EPI affecting the estimated position.

[0118] Advantageously, the determination step DV 1 comprises at least one elementary step comprising: - for at least one stimulation S;, generate GE1 a first estimate ES R; of a value of at least one first characteristic of the cardiac activity of the individual using an initial parametric model MPO1 of which a parameter has an initial value, the anatomical model MA and the position P; of the stimulation S;, - define GMA1 the value of the parameter from the first ESR estimate; and from a reference value of the first characteristic, this reference value coming from the electrical response signal SR;, so as to obtain the parametric model MOPA1 differing from the initial parametric model MOP1 only by the value of the parameter.

[0119] This last step is advantageously carried out in such a way as to limit an error.

[0120] In the embodiment example of [Fig.5], an estimate is made, during step GE1, of the electrical response signal ESR; to stimulation Si.

[0121] The MOP1 model used to calculate this estimate is a global parametric model.

[0122] The global parametric model comprises, for example, several elementary parametric models.

[0123] The global parametric model includes for example an eikonal model and an ionic model and a direct problem resolution model.

[0124] These models each include at least one parameter.

[0125] The processing unit UT is, for example, configured to implement, during the generation step GE1, for each stimulation Si, with i = 1 to P the following sequence of steps, called the sequence of steps of the direct problem in the rest of the text: - generate an activation map from the eikonal model, the position P; and the anatomical model MA, - generate a time and space distribution of transmembrane potentials from the activation map, the ionic model and the MA anatomical model, - generate electrical potentials on the surface of the torso from transmembrane potentials using a direct problem solving model and the MA anatomical model, - calculate the estimated electrical signal ESR; from the electrical potentials on the surface of the torso and the anatomical model MA.

[0126] The step of generating electrical potentials on the surface of the torso from the transmembrane potentials and the anatomical model consists of solving the so-called direct problem of electrocardiography.

[0127] During each of the steps, a model is solved for a geometry defined by anatomical model MA starting from the position P; (to generate the activation map), from the activation map (to generate the transmembrane potentials), from the transmembrane potentials (to generate the electrical potentials on the surface of the torso) so as to generate the result of this step.

[0128] In the initial parametric model M0P1, the parameter(s) of the method have predetermined values called initial values.

[0129] As is known per se, the eikonal model or eikonal equation describes the propagation of an electrical stimulus from a position located on the cardiac tissue in terms of isochronous surfaces.

[0130] The following eikonal equation describes the propagation of an electrical activation front generated at a position P through a non-homogeneous and anisotropic continuum: Tact(x) is the instant at which the activation front reaches the point x and F (x) is the characteristic speed of the wavefront propagation in the medium. TM Tact = 0 on P = 1 on o)h

[0131] D is a second-order diffusion tensor that takes into account the anisotropy of propagation.

[0132] This equation is defined in the article “A conduction velocity adapted eikonal model for electrophysiology problems with re-excitability evaluation” Cesare Corradoa, Nejib Zemzemi, Medical Image Analysis, Volume 43, January 2018, Pages 186-197.

[0133] Thus, the eikonal model describes the propagation of the activation front in the core for a predetermined front propagation speed.

[0134] A first parameter of the eikonal model is the speed F(x) of propagation of the activation front.

[0135] A second parameter of the eikonal model is a diffusion tensor D which takes into account the anisotropy of the propagation of electrical potentials in the core.

[0136] The calculation of the activation map is carried out using a numerical method for solving the eikonal model.

[0137] For example, we know a fast marching resolution method also called FMM, an acronym for the English expression “fast marching methods” or Dijkstra’s algorithm.

[0138] The activation map comprises activation times of a plurality of points of the anatomical model MA defined in relation to the time of stimulation Si, that is to say the time at which the stimulation is carried out at the position P;.

[0139] The calculation of the transmembrane potentials at different times is carried out by a method of numerical resolution of the ionic model making it possible to obtain, from the activation map, a set of transmembrane potentials or action potentials at different times, for points of the mesh of the anatomical model.

[0140] The ionic model is, for example a Mitchell and Schaeffer model described in the article “A Two-Current Model for the Dynamics of Cardiac Membrane, Colleen v. Mitchell and Cavid g. Schaeffer”, Department of Mathematics, Duke University and Center for Nonlinear and Complex System, Bulletin of Mathematical Biology (2003) 65, 767-793.

[0141] The Mitchell and Schaeffer model includes in particular the following parameters: rin , roui, ropen and Tclose which are four time constants each characterizing one of the four respective phases of the action potential. Tin is the time constant characteristic of the depolarization phase in the transmembrane potential, Tout is the time constant characteristic of the plateau phase in the potential transmembrane, T0Pen is the characteristic time constant of the repolarization phase in the transmembrane potential and rclose is the characteristic time constant of return to the initial condition of the electrical state of the heart.

[0142] The values of these time constants affect the duration of the QRS complex and the shape of the response signal.

[0143] Other ionic models can be used. In particular, the Beeler-Reute model or the Noble model are known. These models include at least one parameter such as an ion channel conductance.

[0144] The processing unit UT is configured to use, for example, a method using an implicit, explicit or semi-implicit numerical scheme, for example of the Euler or Runge Kutta type to calculate the transmembrane potentials from the activation map.

[0145] Calculating electrical potentials at the surface of the torso from transmembrane potentials involves solving the direct problem of electrocardiography.

[0146] In a manner known per se, the direct problem is expressed in the following matrix manner:

[0147] Aq>H = q>T

[0148] where q>H and q>T are respectively the electrical potential on the epicardial surface and on the thoracic surface and A is the transfer matrix.

[0149] The calculation of the electrical potentials at the surface of the torso from the transmembrane potentials consists of solving the direct problem of electrocardiography by a boundary element resolution method also called BEM (acronym for the Anglo-Saxon expression "boundary element methods"), the method of fundamental solutions also called MFS (acronym for the Anglo-Saxon expression "method of fundamental solutions"), the finite element method also called FEM (acronym for the Anglo-Saxon expression "method of fundamental solution").

[0150] These resolution methods use a parametric model of the conductivity of the torso comprising a parameter such as a single conductivity of the torso or several conductivities associated with different areas of the torso and therefore of the anatomical model.

[0151] The calculation of the electrical response signals from the electrical potentials on the surface of the torso is carried out by calculating, by methods known to those skilled in the art, the derivation(s) from the positions of the electrodes of the electrocardiograph on the torso, the anatomical model MA and the electrical potentials on the surface of the torso.

[0152] Thus, the processing unit UT is, for example, configured to implement the series of steps of the direct problem using a global parametric model in which the parameters have respective initial values so as to form the initial parametric model M0P1.

[0153] This makes it possible to obtain first ESR estimates; SR response signals; to stimulations.

[0154] The processing unit UT is advantageously configured to calculate deviations between the estimates of the ESR response signals; to the stimulations and the SR response signals; themselves.

[0155] These deviations are, for example, combinations of the inverses of correlation coefficients between the different derivations or of the inverses of errors in the sense of least squares calculated for the different derivations.

[0156] The calculation unit is advantageously configured to determine a value of a parameter of the model, for example a new value of a propagation speed of the eikonal model, such as a value of a function of the deviations between the signals estimated using the initial parametric model M0P1 and the response signals SR; has a higher value than the value of this function calculated for the deviations obtained between the estimated signals, using the new parametric model MOPA differing from the initial parametric model MOP1 only by the value of the propagation speed, and the response signals SR;.

[0157] The function is for example a combination, for example, an average. Alternatively, the function is a maximum of the deviations.

[0158] This step can, for example, be carried out by calculating estimated signals from several parametric models differing only by the values of the propagation speed, by calculating, for each of these values of the propagation speed, the deviations between the estimated signals and the electrical response signals and by calculating the value of the function of these deviations and by comparing the values of the functions obtained for the different values of the propagation speed.

[0159] Alternatively or in addition, during the step GE1 of generating a first estimate of a value of at least one first characteristic of the cardiac activity, it is possible to estimate a value of a characteristic of the cardiac activity distinct from an electrical signal of response to a stimulation.

[0160] In one example, the processing unit UT can be configured to calculate, for each stimulation Si, an estimate of the electrical response signal SR; then to extract values of at least one characteristic of this signal.

[0161] For example, the duration of the QRS complex can be deduced from the estimate of the electrical response signal.

[0162] In another example, the processing unit UT is configured to implement, during this step, only part of the sequence of steps of the direct problem so as to generate an intermediate result and to deduce from this result a value of the characteristic of the electrical response signal of this intermediate result.

[0163] For example, the processing unit UT is configured to generate, for at least one stimulation Si, an activation map from the eikonal model, the position P; of the stimulation and the anatomical model MA and to extract from the activation map a duration of the QRS complex of the electrical response signal SR.

[0164] This extraction is, for example, carried out by determining activation durations at different points of the activation map and by identifying the maximum activation duration.

[0165] The maximum activation duration corresponds to the duration of the QRS complex of an electrical response signal estimated by the parametric method.

[0166] Alternatively or additionally, the processing unit UT is configured to generate an activation map from the eikonal model and to generate a time and space distribution of transmembrane potentials from the activation map, the ionic model and the anatomical model MA.

[0167] The processing unit UT can be configured to extract, from the transmembrane potentials, values of several characteristics, for example of at least one time constant.

[0168] An example of a characteristic is a duration of the action potential at x% repolarization, x is between 1 and 100, for example it is equal to 90.

[0169] Classically, this information is proportional to temporal information that can be extracted from transmembrane potentials.

[0170] The processing unit UT is advantageously configured to extract, from the signal SR;, the reference value of the second characteristic which may be the first characteristic.

[0171] For example, it extracts the duration of the QRS complex from the SR response signal;.

[0172] In the case where the characteristic is the duration of the QRS complex, we determine, for each i with i = 1 to P, a difference between the first estimate of the duration of the QRS complex and the duration extracted from the SR response signal;.

[0173] During step GMA1 of defining the value of the parameter, the processing unit UT uses the calculated deviations to define the value of the parameter.

[0174] It determines for example a function is, for example, a combination or an average of the deviations carried out on the different i with i = 1 to N or a maximum of the deviations.

[0175] For example, the value of the propagation speed of the eikonal model is updated so that the average taken over the different stimulations S;, of the differences between the durations of the QRS complex calculated from the activation maps generated with the new propagation speed and the durations of the QRS complex extracted from the measurements of the electrical response signals SR; is lower than the average taken on the different S; stimulations, differences between the QRS complex durations calculated from the activation maps generated with the initial propagation velocity (before updating) and the QRS complex durations extracted from the measurements of the SR; response electrical signals.

[0176] Of course, other functions are possible.

[0177] This step can be performed by calculating the value of the function for several values of one or a set of parameters and selecting the value of the parameter or set of parameters which makes it possible to obtain the lowest value of the function.

[0178] This step may alternatively or additionally use a correspondence table between values of a parameter and values of a characteristic of the electrical response signal as we will see later. Position estimation

[0179] The processing unit UT is configured to, once the value of one or more parameters of the method has been adapted as a function of the positions P; of the stimulations Si and of the response signals SR; to the simulations, estimate a position of the source of the arrhythmia and a positioning error of the source of the arrhythmia.

[0180] The determination DPE1 of the estimated position PE1 of the source of the arrhythmia affecting the patient uses the global parametric model M0PA1. The values of the parameters of this global model are the initial values for the parameters whose values have not been modified and the new values for the parameters are the values have been modified.

[0181] This step therefore uses each elementary model for which a parameter has been modified.

[0182] This step is advantageously carried out from a number J of simulated response signals SSRj, j = 1 to J. J is an integer greater than 1.

[0183] The simulated electrical response signal SSRj is a simulation of an electrical signal representative of the cardiac activity of the individual obtained in response to stimulation at a predetermined simulated position PSj, i.e. in response to excitation by a source occupying the predetermined position PSj.

[0184] This signal is obtained by simulation. In other words, the simulated electrical response signal SSRj is obtained by a calculation implemented by computer.

[0185] In other words, the simulated electrical response signals SSRj are obtained by a non-invasive method.

[0186] Thus, the determination DPE1 of the estimated position PE1 of the source of the arrhythmia affecting the patient comprises the calculation CASSR of simulated electrical response signals SSRj, with j = 1 to J, from the parametric model M0PA1, the anatomical model MA and the J so-called predetermined simulated positions PSj of the source of the arrhythmia and the selection SEL of a simulated position among the J simulated positions from the first signal of the arrhythmia S.

[0187] In other words, each simulated electrical response signal SSRj is calculated by the processing unit UT using the parametric model M0PA1, the anatomical model MA and a corresponding predetermined simulated position PSj of the source of the arrhythmia.

[0188] This CASSR calculation is, for example, carried out by implementing, for each j = 1 to J, the sequence of steps of the direct problem using the parametric model M0PA1 and a simulated position PSj in place of a position P, so as to generate a simulated electrical response signal SSRj.

[0189] Advantageously J is greater than P.

[0190] Advantageously, J is between 50 and 10,000.

[0191] For example, J is greater than or equal to 100 or 1000.

[0192] Advantageously, the simulated positions PSj are regularly distributed on the epicardium and on the endocardium or in a predetermined area of the epicardium and / or a predetermined area of the endocardium and / or in a predetermined area of the myocardium.

[0193] Advantageously, the estimated position PE1 of the source of the arrhythmia is the simulated position PSj corresponding to the simulated response signal SSRj for which a value of a similarity VSj with the first electrical signal of the arrhythmia SA is the highest among the J values of the similarities calculated for the different j (j= 1 to J).

[0194] Similarity is a quantity representative of a similarity between the two signals. For example, one can calculate a combination of correlation coefficients between the derivations of the two signals or a combination of the inverses of the errors between the derivations of the signals in the least squares sense.

[0195] The implementation of this step is rapid and less dangerous than an invasive stimulation step. It also makes it possible to obtain an estimated position PE closer to the actual position of the source of the arrhythmia affecting the patient than that which would have been obtained solely from the positions P; of the stimulations Si, the number of which is limited. Furthermore, this solution implements an improved parametric method using the stimulations. It is therefore personalized, that is to say adapted to the individual and to the cardiac conditions of the individual during the intervention.

[0196] The CEI calculation of the EPI positioning error affecting the estimated position PE1 is carried out from the response signals SR; and the positions R of the stimulations Si with i = 1 to P.

[0197] This calculation is also carried out from the simulated electrical response signals SSRj calculated for the different simulated positions PSj.

[0198] Therefore, this IEC calculation uses the MOPA1 parametric model to calculate the simulated response signals SSRj.

[0199] For example, the calculation of the positioning error comprises, for each i with i = 1 to P, the selection of the simulated position PSj, taken from among the J simulated positions PSj corresponding to the simulated response signal SSRj for which a value of similarity with the electrical response signal SR; is the highest among the J values of the similarities calculated for the same stimulation Si.

[0200] The simulated position PSj associated with this highest value is the simulated position Pj closest to the position P; of the stimulation Si.

[0201] The calculation of the positioning error comprises, for each i, the calculation of a distance between the simulated position PSj identified as being the closest to the position P; of the stimulation.

[0202] The EPI positioning error is a function, for example a combination, for example an average of these distances.

[0203] The EPI positioning error is, for example, a confidence radius (half of a confidence interval) for a predetermined confidence level. Correspondence table

[0204] As specified previously, the step of defining DV the value of a parameter may comprise a step of using a correspondence table between values of the parameter and first respective reference values of a third characteristic of the cardiac activity.

[0205] For example, we use a table comprising durations of the cardiac action potential as a function of values of pairs of parameters Tin, Tout.

[0206] Values of a third characteristic of the individual's cardiac activity are determined. For example, the P durations of the action potentials from the P SR response signals; to the Si stimulations. These durations are calculated from characteristic durations of the SR response signals;, in particular the duration of the QRS complex of the response signal and the duration of the ST segment of this response signal.

[0207] From the correspondence table, for each stimulation Si, a candidate value of the parameter pair Tin, Tout is selected, the closest associated with the duration of the action potential from the response signal SR;.

[0208] For each stimulation Si, we calculate, from a candidate parametric model, in which the value of this pair of parameters is the candidate value, of the anatomical model MA, for each of J simulated positions PSj, a simulated electrical response signal SSRj and we select, as described previously, the simulated electrical response signal SSRj, called similar, which has the strongest similarity with a reference value which is the response signal SR; to the stimulation Si.

[0209] The value of the parameter is then selected from the result of the comparison.

[0210] For each stimulation S;, we calculate for example an error associated with the value of the pair of parameters, by calculating a distance between the simulated position PSj associated with the simulated electrical response signal called similar and the position P; of the stimulation.

[0211] Thus we obtain an error associated with the candidate value of the pair of parameters for each stimulation Si.

[0212] For example, among the candidate values associated with the different stimulations Si, we select the one which is associated with the smallest error.

[0213] Alternatively, the correspondence table associates the values of a parameter (or more than two parameters) with a value of a characteristic of an electrical response signal

[0214] This step may be the determination step DV1 or an elementary determination step. Determination of simulated response signals

[0215] Advantageously, the location assistance method implemented by the processing unit UT comprises a prior step EPR of calculating the initial simulated response signals SISRj for the J simulated positions PSj.

[0216] In a particular embodiment, the preliminary step EPR comprises an initial step of calculating EIC initial signals of simulated responses SISRj for the J simulated positions PSj using a starting parametric model MODP.

[0217] The starting parametric model MODP differs from the initial parametric model M0P1 only by the value of at least one parameter which is a starting value.

[0218] The initial step is, for example, executed by implementing, for each j = 1 to J, the sequence of steps of the direct problem using the starting values of the parameters the simulated position PSj and the starting parametric model MODP, so as to generate a simulated electrical response signal SISRj.

[0219] The preliminary step EPR then comprises a preliminary improvement step EAMOP1 of the initial parametric model from a second electrical signal of the arrhythmia SA2 so as to generate the initial parametric model MOP1.

[0220] In other words, this step consists of modifying a value of a parameter of the initial parametric model MODP so as to generate the initial parametric model MOP1, from a second electrical signal SA2 of the arrhythmia affecting the individual.

[0221] This step makes it possible to start, in the surgical phase, from an initial parametric model MOP1 which is already efficient, which makes it possible to limit the costs in terms of calculations of the model improvement step in the surgical phase and / or to improve the precision of the estimation of the position of the source of the arrhythmia.

[0222] This step can be implemented in the same way as step DV1. More precisely, the preliminary improvement step EAMOP1 can differ from step DI only in that it uses distinct input data and in particular the J estimated positions Pj instead of the P positions P, the starting parametric model MODP instead of the parametric model MOP1 and the second electrical signal of the arrhythmia SA2 instead of the response signals SR;.

[0223] The initial parametric model MOP1 generated during this step is advantageously the one used by the first elementary determination step in the case of several elementary determination steps.

[0224] The second electrical signal of SA2 arrhythmia may be the electrical signal of SA arrhythmia.

[0225] Alternatively, this second signal SA2 may be distinct from the signal SA.

[0226] For example, the EAMOP1 improvement step of the initial model so as to generating the initial parametric model MOP1 uses a second electrical signal of the arrhythmia measured in the pre-surgical phase and the step of determining the estimated position DPE1 uses an electrical signal of the arrhythmia measured in the surgical phase.

[0227] Advantageously, the preliminary step comprises a step of calculating an initial estimated position PIE of the source of the arrhythmia from the initial parametric model MOP1.

[0228] This step advantageously differs from step DPE1 only by the parametric model used which is the initial parametric model MOP1 instead of the parametric model MOPA.

[0229] The preliminary step also advantageously comprises a step of generating information representative of the initial estimated position PIE so that the system S returns this information to the practitioner.

[0230] In this way, this information is made available to the practitioner via the INTS output interface.

[0231] For example, the INTS output interface displays this information on a set of at least one screen. This display may be of the same type as the display of the estimated position without information relating to a confidence zone.

[0232] Advantageously, this step is implemented prior to the stimulation step.

[0233] This step makes it possible to give an initial estimate of the position to a practitioner who will carry out the stimulation at positions around this initial estimated PIE position as represented in [Fig.2].

[0234] These steps of calculating the initial estimate of the position, of generating the information representative of this estimate and of its restitution can also be implemented in the second embodiment from the initial model M0P2 and the second signal SA2.

[0235] We will now describe, with reference to [Fig.6], the steps implemented by the system according to a second embodiment.

[0236] According to this second embodiment, an inverse parametric model is used to model the propagation of an electrical signal from the individual's torso to the patient's heart.

[0237] In other words, the processing unit UT is configured to implement, during the method of assisting in locating the source of the arrhythmia, a step during which it will solve the inverse problem of electrocardiography by using the inverse model so as to calculate electrograms from an electrical signal SR; in response to a stimulation Si.

[0238] The processing unit UT is configured to implement the following method for assisting in locating a source of an arrhythmia: - generate GA the anatomical model MA, - generate GE2 from the first estimated ESP positions; (with i= 1 to P) of the Si stimulations of the arrhythmia on the individual's heart from an initial parametric model M0P2, the MA anatomical model and the SR response signals; to the Si5 stimulations - define GMA2 the value of a parameter of the model from the estimated ESP positions; and the positions of the stimulations Si of a parameter from the first estimated ESP positions; so as to obtain a parametric model M0PA2 differing from the initial parametric model M0P2 by the value of the parameter, - determine DPE2 an estimated position PE2 of the source of the arrhythmia from the parametric model M0PA2, the anatomical model MA and a first electrical signal of the arrhythmia SA representative of the cardiac activity of the individual generated spontaneously in response to an excitation generated by the source of the arrhythmia affecting the individual, - calculate CE2 a positioning error EP affecting the estimated position PE2 of the arrhythmia source from the position P; of the stimulation Si and the electrical response signal SR; to the stimulation Si, - generate RES information representative of the estimated position PE2 and information representative of the positioning error EP2 affecting the estimated position.

[0239] For example, the GMA2 definition of the parameter includes: - calculate the first distances between the first estimated ESP positions; and the P positions; of the respective Si stimulations, - calculate, for each candidate parametric model of a set of candidate parametric models, second positions estimated from the candidate parametric model, the anatomical model MA and the SRs; response signals of the stimuli Si, - calculate, for each candidate parametric model, second distances between the second estimated positions and the positions of the S stimuli;> - select a parameter value from a function of these distances.

[0240] For example, candidate parametric models differ only in the values of the regularization parameter used.

[0241] In other words, we test several values of the regularization parameter.

[0242] For example, we select the value of the regularization parameter for which the average of the distances obtained for the different stimulations is the lowest and is lower than the average of the first distances.

[0243] The function is, for example, a combination of, for example, an average or a median or a maximum.

[0244] In this embodiment, the processing unit is configured to implement, during the generation step GE2, for each i = 1 to P a so-called inverse method comprising the following sequence of steps, called the sequence of steps of the inverse problem in the rest of the text: - calculate electrical potentials on the surface of the torso from the anatomical model MA and the electrical signal SR; response, - generate electrograms in the heart from electrical potentials using an inverse problem solving model and the MA anatomical model, - generate an activation map from the MA anatomical model and electrograms, - determine a first estimated position ES P, from the activation map and the anatomical model MA.

[0245] Electrograms are curves of extracellular electrical potentials as a function of time. These electrograms are determined at each point of a set of points of the MA anatomical model. These steps are classic steps known to those skilled in the art.

[0246] The step of calculating the electrical potentials on the surface of the torso is carried out in a conventional manner from the electrical response signal SR;.

[0247] The generation of electrograms from the electrical potentials at the surface of the body and the anatomical model is implemented by a method of solving the inverse problem of electrocardiography.

[0248] Given that the transfer matrix A this method solves a model of the regularized inverse problem which includes a regularization parameter X of the inverse matrix.

[0249] This method implements a method known to those skilled in the art for solving the inverse problem, for example described in the Thesis of Amel Karoui “Numerical methods for solving inverse problems in electrocardiography”, University of Bordeaux.

[0250] For example, a boundary element resolution method is implemented, also called BEM (acronym for the English expression “boundary element methods”), the method of fundamental solutions also called MFS (acronym for the English expression “method of fundamental solutions”), the finite element method also called FEM (acronym for the English expression “method of fundamental solution”) or the equivalent single layer method or ESL (acronym for the English expression “Equivalent Single Layer - ESL”).

[0251] for the calculation of the transfer matrix.

[0252] These methods are described in the document “Numerical Investigation of Methods used in Commercial Clinical devices for solving the ECGI Inverse Problem” Narimane Gassa, Vitaly Kalinin, and Nejib Zemzemi.

[0253] Regularization limits the space of solutions to the inverse problem. It therefore places constraints on electrical signals, for example, electrical potentials.

[0254] For example, we can use a Tikhonov regularization or singular value decomposition. These examples are of course not limiting.

[0255] The generation of the activation map from the electrograms and the anatomical model MA is, for example, implemented by a method of internal deflection time also called TID with reference to the English expression “Time of inner deflection” or by a method of mapping the activation direction also called ADM with reference to the English expression “Activation Direction Mapping”. These methods are also described in the previously cited document.

[0256] Thus, in this embodiment, the value of the regularization parameter is advantageously defined.

[0257] Advantageously, the method for assisting in locating the arrhythmia comprises for each stimulation Si: - generating a plurality of estimated positions of the stimulation from the SR response signal; and from the MA anatomical model by so-called methods distinct inverses differing only in that they implement distinct methods for solving the inverse problem and / or distinct methods for calculating an activation map from the electrograms, - selecting an inverse method from among these inverse methods from the estimated positions and the positions of the stimulations Si.

[0258] The selection of the inverse method consists of selecting a method for solving the inverse problem or a method for calculating the activation map or a couple (method for solving the inverse problem, method for calculating the activation map).

[0259] For example, estimated positions are calculated for each stimulation Si by implementing distinct methods differing only in the method used to solve the inverse problem. For example, a method using the boundary element resolution method also called BEM, a method using the fundamental solution method, and a method using the finite element method also called FEM, are used to calculate the transfer matrix.

[0260] Advantageously, during the step of generating a plurality of estimated positions, a plurality of estimated positions are generated from inverse methods implementing distinct inverse problem resolution methods, each of these methods using a distinct respective resolution method taken from the boundary element resolution method, the finite element method also called FEM, the equivalent single layer method or ESL. The selection of the inverse method comprises the selection of a method for solving the inverse problem from these inverse problem resolution methods.

[0261] Alternatively or additionally, in the step of generating a plurality of estimated positions, a plurality of estimated positions are generated from methods implementing distinct activation map calculation methods, each of these methods using a distinct respective resolution method taken from the internal deviation time method and the activation direction mapping method. Selecting the inverse method comprises selecting an activation map calculation method from among these activation map calculation methods.

[0262] The method that gave the most similar estimates is selected according to a predetermined similarity criterion.

[0263] Alternatively, the method for assisting in locating the arrhythmia comprises: - for each stimulation If: • generate a plurality of estimated positions of the stimulation from the SR response signal; and from the MA anatomical model by distinct so-called inverse methods differing in that they implement distinct inverse problem resolution methods and / or separate methods of calculating an activation map from transmembrane potentials and by values of the regularization parameter, - select a pair including the inverse process and the value of the regularization parameter among these processes.

[0264] The CE2 calculation of the EP2 positioning error then includes: - for each stimulation S;, calculation of an estimated position from the selected method using the inverse problem resolution model whose value is determined during the method according to the invention, - calculation of the distances between these estimated positions and the respective positions P; of the stimulations Si, - determination of the error from these distances.

[0265] The positioning error EP2 is a function, for example a combination, for example an average of these distances.

[0266] The positioning error EP2 is, for example, a confidence radius (half of a confidence interval) for a predetermined confidence level.

[0267] The estimated position PE2 is calculated, during the CEI step, from the electrical response signal SR; and from the anatomical model by implementing the possible selected method, and by using the value of the parameter defined during the method. Synchronization

[0268] The input interface INTE of the human-machine interface INT advantageously allows the practitioner to generate a stimulation order to the stimulator STIM so that the stimulation STIM carries out the stimulation at the position of the electrode.

[0269] The processing unit UT is configured to, when it receives a stimulation order from the input interface INTE, control the stimulator STIM from the measurements delivered by the recorder ENRR so that the stimulator STIM carries out the stimulation at the position P; in a predetermined phase of a respiratory cycle.

[0270] This makes it possible to improve the accuracy of the estimation of the position of the source of the arrhythmia. Indeed, the solutions of the direct and inverse methods are affected by the relative position between the heart and the electrodes measuring the response signals and therefore by the respiratory cycle.

[0271] Advantageously, the image(s) on the basis of which the anatomical model MA is generated are taken during a given phase of the respiratory cycle.

[0272] Advantageously, the stimulations S; are carried out during this phase.

[0273] Advantageously, the system S comprises means configured to require that the stimulations Si are carried out during a predetermined phase of a respiratory cycle.

[0274] The phase is for example defined so as to be an instant separated from a predetermined event of the respiratory cycle by a given predetermined duration. This assumes that the duration of the respiratory cycle is considered constant.

[0275] The event is, for example, the start of inspiration or expiration. Material

[0276] From a hardware point of view, the processing unit UT can be seen as a calculator interacting with computer programs.

[0277] The processing unit UT comprises at least one computer, for example, a microcomputer, a computer network, an electronic component, a tablet, a Smartphone or a personal digital assistant (PDA).

[0278] The processing unit UT comprises, for example, a computer, comprising a set of at least one processor, and possibly a memory operationally coupled to the computer.

[0279] The memory comprises for example a computer-readable medium. The computer-readable medium is a tangible device readable by a reader of the processing unit, capable of storing electronic instructions and of being coupled to a communication unit.

[0280] In other words, the computer-readable medium is a tangible medium. In other words, it is not a transient signal per se, such as radio waves or other freely propagating electromagnetic waves, such as light pulses or electronic signals. Such a computer-readable storage medium is, for example, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof.

[0281] For example, the readable medium is an optical disk, a magneto-optical disk, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a random access memory (RAM), a magnetic card or an optical card.

[0282] The readable medium may include an operating system and load the programs according to the invention. It includes registers adapted to record parameter variables created and modified during the execution of the programs. above. A computer program containing software instructions is then stored on the readable medium.

[0283] Alternatively, the program instructions come from an external source and are downloaded via a network. This is particularly the case for applications.

[0284] The processing unit UT comprises a calculator, i.e. at least one electronic data processing circuit designed to manipulate and / or transform data represented by electronic or physical quantities in registers of the evaluation system and / or memories into other similar data corresponding to physical data in the memories of registers or other types of display devices, transmission devices or storage devices.

[0285] The processing unit UT comprises, for example, memories, for storing data, for example measurements, the programs of the parametric methods comprising the values of the parameters capable of being updated, the data calculated during the implementation of the method according to the invention, operationally coupled to the electronic data processing circuit and a reader adapted to read a computer-readable medium.

[0286] The steps of the method according to the invention are, for example, executed by causing the electronic processing circuits of the processing unit UT to read predetermined programs recorded on hardware such as memories so that the electronic data processing circuits execute calculations, control communications and read and / or write data in memories.

[0287] The method according to the invention is, for example, executed on a processing device, for example a single computer, or on a system distributed between several computers (in particular via the use of cloud computing).

[0288] The processing unit UT comprises at least one computer comprising at least elements listed below: a set of one or more processors (for example at least one central processing unit (CPU) and / or at least one graphics processing unit (GPU) and / or at least one microcontroller and / or at least one digital signal processor (DSP)) ASIC capable of interpreting instructions in the form of a computer program and / or a hardware assembly such as an application-specific integrated circuit (ASIC), an in-field programmable gate array (FPGA), a programmable logic device (PLD) of programmable logic arrays (PLA), a system on chip (SOC), and / or an electronic card in which steps of the method according to the invention are implemented in hardware elements.

[0289] The invention relates to a computer program product comprising the computer-readable medium containing instructions which, when executed by the processing circuit, cause the processing unit to implement the localization and arrhythmia assistance process.

[0290] The program product may include the computer-readable recording medium.

[0291] The invention also relates to a computer-readable medium, on which the computer program is recorded.

[0292] Alternatively, the program instructions are obtained from an external source and downloaded via a network. This is particularly the case for applications. In this case, the computer program product comprises a computer-readable data carrier on which the program instructions are stored or a data carrier signal on which the program instructions are encoded.

[0293] The form of the program instructions is, for example, a source code form, a computer-executable form, or any intermediate form between a source code and a computer-executable form, such as the form resulting from the conversion of the source code via an interpreter, an assembler, a compiler, a linker, or a locator. Alternatively, the program instructions are microcode, firmware instructions, state definition data, integrated circuit configuration data (e.g., VHDL), or object code. The program instructions are written in any combination of one or more programming languages, for example, an object-oriented programming language (C++, JAVA, Python), a procedural programming language (e.g., C language).

[0294] The communication unit comprises at least one communication device enabling communication between the elements of the system and possibly between at least one element of the system and a device external to the system. The communication systems can establish a physical link between elements of the system and / or between an element of the system and a device external to the system and / or a remote communication link (wireless) between elements of the system and / or between an element of the system and a device external to the system.

[0295] The communication unit may comprise any hardware, firmware and / or software suitable for communicating information between elements of the device to which the communication unit belongs, for example via a data bus, or to an element external to the device. In order to enable data communication between different devices to which, where appropriate, communication devices belong, these devices comprise firmware and / or software hardware enabling a wired or wireless communication link, for example Wi-Fi, Bluetooth, cellular or Ethernet, to be established between them.

[0296] The user interface INT includes, for example, the interface and output INTS and the input interface INTE.

[0297] The input interface INTE allows a user to enter data or commands so as to be able to interact with the programs according to the invention.

[0298] The input interface INTE comprises, for example, a keyboard or a pointing interface, such as a mouse, a light pen, a touchpad, a remote control, a voice recognition device, a haptic device.

[0299] The processing unit UT is advantageously configured to generate information so that the output interface INTS makes this information available to the practitioner.

[0300] The INTS output interface is designed to restore this information to a user, in a sensory or electrical manner, such as, for example, visually, audibly or haptically. The output interface comprises, for example, a display.

[0301] Previously, a visual provision of position and position error information has been described, but at least one of the information may be a step of providing information by means other than a display, for example haptically or audibly.

[0302] The output interface INTS may be the input device INTE, for example, in the case of a touch tablet.

Claims

Claims

1. System (S) for assisting in locating a source of an arrhythmia affecting the heart of an individual, the system comprising a processing unit (UT) configured to implement, by computer, a method for assisting in locating a source of an arrhythmia comprising: • Generating (GA) a cardiac and thoracic anatomical model (MA) of the individual from at least one image of the individual (I), • Determining (DV) a value of at least one parameter of a parametric model modeling a propagation of an electrical signal of cardiac activity, the value of the parameter being determined from a position (P; ) of at least one stimulation (Si) of the heart of the individual and from a response electrical signal (SR;) representative of the cardiac activity of the individual and measured in response to the stimulation (Si) and from the anatomical model (MA), • Determine (DPE) an estimated position (PE) of the source of the arrhythmia from the parametric model, the anatomical model (MA) and a first electrical signal of the arrhythmia (SA) representative of the cardiac activity of the individual and generated spontaneously by the source of the arrhythmia, • Calculate (CE) a positioning error (EP) affecting the estimated position (PE) of the source of the arrhythmia from the position (P;) and the electrical response signal (SR;), • generate information representative of the estimated position and information representative of the positioning error affecting the estimated position so that the system returns this information to a practitioner.;

2. System according to the preceding claim, in which, determining (DV) the value of the parameter comprises: • generating (GE) a first estimate of a value of a first characteristic of the cardiac activity of the individual using an initial parametric model (MOP), distinct from the parametric model in that it presents an initial value of the parameter distinct from the value, the anatomical model (MA), and an initial data taken from a position or an electrical signal representative of the cardiac activity of the individual, • Determine (GMA) the value of the parameter from the first estimate and a reference value of a second characteristic of the cardiac activity resulting from a measurement taken from the position (P;) of at least one stimulation (Si) and the electrical signal (SR;) of response to the stimulation.

3. System according to the preceding claim, in which the initial data is a position.

4. System (S) according to the preceding claim, in which, determining the estimated position (PE1) of the source of the arrhythmia comprises: • calculating (CASSR) J simulated electrical response signals (SSRj) using the parametric model, the anatomical model (MA) and J so-called predetermined simulated positions (PSj) of the source of the arrhythmia, • selecting a simulated position from among the J simulated positions, from the electrical signal of the arrhythmia (SA).

5. System according to the preceding claim in which, calculating (CEI) a positioning error (EPI) affecting the estimated position (PE1) of the source of the arrhythmia comprises: • Selecting, for at least one stimulation, a simulated position (PSj) taken from among the J simulated positions (PSj) from the electrical response signal (SR;) to the stimulation (Si).

6. System according to any one of claims 4 to 5, in which, determining the value of the parameter comprises, for at least one stimulation (Si): • selecting a candidate value of the parameter, associated with a value of a third characteristic of the cardiac activity closest to a first reference value of the third characteristic from the electrical signal of response to the stimulation in a correspondence table, • calculate for each of J so-called simulated positions (PSj), a candidate estimate of a simulated electrical response signal associated with a source located at the simulated position (PSj) from a candidate parametric model in which the parameter has the candidate value, • select a candidate estimate from among the J candidate estimates.

7. System according to any one of claims 4 to 6, wherein the method for assisting in locating the source of the arrhythmia comprises: • calculating (EIC) J initial signals (SSRj) being simulated electrical signals associated with sources occupying the respective simulated positions (PSj) using an initial parametric model (MODP) and the anatomical model (MA), • modifying a value of at least a first parameter of the initial parametric model (MODP) so as to generate the initial parametric model (MOP), from a second electrical signal (SA2) of the arrhythmia affecting the individual.

8. A system according to any one of claims 3 to 7, wherein the parameter comprises a propagation velocity of an eikonal model or a characteristic time constant of an ionic model.

9. System according to any one of claims 2 to 3, in which the initial data is an electrical signal representative of the cardiac activity of the individual.

10. System according to the preceding claim, in which the parameter is a regularization parameter of the inverse problem.

11. System according to any one of claims 9 to 10, in which the method for assisting in locating the source of the arrhythmia comprises: • for at least one stimulation (Si): • generating a plurality of estimated positions of the stimulation from the response signal (SR;) and the anatomical model (MA) by separate so-called inverse methods differing in that they implement distinct methods for solving the inverse problem and / or distinct methods for calculating an activation map from electrograms, • select an inverse method from among these methods.

12. System according to any one of the preceding claims, comprising: • an imager (IM) configured to generate the image (I), • a stimulator (STIM) to carry out the stimulations (Si), • a recorder (ENRS) configured to record the electrical response signals (SR;) to the stimulations (Si), • a locator (LOC) configured to measure the positions of the stimulations so as to measure the positions (P;).

13. A system according to any preceding claim, wherein the at least one image of the individual is acquired during a predetermined phase of the individual's respiratory cycle and the stimulation is performed during said phase of another respiratory cycle of the individual.

14. System according to the preceding claim, comprising a recorder (ENRR) intended to record the patient's breathing and a stimulator for performing a stimulation, in which the processing unit (UT) is configured to impose, from a measurement of information on the patient's breathing, that the stimulator performs the stimulation (Si) only during a predetermined phase of a respiratory cycle of the patient being the phase during which the image is acquired.

15. A system according to any preceding claim, comprising an output interface comprising a display, the system being configured to display on the display: • a graphical representation of at least a portion of the individual's heart or myocardium, • a visual indicator occupying a position, on the graphical representation of the at least a portion of the individual's heart or myocardium, corresponding to the estimated position, • a graphical representation of an area, around the visual indicator, containing the points of the heart or myocardium myocardium located at a distance from the estimated position less than or equal to the positioning error.

Images