DEVICE FOR IMAGING A PATIENT'S HEART AND ASSOCIATED SYSTEM

Abstract

Title: DEVICE FOR IMAGING A PATIENT'S HEART AND ASSOCIATED SYSTEM Device for imaging a patient's heart, the heart comprising a myocardium, comprising at least one input interface, at least one processing unit, and at least one output interface, the at least one processing unit being configured to implement by computer a processing step comprising: generation of a first black-blood image from first black-blood signals, generation of a first white-blood image from first white-blood signals, generation of a second image of a type between white and black blood from second signals of said type, determination, from said first black-blood image, first white-blood image, and second image of said type, of first localization data for a myocardial lesion and second localization data for adipose tissue. Figure for the abstract: Fig. 2

Description

Title of the invention: DEVICE FOR IMAGING A PATIENT'S HEART AND ASSOCIATED SYSTEM Scope of the invention

[0001] The invention relates to the field of cardiac magnetic resonance imaging (MRI) by late gadolinium enhancement or LGE, in reference to the Anglo-Saxon expression "Late Gadolinium Enhancement".

[0002] The field of application of the invention relates more particularly to methods and systems for characterizing lesions in the heart. This characterization makes it possible, in particular, to guide ablations.

[0003] The reference technique for characterizing regional lesions, including myocardial fibrosis, is bright-blood LGE (BBL) late gadolinium enhancement imaging using inversion recovery, such as the phase-sensitive inversion recovery (PSIR) sequence. In this type of imaging, the viable myocardial signal is induced to cancel out using inversion recovery pulses, allowing visualization of lesions with high contrast between healthy myocardial tissue and the lesions. However, for myocardial lesions adjacent to the heart's blood chambers (right and left ventricles), the high intensity of the signal from the blood, and therefore the low contrast between the lesions and the blood, prevents the automatic, accurate, reliable, and robust characterization of scars, particularly sub-endocardial scars.

[0004] To circumvent this problem, black blood LGE (BL-LGE) imaging techniques have been proposed. These techniques allow for the simultaneous cancellation of signals from both healthy myocardium and blood, thus providing high contrast both between lesions and between blood, and between lesions and healthy myocardium. Healthy myocardium is defined as the largest representative area of ​​the myocardium.

[0005] However, black blood imaging techniques do not allow for proper characterization of lesions, in particular their precise localization in relation to the myocardium, as the contrast between the blood and the healthy myocardium is not sufficiently high.

[0006] The inventors of the present application have previously proposed in patent application FR2203782 a method for acquiring and merging images in black blood and white blood that allows for better localization of heart scars while maintaining a short acquisition time.

[0007] In this type of image, however, scars are difficult to distinguish from tissues such as adipose tissue. It is therefore difficult to discriminate scars from such

[0008]

[0009]

[0010] fabrics. One goal of the invention is to improve the situation. To this end, the invention relates to a device for imaging a patient's heart, the heart comprising a myocardium delimiting a cavity of the heart, comprising: • At least one input interface configured for: • receive, during a pair of interbeats comprising two consecutive interbeats, initial signals in dark blood by magnetic resonance imaging (MRI) in dark blood by late gadolinium enhancement at a first echo duration, • receive, during the inter-beat couple, the first signals in white blood by magnetic resonance imaging (MRI) in white blood by late gadolinium enhancement at the first echo duration, • receive, during the inter-beat couple, second signals of a type chosen from black blood and white blood by magnetic resonance imaging with late gadolinium enhancement at at least one other echo duration different from the first echo duration, • At least one processing unit configured to implement by computer a processing step comprising: • generation of an initial black blood image from the initial black blood signals, • generation of an initial white blood image from the initial white blood signals, • generation of a second image of said type from the second signals of said type, • determination, from the first image in black blood, the second image of said type and possibly the first image in white blood, of first localization data of a myocardial lesion and second localization data of adipose tissue, • At least one output interface configured to return the first localization data of a myocardial lesion and the second localization data of adipose tissue. Myocardial lesion and adipose tissue refers to two areas of interest representing tissue peculiarities near or within a reference area, which is the myocardium. Gadolinium preferentially accumulates in these areas of interest because they are composed of interstitial tissue. Thus, these areas of interest are characterized, in the first black-blood image, by a pixel or voxel intensity exceeding a predetermined threshold. The use of the second image of this type in combination with the first black-blood image (or al alternatively the first image in white blood) allows to highlight the adipose tissues and therefore, to differentiate the two areas of interest previously mentioned.

[0011] In this application, the term "myocardial lesion" refers to a local change in tissue structure compared to the most representative tissue structure of the myocardium. A distinction is made between pathological and benign lesions. For example, some lesions may alter the transmission of a nerve impulse in the myocardium. These lesions are referred to herein as pathological. In another example, some lesions may not alter the transmission of a nerve impulse. These lesions are referred to herein as benign.

[0012] Thus, the device according to the invention makes it possible to determine and return, from three signals during a pair of consecutive inter-beats, two signals being of the same type among white blood and black blood acquired at two different echo durations, a combination of information relating to the existence and position of lesions of the myocardium and cardiac adipose tissue. Thus, the device according to the invention makes it possible to implement more precise clinical observations based on a limited amount of information.

[0013] Advantageously, at least one processing unit is configured to, during the processing step: • Generate a first fused image from a first reference white blood image derived from the first white blood image and the first localization data of the myocardial lesion.

[0014] Advantageously: • At least one input interface is configured to receive, for a number N greater than 1 of inter-beat pairs, a plurality of N first signals in black blood by black blood magnetic resonance and a plurality of N first signals in white blood by white blood magnetic resonance, • At least one processing unit is configured to generate N first images in black blood and N first images in white blood, • The first reference image in white blood is a combination of the first N images in white blood.

[0015] Thus, the quality of the first reference white blood image and the first reference black blood image is better, with for example a better signal-to-noise ratio, and the quality of the first merged image is strengthened.

[0016] Advantageously, at least one processing unit is configured to: • Generate a first phase image and a first magnitude image from a first reference image derived from the first image of said type, • Generate a second phase image and a second magnitude image from a second reference image derived from the second image of said type, • Generate a fat image from the first and second phase images and the first and second magnitude images, • determine, from the fat image, second data on the location of adipose tissue.

[0017] By fat image, we mean an image containing only information about adipose tissue. In other words, the high-intensity pixels of the fat image correspond to the location of adipose tissue. Thus, advantageously, by processing the first reference image and the second reference image, corresponding to two different echo durations, it is possible to obtain the second adipose tissue localization data.

[0018] Advantageously, at least one processing unit is configured to: • when determining said first heart lesion localization data and said second adipose tissue localization data, segment a first reference image from the first white blood image, so as to generate positioning data for a set of at least one wall delimiting the myocardium, • implement a step of geometric characterization of lesions and / or fat of the heart, based on positioning data and second data on the localization of adipose tissue.

[0019] Segmenting the first reference white blood image allows for automatic, robust, reliable, and precise positioning of the walls delimiting the myocardium, because these images exhibit significant contrast between the myocardium and the blood. Black blood images do not provide such good results due to the lack of contrast between healthy myocardium and blood.

[0020] The geometric characterization of lesions and / or fat of the heart which uses, in addition to positioning data from segmentation, the first reference black blood image allows us to obtain good results which would not be possible to obtain on its own with the first reference white blood image containing little or no information on cardiac lesions and adipose tissue of the heart.

[0021] Advantageously, at least one processing unit is configured to: • Generate a second fused image from the first fused image and the second adipose tissue localization data, in which the heart lesions are differentiated from the adipose tissue.

[0022] It is thus possible to generate (and therefore subsequently display and view) this second merged image where one can differentiate and / or combine elements characteristic formations of adipose tissue of the heart and other information characteristic of cardiac lesions.

[0023] Advantageously: • The assembly of at least one wall includes a first wall delimiting and surrounding the myocardium, • at least one processing unit is configured to implement, during lesion and / or fat geometric characterization, a preliminary segmentation step to localize an area likely to contain myocardial lesions and / or adipose tissue on a first reference black blood image from the first black blood image using data from first wall positioning data.

[0024] Advantageously, the first myocardial lesion localization data are obtained from data obtained during the preliminary segmentation step and the second adipose tissue localization data.

[0025] Advantageously, at least one output interface includes a display unit configured to display a first representation of the first myocardial lesion localization data in a first color and a second representation of the second adipose tissue localization data in a second color different from the first color.

[0026] The invention also relates to a magnetic resonance imaging system comprising a magnetic resonance system configured to implement an acquisition step and an imaging device as described above, the acquisition step comprising: • During the inter-beat couple, acquisition of the first signals in dark blood by magnetic resonance imaging in dark blood by late gadolinium enhancement at said first echo duration, • During the inter-beat couple, the first signals are acquired in white blood by magnetic resonance imaging (MRI) using late gadolinium enhancement during the first echo duration. • During the inter-beat couple, acquisition of the second signals of a type taken from black blood and white blood by magnetic resonance by late gadolinium enhancement at at least one other echo duration different from the first echo duration.

[0027] Advantageously, the magnetic resonance system is configured to implement: a step of acquiring second signals in white blood by magnetic resonance imaging using late gadolinium enhancement at the second echo duration,

[0028]

[0029]

[0030]

[0031]

[0032]

[0033] and at least one processing unit is configured to: • generate a second white blood image from the second white blood signals. Advantageously, the magnetic resonance system is configured to implement: • a step of acquiring second black-blood signals by magnetic resonance imaging by late gadolinium enhancement at the second echo duration, and in which at least one processing unit is configured to: • generate a second black blood image from the second black blood signals. The invention also relates to an imaging method for imaging the heart of a patient, the heart comprising a myocardium delimiting a cavity of the heart, the method comprising an acquisition step comprising, for at least one pair of interbeats, two consecutive interbeats: • During the inter-beat couple, initial signals are acquired in dark blood by magnetic resonance imaging (MRI) using late gadolinium enhancement at an initial echo duration. • During the inter-beat couple, initial signals are acquired in white blood by magnetic resonance imaging (MRI) using late gadolinium enhancement during the first echo duration. • During the inter-beat couple, acquisition of second signals of a type chosen from black blood and white blood by magnetic resonance imaging using late gadolinium enhancement at at least one other echo duration different from the first echo duration, the imaging process further comprising a processing step including: • generation of an initial black blood image from the initial black blood signals, • generation of an initial white blood image from the initial white blood signals, • generation of a second image of said type from the second signals of said type, • determination, from the first image in black blood, of the second image of said type and possibly of the first image in white blood, of first localization data of a myocardial lesion and of second localization data of adipose tissue. Thus, the process according to the invention makes it possible to generate, from a minimum acquisition of three signals during a pair of consecutive inter-beats, two signals being of the same type among white blood and black blood acquired at two durations different echoes, a combination of information relating to the existence and position of myocardial lesions and cardiac adipose tissue. Thus, the method according to the invention allows for more precise clinical observations based on a limited amount of information.

[0034] Advantageously, the acquisition step comprises: • Acquisition of second white blood signals by magnetic resonance imaging by late gadolinium enhancement at the second echo duration,

[0035] and the processing step includes: • generation of a second white blood image from the second white blood signals.

[0036] Advantageously, the acquisition step comprises: • Acquisition of second signals in black blood by magnetic resonance imaging using late gadolinium enhancement at the second echo duration,

[0037] and the processing step includes: • generation of a second black blood image from the second black blood signals.

[0038] Advantageously, the processing step comprises: • Generation of a first fused image from a first reference white blood image derived from the first white blood image and the first localization data of the myocardial lesion.

[0039] Advantageously, the acquisition step is implemented for a number N greater than 1 of inter-beat pairs so as to generate N first images in white blood and N first images in black blood, the first reference white blood image being a combination of the N first white blood images and the first reference black blood image being a combination of the N first black blood images.

[0040] Thus, the quality of the first reference white blood image and the first reference black blood image is better, with for example a better signal-to-noise ratio, and the quality of the first merged image is strengthened.

[0041] Advantageously, the processing step comprises: • generation of a first phase image and a first magnitude image from a first reference image derived from the first image of said type, • generation of a second phase image and a second magnitude image from a second reference image derived from the second image of said type, • generation of a fat image from the first and second phase images and the first and second magnitude images, • determination, by computer, from the fat image, of said second data on the localization of adipose tissue.

[0042] By fat image, we mean an image containing only information about adipose tissue. In other words, the high-intensity pixels of the fat image correspond to the location of adipose tissue. Thus, advantageously, by processing the first reference image and the second reference image, corresponding to two different echo durations, it is possible to obtain the second data for the localization of adipose tissue in the heart.

[0043] Advantageously, the determination of the first cardiac lesion localization data and the second adipose tissue localization data comprises: • computer segmentation of a first reference image derived from the first white blood image, so as to generate positioning data for a set of at least one wall delimiting the myocardium,

[0044] the method comprising: • Computer-aided geometric characterization of lesions and / or fatty tissue in the heart, based on the first reference black-blood image, positioning data, and second adipose tissue localization data.

[0045] Segmentation of the first reference white-blood image allows for automatic, robust, reliable, and precise positioning of the walls delimiting the myocardium, as these images exhibit significant contrast between the myocardium and the blood. Black-blood images do not provide such good results due to the lack of contrast between healthy myocardium and blood.

[0046] The geometric characterization of lesions and / or fat of the heart which uses, in addition to positioning data from segmentation, the first reference black blood image allows us to obtain good results which would not be possible to obtain on its own with the first reference white blood image containing little or no information on cardiac lesions and adipose tissue of the heart.

[0047] Advantageously, the processing step further includes the generation of a second fused image from the first fused image and the second adipose tissue localization data, in which the heart lesions are differentiated from the adipose tissue.

[0048] It is thus possible to generate (and therefore to display and visualize) this second merged image where one can differentiate and / or combine information characteristic of the adipose tissues of the heart from other information characteristic of cardiac lesions.

[0049] Advantageously: • The assembly of at least one wall includes a first wall delimiting and surrounding the myocardium, • Lesional and / or fatty geometric characterization includes preliminary segmentation to locate an area likely to contain myocardial lesions and / or fatty tissue on a first reference black blood image from the first black blood image using data from first wall positioning data.

[0050] Advantageously, the first myocardial lesion localization data are obtained from data obtained during the preliminary segmentation step and said second adipose tissue localization data.

[0051] Advantageously, the method includes displaying, on a screen, a first representation of the first localization data of the myocardial lesion and a second representation of the second localization data of the adipose tissue, the first and second representation being displayed in different colors.

[0052] Advantageously, the segmentation of the first reference image uses a first learning function to segment the image into white blood so as to obtain the positioning data.

[0053] Advantageously, the first learning function is a convolutional neural network.

[0054] Advantageously, the convolutional neural network is of the Transformer type.

[0055] Advantageously, the first learning function is trained from a set of training images of the area to be imaged generated from signals acquired during respective acquisition steps by magnetic resonance imaging in white blood by late gadolinium enhancement distinct from inversion recovery sequences.

[0056] Advantageously, the assembly of at least one wall includes a first wall delimiting and surrounding the myocardium.

[0057] Advantageously, the preliminary segmentation uses a second learning function to segment the first reference black blood image from the first black blood image so as to locate an area likely to contain myocardial lesions and / or adipose tissue.

[0058] Advantageously, the second learning function is a convolutional neural network.

[0059] Advantageously, the second learning function is trained from a set of training images of the area to be imaged generated from signals acquired during respective acquisition steps by late gadolinium enhancement dark blood magnetic resonance imaging of inversion recovery sequences.

[0060] Advantageously, the second learning function is trained to determine an intensity threshold.

[0061] Advantageously, the preliminary segmentation includes the selection of pixels from the first reference black blood image exhibiting an intensity greater than a predetermined threshold, the pixels being taken only from among the pixels of the first reference black blood image surrounded by the first wall.

[0062] Advantageously, the assembly of at least one wall comprises a second wall delimiting the myocardium and surrounded by the first wall.

[0063] Advantageously, the geometric lesion and / or fatty characterization of the heart includes a lesion detection step and / or adipose tissue detection step, and / or a geometric lesion characterization step and / or an adipose tissue characterization step.

[0064] Advantageously, the geometric characterization of the lesion includes the calculation of a data representative of a lesion size from data from the positioning data of the first wall and possibly of a second wall surrounded by the first wall, and the second data of localization of the adipose tissues.

[0065] Advantageously, the geometric characterization of the lesion includes the calculation of data representative of a percentage of transmurality of the lesion from data from the first lesion localization data and the positioning data of the first wall and the second wall.

[0066] Advantageously, the first image in black blood, the first image in white blood, the second image in black blood, the second image in white blood, the first reference image and the second reference image are two-dimensional.

[0067] The invention also relates to a magnetic resonance imaging system comprising a magnetic resonance system configured to implement the acquisition step and a processing unit configured to implement the processing step of the process described above.

[0068] Advantageously, the processing unit is configured to implement the segmentation step of the first reference white blood image.

[0069] Advantageously, the processing unit is configured to implement the preliminary segmentation step.

[0070] Advantageously, the processing unit is configured to implement the lesion and / or fatty geometric characterization step of the heart.

[0071] Advantageously, the system includes a set of measurement equipment including a magnetic resonance imaging device capable of implementing black blood magnetic resonance acquisition and white blood magnetic resonance acquisition.

[0072] Advantageously, the processing unit is configured to generate commands for the MRI device so that it implements resonance acquisition magnetic resonance imaging in black blood and magnetic resonance imaging in white blood.

[0073] Alternatively and / or in addition, the processing unit is configured to generate the first (respectively second) white blood and black blood images from the signals acquired during the respective acquisitions.

[0074] Advantageously, the system includes an electrocardiograph configured to acquire an electrocardiogram of the patient during the respective acquisitions.

[0075] The invention also relates to a computer program product comprising instructions which lead the device according to the invention to implement the processing step by computer.

[0076] The invention also relates to a computer-readable medium on which the computer program according to the invention is recorded.

[0077] The invention also relates to an imaging method for imaging the heart of a patient, the heart comprising a myocardium, the method comprising an acquisition step comprising, for at least one pair of inter-beats, two consecutive inter-beats: • During the inter-beat couple, initial signals are acquired in dark blood by magnetic resonance imaging (MRI) using late gadolinium enhancement at an initial echo duration. • During the inter-beat couple, initial signals are acquired in white blood by magnetic resonance imaging (MRI) using late gadolinium enhancement during the first echo duration. • During the inter-beat couple, acquisition of second signals of a type chosen from black blood and white blood by magnetic resonance imaging using late gadolinium enhancement at at least one other echo duration different from the first echo duration,

[0078] the imaging method further comprising a processing step comprising: • generation of an initial black blood image from the initial black blood signals, • generation of an initial white blood image from the initial white blood signals, • generation of a second image of said type from the second signals of said type, • determination, from the first image in black blood, of the second image of said type and possibly of the first image in white blood, of initial location data for a first area of ​​interest and of second location data for a second area of ​​interest, said first area of ​​interest and second area of ​​interest showing in the first image in black blood an intensity greater than a predetermined threshold, and cor responding to tissue structures of a different nature.

[0079] Thus, the method according to the invention makes it possible to obtain localization information for tissue structures of different types from a reference area such as the heart, based on the acquisition of three signals during a pair of consecutive interbeats, two signals being of the same type among white blood and black blood acquired at two different echo durations. Therefore, the method according to the invention allows for more precise clinical observations based on a limited amount of information.

[0080] Among the tissue structures, we can list scars (or lesions), or even adipose tissue.

[0081] Advantageously, the tissue structures corresponding to the first and second areas of interest are areas where gadolinium preferentially accumulates. For example, these are tissue structures composed of interstitial tissue. Thus, the first and second areas of interest are characterized, in the first black blood image, by a pixel or voxel intensity exceeding a predetermined threshold. Brief description of the figures

[0082] Other features and advantages of the invention will become apparent from the following detailed description, with reference to the accompanying figures, which illustrate:

[0083] [Fig-1]: an example of an embodiment of a system according to the invention,

[0084] [Fig.2]: a schematic representation of an elementary acquisition sequence signals for generating first (and / or second and / or other) black blood images and first (and / or second and / or other) white blood images used in the process according to the invention,

[0085] [Fig. 3]: A schematic representation of a black blood MRI acquisition phase and white blood cells produced over a plurality of heartbeats.

[0086] [Fig.4]: a schematic representation of a heart in three dimensions (3D) illustrating different cross-sectional planes distributed along the major axis of the heart and images generated from signals acquired in one of the cross-sectional planes,

[0087] [Fig.5]: a flowchart of an example of a method according to the invention,

[0088] [Fig.6]: a schematic representation of four images comprising at the top On the left, an image in white blood; on the top right, an image in white blood showing the walls detected during the segmentation step; and on the bottom left, an image in black blood and the representation of the walls superimposed on the black blood image.

[0089] [Fig.7]: a schematic representation of three images comprising on the left a A black blood image showing detected lesions and / or fatty tissue of the heart, with a fat image in the middle showing the tissues detected adipose tissue, and on the right a second fused image according to the invention on which the detected lesions and the detected adipose tissue are represented in two different representations,

[0090] [Fig.8]: at the top, an image in black blood with transferred walls delimiting a lesion, on which sectors have been represented, at the bottom left a bullseye type representation of the lesion size and at the bottom right a bullseye type representation of a lesion percentage of transmurality. Invention description

[0091] The invention relates to the field of cardiac imaging by magnetic resonance or MRI, late gadolinium enhancement in black blood and white blood.

[0092] The invention relates to a device, an imaging system and an imaging method for characterizing tissue structures of different nature in or near the heart, and more specifically of at least one cardiac muscle, for example of the myocardium.

[0093] In the continuation of this description, and by way of illustration, tissue structures of different nature include cardiac lesions and adipose tissue.

[0094] By cardiac lesion, we mean an lesion of a muscle of the myocardium.

[0095] Cardiac lesions can be divided into acute lesions resulting from a Acute myocardial injury, such as acute myocardial infarction, and chronic lesions characteristic of chronic heart disease. These lesions are cardiac lesions, for example, of the myocardium or papillary muscles. These lesions include myocardial fibrosis, which frequently develops in the context of hypertrophic or dilated cardiomyopathies, but is also a common sequela of inflammatory heart disease or myocardial infarction.

[0096] The lesions also include myocardial necrosis, i.e. the volumes of myocytes whose cell membrane has been destroyed and the volumes of extracellular matrices and collagen constituting fibrous scars, in the chronic phase of the infarction. Imaging system

[0097] Figure 1 schematically represents an example of an embodiment of a system S according to the invention. The system comprises the hardware and software means for implementing the method according to the invention.

[0098] Advantageously, this system S comprises a set of measurement equipment A including a magnetic resonance imaging (MRI) device B and an electrocardiograph referenced ECR on [Fig.1].

[0099] The system S also includes a processing device C comprising a The system consists of a processing unit (TC) and a human-machine interface (INT). This processing unit can be part of the imaging device (B) or external to it; in the latter case, the system is a single device. Alternatively, the system has a distributed architecture.

[0100] In a manner known per se, the MRI B imaging device comprises a static magnetic field generator GEN_B, a gradient generator GEN_GRAD and a radio frequency (RF) device D_RF.

[0101] The static magnetic field generator GEN_B includes a main polarizing magnet intended to generate, along a longitudinal axis z, a substantially uniform static magnetic field of polarization in a polarization zone (generally a tunnel) intended to include the area of ​​the patient to be imaged, this area to be imaged including the heart.

[0102] The patient is a mammal. Without limitation, the mammal is a man.

[0103] The GEN_GRAD gradient generator comprises three gradient coils (or solenoids) arranged and configured to vary the intensity of the magnetic field in the polarization zone along the respective orthogonal axes x, y, and z fixed with respect to the polarization zone. The choice of intensities circulating in these coils allows the selection, from several possibilities, of a slice, having a given thickness and a cutting plane on which the slice is centered, in which the magnetization of the area to be imaged of the patient received in the polarization zone will be measured.

[0104] The radio frequency device D_RF includes coils or solenoids and is capable of generating MRI acquisition sequences including preparatory sequences of magnetization of the area to be imaged and sequences of reading RF signals from the area to be imaged.

[0105] Each of the preparatory and reading sequences includes at least one radio frequency power source of predetermined and adjustable frequency, shape, duration, phase, amplitude.

[0106] The preparatory sequence is configured to excite, that is to say, to change the direction of the magnetization of the tissues of the area to be imaged.

[0107] The reading sequence is configured to measure the magnetization of the area to be imaged resulting from the preparatory module.

[0108] The ECR electrocardiograph is intended to acquire an electrocardiogram of the patient.

[0109] The TC processing unit is configured to generate commands for the MRI B device, in particular for the RF device D_RF and the gradient generator GEN_GRAD, so that the MRI device generates predefined acquisition sequences of signals from predefined volumes or slices of the area to be imaged.

[0110] The TC processing unit is also configured to generate images of the area to be imaged from the measured signals, using reconstruction techniques known to the person skilled in the art, and to process these images as we will see in more detail later in the description. Acquisition sequence

[0111] Figure 2 shows an example of an elementary acquisition sequence SE1 of an RF signal MRI acquisition sequence for generating images of the heart and an electrocardiogram (ECG) E measured by the ECR electrocardiograph during the elementary sequence SEL

[0112] The acquisition sequence comprises a series of elementary acquisition sequences SE1 such as that shown in [Fig.2].

[0113] The lower part of [Fig. 2] represents the variation of the longitudinal magnetization Mz of the tissues in the area to be imaged as a function of time t during this elementary SEL sequence

[0114] The elementary acquisition sequence SE1 comprises a so-called black blood acquisition ACQ1 followed by a so-called white blood acquisition ACQ2, which will be described later. The black blood acquisition ACQ1 acquires initial black blood signals from the area to be imaged, generating a first elementary black blood image IM1 of the area to be imaged and, optionally, second black blood signals from the area to be imaged, generating a second elementary black blood image IM1'. The white blood acquisition ACQ2 acquires initial white blood signals from the area to be imaged, generating a first elementary white blood image IM2 of the area to be imaged and, optionally, second white blood signals from the area to be imaged, generating a second elementary white blood image IM2'.These acquisition stages ACQ1, ACQ2 each include a preparatory sequence also called preparatory module PREPI, PREP2 and a preparatory sequence also called reading module LE1, LE2.

[0115] In this patent application, the term "module" means a step comprising a radio frequency pulse or a series of radio frequency pulses.

[0116] It should be noted that during the entire duration of the elementary acquisition sequence SE1 and preferably during the entire duration of the MRI acquisition sequence, the static magnetic field generator GEN_B is controlled by the TC processing unit to generate a fixed static magnetic field along the z-axis.

[0117] The gradient generator GEN_GRAD is controlled for the TX processing unit so that the radio frequency device D_RF acquires signals from a predefined slice having a predefined thickness during the elementary acquisition sequence SEL

[0118] The acquisition sequence is a late gadolinium enhancement acquisition sequence implemented following the injection of a Ga-based contrast agent Gadolinium is administered intravenously to the patient 10 to 15 minutes before the acquisition sequences begin in order to obtain images with maximum contrast between lesions and healthy tissue and blood. In the heart, the contrast agent is rapidly eliminated from healthy myocardium, which is poor in interstitial tissue, but accumulates for a prolonged period in myocardial lesions. Gadolinium has an extracellular distribution; that is, it does not cross the membranes of cardiomyocytes.

[0119] Gadolinium has the effect of shortening the relaxation time Tl of the tissues where it accumulates. The relaxation of the magnetization of the lesions following a magnetization reversal pulse is thus faster than that of blood and healthy myocardium. Acquisition in black blood

[0120] Initially, we seek to generate first elementary images in black blood IML. In an image of this type, the intensity of the pixels corresponding to blood and healthy muscle is zero (black pixels) or substantially zero.

[0121] In order to generate such a first elementary image in black blood IM1, the RF device D_RF implements an inversion-recovery black blood acquisition step ACQ1. This ACQ1 black blood acquisition step includes a longitudinal inversion pulse, denoted 180° in [Fig. 2], which flips the longitudinal magnetization of the tissues in the imaged area in the opposite direction, i.e., it reverses the longitudinal magnetization of these tissues. In [Fig. 2], it can be seen that the magnetization of the area to be imaged changes from Mz to -Mz under the effect of the inversion pulse. Due to longitudinal relaxation, the longitudinal magnetization of the different tissues present in the imaged area increases to return to its initial value, passing through zero. Naturally, the relaxation kinetics of the different tissues are different.

[0122] In a manner known per se, the ACQ1 black blood acquisition also includes a PREPI preparatory module implemented after the 180° longitudinal inversion pulse, for example, an adiabatic Tl-rho (Tlp) module of duration denoted TSL (acronym for the Anglo-Saxon expression "Time of Spin Lock") or a T2-weighted module, or of the MTC type (acronym for the Anglo-Saxon expression "Magnetization Transfer Contrast") or a combination of two or three of these modules.

[0123] The PREPI preparatory module is configured so that the longitudinal magnetization of blood A(Blood) and that of healthy myocardium A(Musc) cancel each other out at the same instant te.

[0124] At this same instant te, the longitudinal magnetization of the A(Cica) lesions is clearly greater than zero. By acquiring the signals from the area to be imaged at this instant te, we obtain an image exhibiting very high contrast between the pixels or Voxels corresponding to blood and healthy myocardium are black, and pixels or voxels corresponding to lesions are generally white.

[0125] The first ACQ1 acquisition step in inversion-recovery then comprises a first LE1 readout sequence including a 90° pulse applied at time te and a readout gradient to read the transverse magnetization of the area to be imaged. The inversion time TI is the time separating the 180° pulse of the first LE1 readout sequence from the elementary ACQL sequence. In order to obtain the best contrast between myocardial lesions and blood, as well as between myocardial lesions and healthy myocardium, the LE1 readout sequence is advantageously started at time te when the longitudinal magnetizations of the blood and myocardium cancel each other out in order to generate the image with the best contrast.

[0126] In the example in [Fig. 2], the first LE1 readout module of the black blood acquisition is temporally separated from the PREPI preparatory module of the black blood acquisition. Alternatively, the first LE1 readout module begins as soon as the PREPI preparatory module ends. The same applies to the relative temporal positioning between the PREP2 preparatory module of the white blood acquisition and the LE2 readout module of the white blood acquisition.

[0127] In the example in [Fig.2], the IMPI reversing pulse is generated before the PREPI preparatory module. Alternatively, the PREPI preparatory module is generated before the IMP1 reversing pulse.

[0128] The first LE1 readout module of the ACQ1 acquisition sequence allows the acquisition of first black blood signals enabling the generation of a first IM1 black blood image.

[0129] In an embodiment called "Black Dixon" which will be detailed later, the second black blood signals are also acquired during the first acquisition step ACQ1, allowing the generation of a second elementary black blood image IM1' by a second reading sequence LE1' of the acquisition sequence ACQ1 starting at a time having a second echo duration TE2 different from the first echo duration TE1 of the time at which the first reading sequence LE1 begins.

[0130] By echo duration of a reading sequence, we mean the duration between the instant of the 90° RF pulse applied at time te (denoted TE in the prior art) by the first reading module LE1 and the echo which will be generated by this pulse and which will be recorded by the reading sequence.

[0131] The second black blood elementary image IM1' generated from the second black blood signals is advantageously used with the first black blood elementary image IM1 in an implementation of a Dixon method called "black Dixon" detailed later.

[0132] As will be seen later, the echo durations TE1 and TE2 are predetermined so that so-called water signals and so-called fat signals are respectively in phase (e.g., both appearing dark) and out of phase (e.g., some appearing light and others appearing dark). The water signals are characteristic of water molecules in the heart and the fat signals are characteristic of adipose tissue. The echo durations TE1 and TE2 depend on the characteristics of the system S, in particular the B-MRI imaging device, and on the characteristics of the 90° pulse applied at time te.

[0133] Advantageously, signals can also be acquired to generate, during the first acquisition step ACQ1, other elementary black blood images at echo durations different from the first and second echo durations TE1 and TE2. These other echo durations correspond to times when the water and fat signals are either in phase or out of phase. These other elementary black blood images can be used in combination with the first elementary black blood image IM1 and the second elementary black blood image IM1' in an implementation of a Dixon method. Acquisition in white blood

[0134] The first elementary white blood image IM2 of the area to be imaged is generated from first white blood signals acquired by implementing the white blood acquisition step ACQ2 comprising a preparatory module PREP2 followed by a second reading module LE2 of the white blood acquisition step ACQ2.

[0135] Advantageously, the PREP2 preparatory module is identical to the PREPI preparatory module of the ACQ1 black blood acquisition step, but the invention also applies when these modules are distinct.

[0136] The PREP2 preparatory module is, for example, an adiabatic Tlrho sequence.

[0137] Alternatively, the PREP2 module comprises at least one preparatory sequence taken from a T2-weighted module and an MTC (Magnetization Transfer Contrast) type preparatory module or a combination of two or three of these modules.

[0138] The second reading module LE2 includes a reading gradient for reading the transverse magnetization of the area to be imaged. This second reading module LE2 can be implemented using gradient echo or spin echo, just like the first reading module LE1 of the ACQ1 black blood acquisition step. The first and second reading modules LE1 and LE2 can be identical or different.

[0139] The time interval D2 separating the second reading module LE2 from the white blood acquisition ACQ2 is defined such that the longitudinal magnetization of blood A(Blood) is greater than that of the Myocardium A(MUSC), which leads to the generation of an image in which the blood pixels or voxels are white, i.e., with high luminance, and in which the pixels of the myocardial tissue are slightly less luminous than those of the blood, as can be deduced from the curves shown in [Fig. 2]. These images allow for perfect visualization of the cardiac anatomy, making it possible to delineate the myocardium in this image, which is not possible with a black blood image.

[0140] The second LE2 reading module of the ACQ2 white blood acquisition allows the acquisition of first white blood signals enabling the generation of a first IM2 white blood image.

[0141] In an embodiment called "White Dixon" which will be detailed later, during the white blood acquisition step ACQ2, second white blood signals are acquired by the second reading module LE2, allowing the generation of a second elementary white blood image IM2' at an echo duration TE2 different from the echo duration TE1 of the instant at which the second reading sequence LE2 begins.

[0142] This second elementary image in white blood IM2' is advantageously used with the first elementary image in white blood IM2 in an application of a Dixon method called "White Dixon" detailed later.

[0143] As with black blood acquisition, the TE1 and TE2 echo durations are predetermined so that the water and fat signals are, respectively, in phase at echo duration TE1 and out of phase at echo duration TE2. The water signals are characteristic of water molecules in the heart and the fat signals are characteristic of the heart's adipose tissue.

[0144] For example, the second reading sequence LE2 begins with the emission of a 90° RF pulse causing the protons to vibrate, which generates echoes at different times subsequent to the time of emission of the 90° RF pulse. For example, the echo durations TE1 and TE2 correspond to the first and second subsequent times.

[0145] The echo durations TE1 and TE2 depend on the characteristics of the system S, in particular the MRI imaging device B, and on the characteristics of the 90° pulse applied at time te.

[0146] Advantageously, during the second acquisition step ACQ2, other elementary white blood images can also be generated at echo durations other than the echo durations TE1 and TE2. These other echo durations correspond to times when the water and fat signals are either in phase or out of phase. These other elementary white blood images can be used in combination with the first elementary white blood image IM2 and the second elementary white blood image IM2' in an implementation of a Dixon method.

[0147] It will be explained later how, using the IM1, IM2, and IM1' images and possibly the other elementary black blood images in the "Black Dixon" embodiment, or IM2' and possibly the other elementary blood images white in the "Dixon white" embodiment, it is possible to improve the detection of lesions and their distinction from the adipose tissues of the heart, and therefore their characterization, in particular their sizing.

[0148] Synchronization of acquisition steps with cardiac cycles

[0149] Preferably, the TC processing unit is configured to synchronize the SE acquisition sequence with electrocardiogram E.

[0150] For this purpose, the TC processing unit uses the electrocardiogram E to generate trigger commands for acquisition sequences destined for the RF device, the gradient generator and possibly the main magnetic field generator.

[0151] Advantageously, the acquisition sequence SE comprises, as seen in [Fig.3], a plurality of elementary acquisition sequences SEi, with i = 1 to N, where N is greater than 1, where i is the index of the elementary sequence, whose acquisition steps ACQ1, ACQ2 are identical. i = 1 on [Fig.2].

[0152] Each elementary acquisition sequence SEi is advantageously implemented during two consecutive cardiac cycles, preferably during two consecutive interbeats Cl, C2 referenced in [Fig. 2] constituting a pair of interbeats CBi referenced in [Fig. 1]. One advantage is to minimize the acquisition time and therefore minimize heart movements between the different acquisitions and the spatial shifts between the IM1 and IM2 images, and IM1' and IM2', as well as possibly the other elementary images in black blood or the other elementary images in white blood, if applicable.

[0153] In the rest of the text, a beat is defined as a QRS complex, and an interbeat as a phase of a cardiac cycle located between two consecutive beats.

[0154] Advantageously, the consecutive elementary acquisition sequences SEi are implemented during consecutive inter-beat pairs CBi.

[0155] Each elementary SEi acquisition sequence comprises: - During the first inter-beat Cl of the inter-beat couple CBi, the black blood acquisition stage ACQ1; - During the second inter-beat C2 of the inter-beat couple CBi, the white blood acquisition stage ACQ2.

[0156] One interest is to minimize the acquisition time and therefore to minimize the movements of the heart between the different acquisitions and the spatial shifts between the IM1 and IM2 images, and IM1' and IM2', as well as possibly the other elementary images in black blood or the other elementary images in white blood as appropriate, acquired during the different elementary SEi sequences.

[0157] Advantageously, as shown in [Fig. 2], the SE acquisition sequence and the E electrocardiogram are synchronized so that the first and second LE1, LE2 reading modules are implemented during the same phase of their respective cardiac cycles Cl, C2.

[0158] Advantageously, this phase is an inter-beat.

[0159] Advantageously, this phase is diastole.

[0160] Advantageously, the acquisition sequence SE and the electrocardiogram E are synchronized so that the reading modules LE1, LE2 of the black blood and white blood acquisition steps ACQ1, ACQ2 are implemented at the same times of these respective cardiac cycles Cl, C2.

[0161] These instants are defined with respect to the same time reference of cardiac cycles Cl, C2. The time reference is, for example; the maximum of the QRS complex.

[0162] These instants are separated by the same duration D2' from the maximum of the wave R in the example of [Fig.2].

[0163] Synchronization of the first and second LE1 LE2 reading modules of the black blood and white blood acquisition sequences ACQ1, ACQ2 and, as we will see later, of different reading modules of white blood acquisition stages on the one hand and of the different reading modules of black blood acquisition stages on the other hand, makes it possible to generate images of the heart at times when the heart occupies the same position in a fixed reference frame with respect to the main magnet which makes it possible to superimpose the images obtained without the need for registration or by performing a simple registration.

[0164] The invention also applies when the order of the acquisition steps is different. For example, several black blood ACQ1 acquisition steps can be implemented during consecutive inter-beats, then several white blood ACQ2 acquisition steps during consecutive inter-beats, and vice versa.

[0165] It is also possible to implement at least one black blood acquisition step ACQ1 and at least one white blood acquisition step ACQ2 during the same interbeat.

[0166] Alternatively, at least one ACQ2 white blood acquisition step is separated from the nearest temporally close ACQ1 black blood acquisition step. Acquisition of cups

[0167] Advantageously, the TC control processing unit is configured to generate commands to the MRI device to acquire signals from respective slices distributed along a predefined axis of the heart.

[0168] The axis is advantageously the major axis of the heart. The resulting slices are then called minor axis slices. One advantage is that they allow excellent visualization of both ventricles. However, the invention also applies to cases where the slices are distributed along another axis of the heart, for example, a two-chamber axis (long axis). vertical) or 4 cavities (horizontal long axis) of the core. Each section has a defined thickness along the axis and is centered on a predefined cutting plane perpendicular to the axis. In this application, a section is defined as a slice or layer perpendicular to the axis g and having a predefined thickness along the axis.

[0169] Advantageously, the cuts are contiguous along the axis.

[0170] This is achieved by the commands generated by the choice of commands generated for the gradient generator GEN_GRAD by synchronizing the commands for the gradient generator GEN_GRAD and for the RF device D_RF.

[0171] Advantageously, the signals are acquired in adjacent or partially overlapping slices. This allows for complete imaging of the heart.

[0172] Advantageously, the gradient generator GEN_GRAD is controlled so that several two-dimensional images of each slice can be generated from the signals acquired during the SE acquisition sequence.

[0173] Figure 4 represents a three-dimensional (3D) view of a core comprising a plurality of section planes, denoted PCk, distributed along the major axis g, with k = 1 to K, K being an integer greater than 1. The section planes are distributed from the apex AP to the base BA of the core CO. Consequently, the sections are referred to as "minor axis sections".

[0174] Only the images from the section centered on the cutting plane PCksont are shown on [Fig.4].

[0175] Preferably, several first (and possibly second and other) elementary images in black blood IMlk(j) and / or several first (and possibly second and other) elementary images in white blood IM2k(j) are generated, with j = 1 to J, J being an integer greater than or equal to 2 for at least one slice plane PCk, for example, for each slice plane PCk. This makes the analysis of the obtained images more robust.

[0176] Alternatively, the system is configured to acquire signals from a volume, for example from the entire core so as to allow the generation of three-dimensional images.

[0177] Advantageously, the SE acquisition sequence is implemented while the patient is holding their breath. One advantage is obtaining perfectly registered images, which makes it possible to limit, simplify, or eliminate the need for image registration.

[0178] Alternatively, the SE acquisition sequence is implemented during free breathing. Free breathing acquisition offers temporal advantages. Indeed, acquisition during apnea must be rapid, which necessitates reducing the acquisition time from a few seconds to a few minutes and often implies having to reduce the area to be imaged, for example, by limiting it to a 3D portion. However, the sequence Since SE acquisition is synchronized with the ECG so that the reading sequences are implemented at the same time marker of different cardiac cycles, apnea acquisition leads to the generation of a limited number of images. Image generation

[0179] The TC processing unit is configured to generate GEN, by reconstruction techniques known to those skilled in the art, images of the area to be imaged.

[0180] One can, for example, use the GRAPPA algorithm or the SENSE algorithm (and its iterative version).

[0181] The generation step includes a step of generating two-dimensional (2D) or three-dimensional (3D) elementary images of the area to be imaged from the signals acquired during the SE acquisition sequence.

[0182] In the 2D case, advantageously, for each black blood acquisition step ACQ1, a first black blood image IM1 is generated, and optionally a second black blood image IM1' and other black blood images in the "Dixon black" mode, from the signals acquired during the black blood acquisition step ACQ1 and, for each white blood acquisition step ACQ2, a first white blood image IM2 is generated, and optionally a second white blood image IM2' and other white blood images in the "Dixon white" mode, from the signals acquired during the white blood acquisition step ACQ2.

[0183] The elementary images IM1, IM2, IM1', IM2' are advantageously generated in grayscale. Each image comprises a set of unit elements of the pixel or voxel type, each characterized by an intensity I that can take a set of N values ​​(N being a finite integer greater than 1) corresponding to N grayscale levels ranging from 0 to N-1. For example, this value can take 256 values ​​between 0 and 255, but N is not limited to 256. This value can advantageously take 4096 values ​​between 0 and 4095.

[0184] The TC processing unit is also configured to use the elementary images obtained to characterize lesions of the heart and / or adipose tissue, for example of the myocardium as we will see in more detail later in the text.

[0185] The elementary images generated by the TC processing unit can be intended to be displayed on a screen of the INT human-machine interface. Recalibration

[0186] Advantageously, the process is devoid of an image registration step.

[0187] Alternatively, the image generation step GEN includes the registration of first elementary black blood images with each other, and / or the registration of second elementary black blood images with each other, and / or the registration of other elementary black blood images with each other, and / or the registration of first elementary white blood images between them, and / or the registration of second elementary images in white blood, and / or the registration of other elementary images in white blood with each other, and / or the registration of first (and / or second and / or other) elementary images in black blood and white blood with each other. This allows, particularly when the patient is breathing freely during the SE acquisition sequence, to limit the effects of respiration on the position of the heart and thus avoid the spatial shifts induced by respiration on the images, which could affect the accuracy and reliability of the analysis of these images or combinations of these images. Indeed, the respiratory rate is a priori different from the heart rate, but even if correlations exist between these two rhythms, it is possible for respiration to accelerate while the heart rate remains stable, or vice versa.

[0188] Advantageously, the process includes the registration of first elementary images in black blood and / or second elementary images in black blood and / or other elementary images in black blood of the same section.

[0189] Advantageously, the registration is carried out using an implementation of a non-rigid image registration algorithm.

[0190] Advantageously, the process includes the registration of first elementary images in white blood and / or second elementary images in white blood and / or other elementary images in white blood of the same section.

[0191] Advantageously, the registration is carried out using an implementation of a non-rigid image registration algorithm.

[0192] Advantageously, the process includes the registration of first elementary images in black blood IM1 and in white blood IM2 with respect to each other.

[0193] Advantageously, the process includes the registration of second elementary images in white blood IM2' and second elementary images in black blood IM1' of the same slice.

[0194] Advantageously, the method comprises registering other elementary images in white blood IM2'i and other elementary images in black blood IMl'i of the same slice

[0195] Advantageously, this registration is carried out using an implementation of a non-rigid image registration algorithm.

[0196] Advantageously, these algorithms are identical. It is possible to choose a different algorithm for processing the first, second, and other elementary images in black and white blood, but it is preferable to choose the same algorithm for ease of implementation. Registration significantly improves the quality, particularly the contrast, of an image derived from a plurality of first, second, and other elementary images of the same slice. Furthermore, it reduces artifacts related to respiration.

[0197] According to a first example, at least one of the non-rigid algorithms is based on the mutual information method between images, founded on statistical relationships. The function to be optimized can be implemented using a statistical similarity criterion. One advantage of this method is that the matching of homologous attributes of images in the same slice is independent of their geometric position. Furthermore, this method is particularly effective for registering first and / or second and / or other elementary images exhibiting different contrasts, such as black and white blood images.

[0198] According to a second example consistent with the first example, at least one of the non-rigid algorithms is based on a transformation model. The transformation model allows the determination of functions for minimizing the difference between two images. The difference can be expressed as a geometric error to be minimized. Different approaches can be used, such as those based on extracting geometric primitives or shape descriptors, such as salient points, shape singularities, or contours, from each of the images. A parametric or non-parametric approach can be used.

[0199] According to an example of optimization of a transformation model or a similarity criterion, the least squares method can be used.

[0200] Other optimization methods can be implemented, such as gradient descent. However, this latter method is more specifically applied to image intensities and is not optimal within the scope of the invention, since the aim is to optimize the sharpness and contrast of the merged image. Nevertheless, the invention includes this embodiment.

[0201] Registration can be performed by choosing a reference image and determining a transformation function for the other images of the same section with respect to this image. Each image is then registered by optimizing a transformation to obtain the reference image according to a geometric criterion from the image under consideration.

[0202] During three-dimensional image acquisition, it is possible to acquire several three-dimensional images of the core, which can optionally be registered. Image combination

[0203] When a plurality of elementary images of the same section is generated, it is possible to carry out operations aimed at combining, that is to say merging these images in order to produce a single combined image per section.

[0204] Advantageously, the GEN image generation step includes, for at least one slice of index k, for example for each slice, the combination of first (or second or other) white blood images of the slice IM2k(j), for example for j = 1 to J so as to obtain a first combined ISBken white blood image of the slice.

[0205] In the "Dixon white" embodiment, a second combined ISB2ken white blood image of the cut is also obtained, and possibly other combined white blood images of the cut.

[0206] Advantageously, the method includes, for at least one slice, for example for each slice, the combination of first (or second or other) black blood images IMlk(j), for example for j = 1 to J of the slice with index k so as to obtain a combined black blood image ISNk of the slice.

[0207] In the "Dixon black" embodiment, a second combined image of the cut is also obtained in black blood ISN2ken, and possibly other combined images in black blood of the cut.

[0208] This helps to reduce noise and increase the signal-to-noise ratio.

[0209] Image combination can be performed before, during or after image generation GEN.

[0210] According to one example, the combination is an averaging. The averaging is, for example, performed in image space or in Fourier space (i.e., the frequency domain, before image reconstruction). These solutions are computationally inexpensive and fast.

[0211] Averaging has the advantage of preserving image detail, since it increases the signal-to-noise ratio (SNR). This technique smooths noise to reduce residual image artifacts. Furthermore, averaging improves the bit depth of the digital image beyond what is possible with a single image.

[0212] One advantage of the image averaging step for images acquired from the same slice is to reduce the maximum deviation. The noise amplitude decreases as the square root of the number of images used; that is, with only 4 images, the noise amplitude can be reduced by a factor of two. For example, in a 2-minute free-breathing acquisition, it is possible to collect 4 to 5 images per slice, which allows for good noise reduction performance.

[0213] In one embodiment, image combination can alternatively be implemented by motion-compensated iterative reconstruction. In other words, this type of combination is implemented during image reconstruction. Compensated MRI reconstruction techniques are described in particular in the following articles: Odille F, et al., “Generalized reconstruction by inversion of coupled systems (GRICS) applied to free-breathing MRI,” Magnetic Resonance in Medicine, 2008; and “3D whole-heart isotropy sub-millimeter resolution coronary magnetic resonance angiography with non-rigid motion-compensated PROST,” Bustin A, et al., Journal of Cardiovascular Magnetic Resonance, 2020.

[0214] Advantageously, the first (or second or other) elementary images of a part in black blood and part in white blood are respectively combined in such a way as to produce a combined black blood image ISNk and a combined white blood image ISBk per slice.

[0215] In the "Dixon white" embodiment, advantageously a second combined ISB2ken white blood image of the cut is obtained, and possibly other combined white blood images of the cut.

[0216] In the "Dixon black" embodiment, a second combined image of the cut is advantageously obtained in black blood ISN2ken, and possibly other combined images in black blood of the cut.

[0217] In the following, it will be understood by, first reference white blood image, an image derived from the first white blood image.

[0218] For example, the first reference white blood image may be a combination of N first white blood images (combined or elementary) generated from ACQ2 white blood acquisitions of respective SEi elementary acquisition sequences, with i = 1 to N, where N is greater than 1 and where i is the index of the elementary sequence.

[0219] Alternatively, the first reference white blood image is a white blood image (combined or elemental).

[0220] Similarly, it will be understood by, first reference black blood image, as an image derived from the first black blood image.

[0221] For example, the first reference black blood image may be a combination of N first black blood images (combined or elementary) generated from ACQ1 black blood acquisitions of respective elementary acquisition sequences SEi, with i = 1 to N, where N is greater than 1 and where i is the index of the elementary sequence.

[0222] Alternatively, the first reference black blood image is a black blood image (combined or elementary). The expressions "second reference white blood image" and "second reference black blood image" are defined in the same way, in relation to the second white blood image and the second black blood image, acquired at the second echo duration TE2 of an inter-beat pair.

[0223] During the acquisition of three-dimensional images, it is possible to acquire several three-dimensional images of the heart, which can be combined, for example averaged to obtain a single three-dimensional image of the heart.

[0224] Characterization of lesions and / or adipose tissue

[0225] As seen previously, during the SE acquisition sequence, first black blood signals are acquired for the generation of first elementary black blood images IM1 and first white blood signals for the generation of first elementary white blood images IM2.

[0226] Similarly, in the so-called "White Dixon" embodiment, second white blood signals are acquired for the generation of second elementary white blood images IM2' and possibly other elementary white blood images.

[0227] In the so-called "Dixon black" embodiment, second black blood signals are acquired for the generation of second elementary black blood images IM1' and possibly other elementary black blood images.

[0228] These first, second and other elementary images IM1, IM2, possibly IM1', and possibly IM2' are generated from the acquired signals.

[0229] Advantageously, but not necessarily, these first, second, and other elementary images IM1, IM2, IM1', IM2' are generated from signals acquired by implementing an acquisition sequence. This acquisition sequence comprises ACQ2 blood-white acquisition steps. Each ACQ2 blood-white acquisition step is distinct from an inversion-recovery sequence.

[0230] In other words, this sequence is devoid of a reversal pulse of the longitudinal magnetization of the area to be imaged.

[0231] Therefore, unlike PSIR imaging during white blood acquisition, the longitudinal magnetization of the myocardium is not canceled, which allows for images with a higher contrast between lesions and blood, thus facilitating diagnosis and image processing.

[0232] Alternatively, at least one white-blood acquisition step is a recovery inversion sequence, for example a PSIR sequence.

[0233] Advantageously, the white blood acquisition is configured so that, during the implementation of the LE2 readout module, the respective longitudinal magnetizations of the healthy myocardium, blood, and lesions are positive, and the longitudinal magnetization of the lesions is between the magnetization of the healthy myocardium and that of the blood. This is achieved by the configuration of the PREP2 preparatory module and that of the LE2 readout module, and by the relative temporal positioning between these two modules.

[0234] As can be seen in [Fig. 5], the imaging method comprises: - the determination DET, from a first image in black blood, a first image in white blood, a second image of a type taken from white blood in the embodiment "Dixon white" and black blood in the embodiment "Dixon black", and possibly another image of a type taken from white blood in the embodiment "Dixon white" and black blood in the embodiment "Dixon black", of first localization data of a myocardial lesion and second localization data of adipose tissue.

[0235] This step is implemented by the TC processing unit, which uses for this purpose first or second or other images white blood ISB, ISB2, and black blood ISN, ISN2 generated by the process described above.

[0236] In an embodiment referred to as the "fusion mode," it is advantageously possible to generate an image in which the first data for the localization of a myocardial lesion and the second data for the positioning of adipose tissue are differentiated. In a detailed example to follow, it will be described that the differentiation can be achieved by using different colors or representative patterns.

[0237] In an embodiment referred to as the "segmentation mode," it is advantageously possible to determine quantitative data based on the first myocardial lesion location data and the second adipose tissue positioning data. In this embodiment, a SEG segmentation step and a preliminary SEG segmentation step are advantageously implemented.

[0238] Each first white blood image ISB, or second white blood image ISB2, or other white blood image respectively black blood image ISN, ISN2, ISN2, is a first (or second or other) elementary image IMlk(j), IM2k(j) or a first (or second or other) combined image in white blood ISBk, ISB2k, respectively in black blood ISNk, ISN2k.

[0239] In the following text, it is considered, as in the example of [Fig.5], that each first white blood image ISB, or second white blood image ISB2, or other white blood image is a first (or second or other) combined image in white blood ISB k, ISB2ket that each first black blood image ISN, or second black blood image ISN2, or other black blood image is a first (or second or other) combined image in black blood ISNk, ISN2k.

[0240] The invention makes it possible to characterize, in an automatic, reproducible, reliable and precise way, a cardiac lesion, and more specifically cardiac muscles, in particular myocardial lesions, as well as adipose tissue. Segmentation

[0241] According to one example, a SEG segmentation is implemented using at least one image from the first white blood image ISB, or the second white blood image ISB2 in the case of the "Dixon white" embodiment, so as to generate positioning data of a first wall L1 and a second wall L2 delimiting the myocardium M and surrounding and delimiting a cavity of the heart, the first wall L1 surrounding the second wall L2.

[0242] Positioning data relating to a wall corresponds, for example, to the identification of the pixels constituting the wall.

[0243] The result of this segmentation is visible in [Fig. 6], schematically representing, at top left, a first white blood image of a section of the ISBk heart and, at top right, the first white blood image of the section on which the first wall L1 and the second wall L2 obtained during the segmentation step are represented in thick black lines.

[0244] The generated images are, for example, in greyscale. The dotted, checkered and triangular areas of [Fig.6] represent areas of greater intensity than the bricks.

[0245] In the example of [Fig.6], the heart cavity is the left ventricle and the SEG segmentation is implemented so as to delimit the walls L1, L2 of the part of the myocardium surrounding and delimiting the left ventricle.

[0246] It should be noted that in the present description the invention is described in the case where the cavity of the heart is the left ventricle, but the invention is applicable to any cavity of the heart, such as the right ventricle and the atria which are also surrounded and delimited by the myocardium and subject to cardiac lesions.

[0247] The second wall L2 is the wall delimiting the myocardium and the left ventricle (LV). The first wall L1 surrounding the second wall L2 is the external wall, that is, facing outwards from the left ventricle (LV), of the part of the myocardium surrounding the left ventricle (LV). This is the epicardium.

[0248] The second wall L2 is the wall of the myocardium delimiting the left ventricle. This is the endocardium.

[0249] On the images of [Fig.6] which are sections of the heart these walls L1, L2 form closed curves in that they completely surround the left ventricle LV in short axis sections.

[0250] It is easy to understand that in 3D these walls form surfaces.

[0251] Alternatively, the segmentation is implemented so as to generate positioning data for only one of these two walls, for example the outer wall of the myocardium.

[0252] SEG segmentation is achieved by implementing a first learning function or algorithm, for example an artificial neural network, to segment a white blood image so as to delimit at least one wall of the myocardium surrounding and delimiting a cavity of the heart.

[0253] In a non-limiting example, the first learning function is a neural network.

[0254] The artificial neural network used for segmentation is advantageously a convolutional neural network.

[0255] The convolutional neural network is, for example, of the U-Net type or of the transformer type also called a self-attentive model, for example, of the type commonly called a swin transformer.

[0256] The neural network, or more generally the first learning function, is implemented on two-dimensional (2D) images and / or on three-dimensional images sionnelles (3D). In other words, it is trained to perform the desired segmentation by receiving 2D and / or 3D images as input.

[0257] Advantageously, the first learning function is trained, prior to the implementation of the method according to the invention, from white blood images of the heart, generated from signals acquired during separate white blood acquisition steps of inversion recovery sequences, and labeled by specialists, i.e. segmented by specialists, so that the first trained learning function receiving input data including a white blood image of the heart is able to segment in such a way as to delimit at least one wall of the myocardium surrounding and delimiting a cavity of the heart.

[0258] The first learning function is, for example, configured to deliver, from a white blood input image generated from signals acquired during a white blood acquisition step separate from an inversion recovery sequence, an output image in which the pixels or voxels corresponding to the walls or contours L1 and L2 are colored in a predetermined intensity or color or in respective predetermined colors. Spread

[0259] The method may include a step of propagating the walls detected during the SEG segmentation step onto at least one first black blood image (and / or a second or other black blood image in the "Dixon black" embodiment). In other words, the method may include a REP transfer step, i.e., propagation, comprising the identification, on a first (or second or other) ISNk, ISN2k black blood image, of the pixels or voxels corresponding to the L1 and L2 walls identified during the SEG segmentation.

[0260] On [Fig.6], a first black blood ISNk image of the heart section is schematically represented at the bottom left and at the bottom right the first black blood image on which the L1 and L2 walls detected during the SEG segmentation step are represented in thick black lines.

[0261] The identification, on the first black blood image ISNk, of the pixels or voxels corresponding to the first wall L1 and respectively to the second wall L2 is determined from the positions of the pixels or voxels corresponding to these walls on the first white blood image ISBk.

[0262] These pixels or voxels can have the same respective positions on the first image in white blood and on the first image in black blood when these images are considered to be spatially registered and because these images have the same size and the same resolution.

[0263] A predetermined or calculated spatial offset may alternatively be applied to these pixels or voxels when it is estimated that a spatial offset exists between these images.

[0264] The report may include the annotation or colouring of the pixels or voxels corresponding to the walls L1 and L2.

[0265] Advantageously, the characterization includes segmentation for each first (or second or other in the "Dixon white" embodiment) ISBk white blood image so as to generate respective positioning data obtained from the first (or second or other in the "Dixon white" embodiment) respective ISBk white blood images.

[0266] Advantageously, the characterization includes mapping onto each first (or second or other in the "Dixon black" embodiment) black blood image ISNk, the pixels or voxels corresponding to the L1 and L2 walls identified during the SEG segmentation, from one of the first (or second or other in the "Dixon white" embodiment) white blood images ISBk. The pixels or voxels mapped onto the different black blood images or combined black blood images are advantageously identified from the respective first (or second or other in the "Dixon white" embodiment) white blood images.

[0267] Advantageously, the positioning data used for plotting on an ISNk black blood image are generated from a first ISBk white blood image (or a second or other ISB2k white blood image) generated from signals measured during the same elementary acquisition sequence.

[0268] Thus, the positioning data generated from a combined white blood image ISBk (or from a second or other white blood image ISB2k of order k) is advantageously transferred to a combined black blood image ISNk of order k.

[0269] Lesion and / or fatty geometric characterization step CAR

[0270] The lesion and / or fat geometric characterization step CAR consists of characterizing the heart from the point of view of lesions and / or adipose tissue using one or more first (or second or other in the "Dixon black" embodiment) black blood image(s) ISNk ISN2k, and positioning data of at least one wall, for example of the second wall L2, obtained from one or more first (or second or other in the "Dixon white" embodiment) white blood images ISBk ISB2k.

[0271] This step advantageously allows the generation of data characterizing the heart from the point of view of lesions and / or adipose tissue.

[0272] This step is implemented by the CT processing unit.

[0273] The CAR characterization step may include a lesion detection step and / or DEG detection of adipose tissue, and / or a CAE lesion geometric characterization step and / or a CAG characterization step of cardiac adipose tissue.

[0274] Advantageously, when a lesion is detected, the CAE lesion characterization step is implemented. In other words, the CAE lesion characterization step can to be implemented only if a lesion is detected during the DE detection step.

[0275] Alternatively, the DE detection step is implemented after the CAE lesion characterization step or after one of the steps of this CAE effective lesion characterization step.

[0276] Alternatively, the CAE lesion geometric characterization step is devoid of a DE detection step.

[0277] Advantageously, when adipose tissue is detected, the cardiac adipose tissue CAG characterization step is implemented. In other words, the cardiac adipose tissue CAG characterization step can be implemented only if cardiac adipose tissue is detected during the DE detection step.

[0278] Alternatively, the DEG detection step of cardiac adipose tissue is implemented after the CAG characterization step of cardiac adipose tissue or after one of the steps of this CAG characterization step of cardiac adipose tissue.

[0279] Alternatively, the CAG characterization step of cardiac adipose tissue is devoid of a DEG detection step of cardiac adipose tissue.

[0280] As mentioned previously, in black blood images, lesions (also called scars) are often difficult to distinguish from tissues such as adipose tissue. Thus, the method according to the invention advantageously allows for the initial detection and characterization of adipose tissue in the heart from one or more black blood images, so as to subsequently distinguish the areas corresponding to one or more lesions.

[0281] Thus, in a preliminary SP segmentation step, an area likely to contain lesions and / or adipose tissue is detected in at least one ISNk black blood image so as to generate positioning data for at least one cardiac lesion and cardiac adipose tissue possibly present in the figure, the preliminary SP segmentation using positioning data for the first wall L1 and possibly those for the second wall L2, generated during the SEG segmentation step.

[0282] Then, the step of characterizing CAG of cardiac adipose tissue and / or detecting DEG of cardiac adipose tissue is implemented.

[0283] The results of these steps allow the CAE lesion characterization and / or DE lesion detection step to be implemented.

[0284] The CAE lesion characterization step may include the following steps: - CTA calculation of CIC lesion size from the positioning data of the first wall L1 and possibly the second wall L2 from the SEG segmentation step, and possibly data from the step CAG characterization of adipose tissue and / or DEG detection of adipose tissue - CTT calculation of at least one degree of TR transmurality of the CIC lesion from the first wall L1 and the second wall L2 resulting from the SEG segmentation step.

[0285] These calculations are performed using a set of at least one first (or second and / or other in the "Dixon black" embodiment) image in black blood.

[0286] By size of a lesion, we mean a data representative of the dimensions of the lesion, such as a volume or an area, for example, or a number of pixels or voxels. Preliminary segmentation SP

[0287] The preliminary SP segmentation uses one or more first (or second or other in the "Dixon black" embodiment) black blood image(s) ISNk, ISN2k, and positioning data of the first wall L1 and possibly those of the second wall L2 from the SEG segmentation.

[0288] This positioning data can be positioning data generated during the SEG segmentation step or positioning data from the REP reporting step. Alternatively, the preliminary segmentation step SP includes the reporting step.

[0289] This step makes it possible to locate a Zc zone likely to contain both cardiac lesions and adipose tissue, i.e. to generate localization data for this Zc zone.

[0290] These location data include, for example, the identification or positions of the pixels or voxels corresponding to this Zc zone.

[0291] The preliminary segmentation step SP is advantageously implemented by thresholding.

[0292] It advantageously includes the identification of pixels or voxels exhibiting an intensity greater than or equal to a predetermined intensity threshold only in a predetermined area of ​​at least one first (or second and / or other in the "Dixon black" embodiment) black blood image (ISNk, ISN2k) delimited by the first wall L1 and / or the second wall L2. Indeed, as can be deduced from [Fig. 2], lesions and adipose tissue exhibit, on black blood images, a high intensity compared to healthy myocardium and blood.

[0293] This zone is determined from the positioning data of the first wall L1 and possibly those of the second wall L2 from the SEG segmentation.

[0294] This refers, for example, to the area of ​​a first ISNk black blood image delimited by the pixels or voxels of the first wall L1 and / or the pixels or voxels of the second wall L2 reported on this first image black blood ISNk.

[0295] Advantageously, the area of ​​the first image in black blood is the area surrounded and delimited by the first wall Ll.

[0296] In other words, the preliminary segmentation step SP includes the search for pixels or voxels with an intensity greater than or equal to a predetermined intensity threshold only in the area delimited and enclosed by the first wall L1 on one or more first (or second or other in the embodiment) "Dixon "black") black blood image(s) ISNk, ISN2k or ISNik. In other words, these pixels or voxels are taken only from the pixels or voxels of an area of ​​the first (or second or other in the "Dixon "black" embodiment) black blood images surrounded and delimited by the first wall L1. This makes it possible to avoid the erroneous detection of lesions beyond the epicardium, by avoiding confusion between lesions and the fat surrounding the epicardium and represented, on the first (or second or other in the "Dixon "black" embodiment) black blood images, by high intensity pixels.

[0297] Alternatively, the Z zone is the zone of the first or first (or second or other in the "Dixon "black" embodiment) black blood images ISNk delimited by the first wall L1 and by the second wall L2.

[0298] In other words, the preliminary segmentation SP includes searching for pixels with an intensity greater than or equal to a predetermined intensity threshold only in the area delimited by the two walls L1 and L2 of one or more first (or second or other in the "Dixon "black" embodiment) black-blood images ISNkISN2k. This variant has the advantage of identifying pixels or voxels of lesions and / or myocardial adipose tissue only. Indeed, some patients have necrosis of the papillary muscles (located in the area delimited by the L2 wall). In these patients, the muscles appear white on the black-blood image, which can lead to errors in lesion characterization when segmenting lesions throughout the entire area delimited by L1.

[0299] Alternatively and / or in addition, the preliminary SP segmentation includes the search for pixels with an intensity greater than or equal to a predetermined intensity threshold only in the area surrounded by the L2 wall. This step makes it possible to identify the pixels or voxels of the papillary muscles only.

[0300] Alternatively, the preliminary SP segmentation is implemented using a second learning function, for example, a neural network, for example, a convolutional neural network trained to segment cardiac lesions in an area delimited by the walls L1 and / or L2 when it receives as input the positioning data of the corresponding wall(s) from the first (or second or other in the "Dixon black" embodiment) black blood image, or by using at less an active contour segmentation algorithm, that is, a segmentation algorithm using an active contour model.

[0301] A common detection step (DEC) is implemented, comprising detecting the absence or presence of the Z-zone that may contain lesions and / or adipose tissue using a first (or second or other in the "Dixon black" embodiment) blood-black ISNk, ISN2k image and positioning data of at least one wall, for example, the second wall L2. This step outputs an indication of the presence or absence of this Z-zone.

[0302] The common detection step (CDS) can be performed by thresholding or using a neural network, similar to the segmentation step (SEG). It can consist of determining whether a number of contiguous pixels or voxels exceeding a predetermined threshold exhibits an intensity exceeding a predetermined threshold within the area delimited by the L1 and / or L2 walls, the positioning of which is defined during the myocardial segmentation step (SEG). The presence of the Zc zone, which may contain cardiac lesions and / or adipose tissue, is detected if this condition is met, and the absence of this Zc zone is detected if this condition is not met.

[0303] The neural network is, for example, a convolutional neural network. The neural network is, for example, trained to detect the presence or absence of cardiac lesions in an area delimited by the L1 and / or L2 walls when it receives as input the positioning data of the corresponding wall(s) and the black blood image.

[0304] DEG detection and / or CAG characterization of cardiac adipose tissue

[0305] These steps are implemented using an IG fat image generated from of images generated in the previously described process. These steps allow, among other things, the determination of secondary data on the location of adipose tissue.

[0306] The determination of the second adipose tissue localization data is implemented using a Dixon method. The Dixon method is generally used to distinguish adipose tissue from aqueous areas in images of biological samples. It thus allows for obtaining, on the one hand, an image containing only the adipose tissue of the biological sample studied, and on the other hand, an image containing only the aqueous areas of the biological sample studied.

[0307] To do this, the process includes a GENG generation step of an IG fat image comprising a first step GENG1, a second step GENG2 and a third step COMB which will be detailed below.

[0308] Dixon's method uses as input a first image of the sample under study where the adipose tissues and aqueous areas are in phase, and a second image of the sample under study where the adipose tissues and aqueous areas are out of phase. In a Alternatively, the method also receives at least one third image as input.

[0309] We describe here the case where only a first image and a second image are received as input.

[0310] In the "Dixon white" embodiment, the first image and the second image consist of the first image in white blood ISB, corresponding to an echo duration TE1, and the second image in white blood ISB2, corresponding to an echo duration TE2.

[0311] In the "Dixon black" embodiment, the first image and the second image consist of the first image in black blood ISN, corresponding to the echo duration TE1, and the second image in black blood ISN2, corresponding to an echo duration TE2.

[0312] Since protons rotate at different speeds in adipose tissue and in water, it is possible to find TE1 and TE2 echo durations such that the adipose tissue and aqueous areas are either in phase or out of phase. Typically, the time difference between the TE1 and TE2 echo durations is very small, resulting in differences in phase information and very small differences in magnitude information. For example, the TE1 and TE2 echo durations are approximately equal to 1.2 ms.

[0313] A first phase image and a first magnitude image are generated in a first step GENG1 from the first image II. A second phase image and a second magnitude image are generated in a second step GENG2 from the second image 12.

[0314] The first phase image, the first magnitude image, the second phase image, and the second magnitude image are then combined in a third COMB step so as to obtain a so-called IG fat image.

[0315] The GI fat image contains only information about the adipose tissue of the sample studied.

[0316] To obtain the IG fat image, information is typically subtracted from images where adipose tissue and aqueous areas are out of phase from information from images where adipose tissue and aqueous areas are in phase.

[0317] For a detailed description of the combination of the first phase image, the first magnitude image, the second phase image, and the second magnitude image to obtain the IG fat image, reference may be made to the article “Multiecho Dixon fat and water separation method for detecting fibrofatty infiltration in the myocardium”, Kellman P, Hernando D, Shah S, et al., Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine, 2009, vol. 61, no 1, p. 215-221.

[0318] In the variant where the method also receives a third image as input, the The first, second, and third images are combined to obtain the GI fat image. An example of this combination is described in the article "Three-point Dixon technique for true water / fat decomposition with b inhomogeneity correction," GH Glover and E. Schneider, Magnetic Resonance in Medicine, 18, 1991.

[0319] Once the GI fat image has been generated, it is possible to process it in order to detect and characterize the adipose tissues that are visible in it.

[0320] The DEG detection step includes detecting the absence or presence of adipose tissue using the IG fat image and positioning data of at least one wall, for example, the second wall L2. It generates as output an indication of the presence or absence of adipose tissue.

[0321] The DEG detection step can be performed by thresholding or using a neural network, as in the preliminary segmentation step SP. This step can consist of determining whether a number of contiguous pixels or voxels exceeding a predetermined threshold exhibits an intensity exceeding a predetermined threshold.

[0322] Advantageously, a propagation step of the walls detected during the SEG segmentation step on the IG fat image can be implemented, so as to differentiate epicardial adipose tissue from adipose tissue that has infiltrated the myocardium and to focus on the latter during clinical observation. In this case, a second REP transfer step, i.e., propagation, comprising the identification, on an IG fat image, of the pixels or voxels corresponding to the L1 and L2 walls identified during the SEG segmentation, is implemented.

[0323] The DEG detection step advantageously includes identifying pixels or voxels with an intensity greater than or equal to a predetermined intensity threshold only within a predetermined area of ​​the fat image delimited by the first wall L1 and / or the second wall L2. In other words, it is determined whether a number of contiguous pixels or voxels exceeding a predetermined threshold have an intensity greater than a predetermined threshold in this area. The presence of myocardial-specific adipose tissue is detected if this condition is met, and the absence of myocardial-specific adipose tissue is detected if this condition is not met.

[0324] The neural network is, for example, a convolutional neural network. The neural network is, for example, trained to detect the presence or absence of cardiac adipose tissue in an area delimited by walls L1 and / or L2 when it receives as input the positioning data of the corresponding wall(s) and the IG fat image.

[0325] In [Fig.7], in the middle, is represented an image of IG fat, where the pixels showing the presence of adipose tissue specific to the left myocardium are represented in Diagonal hatching. These pixels come from the DEG detection step implemented on the IG fat image.

[0326] The pixels of the IG fat image where the presence of adipose tissue is detected can be identified. For example, the coordinates of these pixels can constitute the second data for the localization of adipose tissue.

[0327] Detection of DE lesions and / or geometric characterization of CAE lesions

[0328] These steps use the first myocardial lesion localization data and the second adipose tissue localization data as will be described below.

[0329] In the "fusion mode" embodiment, a first fused image is generated by combining a first reference white blood image, as defined above, with the first myocardial lesion localization data. For example, the first myocardial lesion localization data may correspond to the pixel coordinates of a first black blood image with an intensity greater than a certain value. As mentioned above, in black blood images, the contrast is very high between the pixels or voxels corresponding to blood and healthy myocardium, which are black, and the pixels or voxels corresponding to lesions and adipose tissue, which are generally white.

[0330] Then, a second fused image is generated by combining the first fused image with the second adipose tissue positioning data.

[0331] For example, as seen previously, the pixels of the IG fat image where the presence of adipose tissue is detected can be identified by their coordinates, which can advantageously constitute the second data for the localization of adipose tissue. These pixels can be colored in a first color.

[0332] Also, the pixels corresponding to the first myocardial lesion localization data can be colored in a second color different from the first color.

[0333] Since the white blood images allow for perfect visualization of cardiac anatomy, the second fused image advantageously includes cardiac anatomy, colored pixels or voxels in the first color corresponding to adipose tissue, and colored pixels or voxels in the second color corresponding to lesions and / or adipose tissue.

[0334] By playing in particular on the contrast of the first reference black blood image and the fat image, for example by adjusting the detection threshold, it is possible to adjust the distribution of colours in the second merged image in order to accentuate the visualization of one or another structure.

[0335] Thus, in the event of overlap of certain high-intensity pixels of the first reference black blood image with other colored pixels of the fat image, it It is possible to increase the intensity of a color between the first color and the second color in order to preferentially distinguish cardiac lesions (or conversely adipose tissue).

[0336] In the "segmentation mode" embodiment, the first (or second or other in the "Dixon black" embodiment) black blood image(s) ISNk, ISN2k used during the preliminary segmentation step SP, the positioning data of the first wall L1 and possibly those of the second wall L2 from the SEG segmentation, and the data from the DEG detection and / or CAG characterization steps of the heart's adipose tissue are used.

[0337] For example, the pixels or voxels of the fat image IG where the presence of adipose tissue has been detected can be identified on the first (or second or other in the "Dixon black" embodiment) black blood image(s) ISNk, ISN2k so as to constitute a first group of pixels or voxels G.

[0338] These pixels or voxels can have the same respective positions on the IG fat image and on the first (or second or other in the "Dixon black" embodiment) black blood image(s) ISNk, ISN2k used during the preliminary segmentation step SP, when these images are considered to be spatially registered and because these images have the same size and resolution.

[0339] Thus, the pixels or voxels identified during the preliminary segmentation step SP but not belonging to the first group of pixels or voxels G together form a second group L that can be identified as corresponding to one or more cardiac lesions. The lesion detection step DE can consist of identifying the pixels in this second group.

[0340] In [Fig.7] on the left, the first ISNk black blood image of the heart section of [Fig.6] can be observed. On this image, the pixels or voxels identified as having an intensity greater than or equal to a predetermined intensity threshold are visible in squares only in the predetermined area delimited by the first wall L1 and / or the second wall L2.

[0341] For example, it is possible to annotate or color differently the pixels (or voxels) of the first group G and the second group L on the first(s) (or second(s) or other(s) in the "Dixon black" embodiment) black blood image(s) ISNk, ISN2k used during the preliminary segmentation step SP.

[0342] On [Fig.7] on the right, we can observe a lopt image obtained from the first black blood ISNk image of the heart section of [Fig.6] following such an annotation, in diagonal hatching for the pixels of the first group, corresponding to the presence of adipose tissue, and in squares for the pixels of the second group, corresponding to the presence of lesions. Lesion size calculation

[0343] CAE lesion characterization advantageously includes a CTA calculation step of lesion size.

[0344] This CTA calculation step includes the determination of at least one elementary data representative of the size of at least one cardiac lesion, for example of the myocardium, using location data of the location data of the first and / or second walls L1, L2 which are for example directly the positioning data from the SEG segmentation of the myocardium or data from these data, for example, data from the transfer step or lesion positioning data obtained during the DE lesion detection and CAE lesion geometric characterization steps.

[0345] On [Fig.8], a first ISNk black blood image has been represented, on which the L1 and L2 limits identified during the SEG segmentation and reported, i.e. propagated, on this first ISNk black blood image have been represented in thick lines, as well as the pixels identified, during the DE lesion detection step, as being pixels of a lesion.

[0346] A representative elementary datum of a lesion size may be a percentage of a myocardial surface occupied by a lesion in a SEC sector of a first (or second or other in the "Dixon black" embodiment) black blood image ISNk, ISN2k, ISNik or a volume or mass of the lesion in this SEC sector originating from axis 1 parallel to the p-axis and passing substantially through the center of the cardiac cavity on the ISNk black blood image and delimited by two rays R originating from axis 1 as seen on the first ISNk black blood image

[0347] The percentage of the myocardial surface occupied by the lesion in the SEC sector can be calculated from the ratio between the number of pixels corresponding to the lesion in this SEC sector and the number of pixels corresponding to the myocardium in this SEC sector.

[0348] The number of pixels corresponding to the lesion in this SEC sector can be calculated from the localization data obtained during the lesion segmentation step or can be calculated directly during the CTA calculation step, for example by selecting, by thresholding, the number of pixels having an intensity greater than a predetermined threshold in the portion of the SEC sector delimited by the walls L1 and L2 or by the wall LL. The CTA step can include the calculation of a data representative of the lesion size in a SEC sector from several elementary data representative of the calculated lesion size, in this SEC sector, for several first (or second or other in the "Dixon black" embodiment) black blood images ISNk, ISN2k distributed along the p-axis.

[0349] For example, a combination or average of the elementary data is calculated.

[0350] The lesion volume on a SEC sector can be calculated from the ratio between the number of pixels corresponding to the lesion on this sector and the number of pixels corresponding to the myocardium on this sector, from the thickness the slice corresponding to a first (or second or other in the "Dixon black" embodiment) image in black blood ISNk, ISN2k, when the image is two-dimensional.

[0351] The size and / or volume are advantageously also calculated from the predetermined resolution of the images.

[0352] It should be noted that the density of the myocardium is 1.06 g / mL. Therefore, the mass of a lesion is considered to be approximately equal to its volume, which allows for the estimation of the lesion's mass.

[0353] The CTA calculation step may, for example, include dividing the first (or second or other in the "Dixon black" embodiment) black blood image ISNk, ISN2k into a first predefined number, equal to 12 in the non-limiting example of Figure 1, of predefined SEC sectors of the same angle oAk opening pointing towards axis 1 and calculating the percentage of the myocardial areas occupied by the lesion over the different SEC sectors.

[0354] The first number of sectors and the opening angle "U" can vary depending on the cutting plane PCk. For example, the closer the cutting plane PCk is to the apex along the major axis, the lower the number of sectors and the higher the opening angle "1^".

[0355] This step can be implemented for different first (or second or other in the "Dixon black" embodiment) black blood images ISNk, ISN2k of different slices centered on respective slice planes PCk with k = 1 to K. The method advantageously includes, a GENR generation step, by computer, for example by the processing unit, of a set of at least one representation of lesion and / or fat geometric characterization data and an AFFD display step of at least one representation of the set of at least one representation on a screen of the human-machine interface.

[0356] The generation step includes, for example, the generation of data representative of the result of the DE lesion detection step and / or the DEG cardiac adipose tissue detection step, i.e. the absence or presence of lesion and / or cardiac adipose tissue, and the display step includes the display of this data.

[0357] As mentioned previously, the lopt image of [Fig.7] on the right, was obtained from the first ISNk black blood image of the heart section of [Fig.6] following an annotation, in diagonal hatching for the pixels of a first group of pixels G, corresponding to the presence of adipose tissue, and in squares for the pixels of a second group L, corresponding to the presence of lesions.

[0358] In another example, at least one representation of characterization data Geometric lesion and / or fat imaging involves a fusion of data from the first (or second) reference white blood image, the first reference black blood image, and a thresholded fat image resulting from a segmentation of the fat image. Fusion refers to the superimposition of these images.

[0359] In another example, the set of at least one representation advantageously includes a first REPT representation of the representative data or data of the lesion size calculated during the CTA calculation step.

[0360] For example, one can generate, as seen in [Fig.8], a bull's eye type representation of the percentages or data representing the percentages of the myocardial areas occupied by the lesion in different sectors of first (or second or other in the "Dixon black" embodiment) black blood images ISNk, ISN2k taken according to the respective PCk slice planes.

[0361] The Bull's eye representation is defined by the American Heart Association (AHA) in reference to the Anglo-Saxon expression "American Heart Association" and described in the following article: "Standardized Myocardial Segmentation and Nomenclature for Tomography Imaging of the Heart: A Statement for Healthcare Professionals From the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association.", Manuel D. Cerqueira et al., Circulation, 2002; 105:539-42.

[0362] The bullseye-type representation comprises a plurality of concentric circles C separated in pairs by rings C. Each ring CO is assigned to a slice or set of contiguous slices, such that the closer the CO ring corresponds to a slice or set of slices to the apex, the closer it is to the center of the circles. Each CO ring is divided into portions of PSE sectors in which are displayed, as in the example of [Fig. 7], the percentages of the myocardial surface occupied by a lesion and calculated for the respective sectors of the first (or second or other in the "Dixon black" embodiment) black blood image ISNk, ISN2k of the corresponding slice or combinations, for example, means, of percentages calculated from the percentages of the myocardial surface occupied by a lesion calculated for the sectors of the black blood images of the corresponding set of slices.

[0363] Alternatively and / or in addition, the intensity of the pixels in the different portions of the crowns depends on the calculated percentage. The lower this percentage, the higher the intensity of the corresponding crown.

[0364] For example, in [Fig. 8], the display is represented in the form of a known bullseye-type representation, showing the respective means of the percentages of the lesion size calculated in the respective sectors defined on three sets contiguous black blood images distributed along the p axis associated with the three respective crowns corresponding respectively to a section of the apex (inner crown), the mid-ventricle (middle crown) and the basal zone (outer crowns).

[0365] Portions of sectors associated with a percentage greater than 80% are represented in dotted lines and those associated with a percentage less than or equal to 80% are represented in white.

[0366] When generating a 3D image of the core, the image is advantageously divided into several layers along the p-axis and the same data is calculated from these different layers as from 2D images. Transmurality

[0367] The CAE lesion geometric characterization step advantageously includes a step for calculating a representative value of a lesion's transmurality percentage. Transmurality percentage refers to the percentage of myocardial thickness occupied by a lesion.

[0368] This CTT calculation step includes determining at least one data point representative of the percentage of transmurality of at least one myocardial lesion using lesion localization data and first and / or second wall localization data L1, L2 from the SEG segmentation step.

[0369] This data can be a percentage of the thickness of the myocardium occupied by a lesion on a sector of the first (or second or other in the "Dixon black" embodiment) black blood image ISNk, ISN2k starting from axis 1.

[0370] The percentage of myocardial thickness occupied by the lesion in the sector can be calculated from the ratio between the number of pixels corresponding to the lesion thickness in that sector and the number of pixels corresponding to the myocardial thickness in that sector. The number of pixels corresponding to the lesion thickness can be a number obtained from the results of the CAE lesion geometric characterization step or be calculated, for example by thresholding, during the CTT calculation step from the L1 and possibly L2 positioning data from the SEG segmentation step.

[0371] The number of pixels corresponding to the thickness of the lesion on a sector can be an average or a maximum of numbers of pixels, corresponding to the thickness of the lesion, calculated according to different radii of the sector.

[0372] The number of pixels corresponding to the myocardial thickness in a sector can be an average or a maximum number of pixels, corresponding to the myocardial thickness, calculated according to different radii of the sector. These numbers are calculated from positioning data of the L1 and L2 walls.

[0373] The CTT calculation step may, for example, include dividing the first (or second or other in the "Black Dixon" embodiment) image into a black blood ISNk , ISN2k in a second predefined number of sectors of the same angle a2k of opening pointing towards the center of the cardiac cavity and the calculation of the percentage of the myocardial surfaces occupied by the lesion on the different sectors.

[0374] The second number of sectors, and therefore the opening angle a2k, can vary depending on the cutting plane PCk. For example, the closer the cutting plane PCk is to the apex, the lower the number of sectors and the higher the opening angle d2k.

[0375] Advantageously the second number is greater than the first number.

[0376] The CTT step may include the calculation of a representative trans-murality data in a sector from several elementary representative trans-murality data calculated in that sector for several first (or second or other in the "Dixon black" embodiment) black blood images ISNk, ISN2k distributed along the p-axis.

[0377] For example, a combination or an average of the elementary data is calculated.

[0378] This step can be implemented for different first (or second or others in the "Dixon black" embodiment) black blood images ISNk> ISN2k of different sections centered on respective PCk section planes.

[0379] The GENR step advantageously includes the generation of a REPTR representation of data representative of a percentage of transmurality. The AFFD display step advantageously includes the display of this representation.

[0380] For example, one can generate a bullseye-type representation of combinations, for example of means, of percentages of the transmurality of the lesion in sectors of contiguous ISNk black blood image sets taken according to the respective PCk slice planes.

[0381] The intensity of the pixels in this image advantageously, but not necessarily, represents the percentage of transmurality.

[0382] For example, in [Fig. 8], the display is shown as a bullseye-type REPTR representation of the average lesion transmurality percentages calculated in sectors defined on several sets of contiguous black blood images taken from respective slice planes distributed along the P- axis.

[0383] Just as before, the bullseye type representation includes a plurality of sector portions whose intensity corresponds to the combination of the percentage of transmurality calculated for that sector portion.

[0384] The lower this percentage, the higher the intensity of the corresponding corona. Alternatively and / or in addition, the percentages are displayed in the sector portions.

[0385] The steps previously described for calculating lesion size and transmurality can advantageously be carried out for adipose tissue in such a way Similar based on the second set of data regarding the location of adipose tissue. Advantages

[0386] The proposed solution makes it possible to obtain images with sufficient resolution and contrast to detect and accurately characterize tissue singularities, for example lesions, in a reference area such as the heart, in a reliable and reproducible manner, and in particular to discriminate more accurately between tissue singularities of different natures, such as lesions and adipose tissue.

[0387] For example, by separating three important pieces of information—namely, the anatomy of the heart, adipose tissue, and lesions—onto three distinct images, namely, respectively, the first (or second or other) images in white blood and the first (or second or other) images in black blood and the fat images, it allows for the implementation of an automated lesion characterization process. This automation results in a significant gain in time and reproducibility compared to prior art solutions.

[0388] The black blood and white blood acquisition sequence of the method according to the invention requires a relatively short acquisition time, particularly when acquiring signals to generate 2D images that involve little computation. This advantageously allows the acquisition sequence to be implemented during breath-holding and limits cardiac movement between images, thus reducing the corrections required. This, in turn, limits computational resources and enables real-time implementation of the method. It also limits artifacts that impair image clarity. These artifacts increase the difficulty of reconstructing clear and precise images to locate and detect the lesion. Furthermore, long MRI acquisitions are uncomfortable for the patient. A duration of 10 to 20 minutes is considered very long, and it is difficult for the patient to remain still within the MRI scanner.

[0389] Furthermore, in the case of 2D image generation by the method according to the invention, artifacts are avoided in an image extracted from a cross-section of the 3D image, which could lead to situations where it is impossible to distinguish the presence of a potential lesion from the presence of blood located near the muscle. Indeed, in some cases, the lesion is so close to the blood—it is called subendocardial—that it is difficult to determine, in images exhibiting artifacts, whether it is a lesion, blood, or an image artifact.

[0390] In the case where the tissue singularities are lesions and adipose tissue, these can become intertwined in the case of several pathologies, and it is therefore interesting to observe this phenomenon, such as, for example, the presence of adipose tissue inside the myocardium, or the development of adipose tissue around old lesions.

[0391] The device and method according to the invention thus make it possible to improve the accuracy of clinical observations, and in particular to base a clinical decision with little risk of diagnostic error on the presence or absence of a lesion. Material

[0392] From a hardware point of view, the TC processing unit can be seen as a computer interacting with computer programs.

[0393] The TC processing unit includes at least one computer, for example, a microcomputer, a network of computers, an electronic component, a tablet, a smartphone or a personal digital assistant (PDA).

[0394] The processing and control unit TC includes, for example, a computer, comprising a set of at least one processor, and optionally a memory operationally coupled to the computer.

[0395] The memory includes, 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 the COI, CO2 communication system.

[0396] In other words, the computer-readable medium is a tangible medium. That is to say, it is not a transient signal in itself, 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.

[0397] By way of example, the readable medium is an optical disc, a magneto-optical disc, a read-only memory (ROM), an erasable and 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.

[0398] The readable medium may include an operating system and load the programs according to the invention. It includes registers adapted to store parameter variables created and modified during the execution of the aforementioned programs. A computer program containing software instructions is then stored on the readable medium.

[0399] Alternatively, the program instructions are taken from an external source and downloaded via a network. This is particularly the case for applications.

[0400] The processing and control unit includes a computer, i.e. at less 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 register memories or other types of display devices, transmission devices or storage devices.

[0401] The TC processing unit includes, for example, memories for storing data, for example black-blood and white-blood images, operationally coupled to the data processing circuit and a reader adapted to read a computer-readable medium.

[0402] The steps of the method according to the invention are, for example, carried out by causing the processing circuits of the processing unit TC to read predetermined programs stored on materials such as memories in such a way that their data processing circuits perform calculations, control communications and read and / or write data in memories.

[0403] The characterization is, for example, carried out 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).

[0404] The processing unit TC comprises at least one computer comprising at least the following elements: 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 a microcontroller and / or a digital signal processor (DSP)) ASICs capable of interpreting instructions in the form of a computer program and / or a hardware set such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a programmable logic device (PLD), programmable logic arrays (PLAs), a system-on-chip (SOC), and / or an electronic board in which steps of the method according to the invention are implemented in hardware elements.

[0405] The invention relates to a computer program product comprising the computer-readable medium containing instructions which, when executed by the processing circuit, cause the system S to implement the steps of the process according to the invention.

[0406] The product-program may include the computer-readable recording medium.

[0407] The invention also relates to a computer-readable medium on which the computer program is recorded.

[0408] Alternatively, the program instructions are taken from an external source and downloaded via a network. This is particularly the case for applications. In this case, the A computer program product includes a computer-readable data carrier on which program instructions are stored or a data carrier signal on which program instructions are encoded.

[0409] The form of program instructions is, for example, a form of source code, a computer-executable form, or any intermediate form between source code and a computer-executable form, such as the form resulting from the conversion of the source code via an interpreter, assembler, compiler, linker, or locator. Alternatively, program instructions are microcode, firmware instructions, state definition data, integrated circuit configuration data (e.g., VHDL), or object code. 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).

[0410] The communication unit includes at least one communication device enabling communication between system elements and optionally between at least one system element and a device external to the system. Communication systems can establish a physical link between system elements and / or between a system element and a device external to the system and / or a remote (wireless) communication link between system elements and / or between a system element and a device external to the system.

[0411] For example, at least one communication device includes at least one input interface configured to receive external data, such as first black blood signals, first white blood signals, second signals of the type taken from black blood and white blood, a plurality of N first black blood signals by black blood magnetic resonance, and a plurality of N first white blood signals by white blood magnetic resonance. This external data can be processed by the TC processing unit during a processing step implemented by the TC processing unit as described in this application.

[0412] For example, at least one communication device includes at least one output interface configured to return data from calculations or a processing step implemented by the TC processing unit as described in this application.

[0413] The at least one communication device may include any hardware, firmware and / or software suitable for communicating information between elements of the device to which the communication device belongs, for example via a data bus, or to an element external to the device. In order to to enable data communication between different devices which may include communication devices, these devices include hardware, firmware and / or software enabling a wired or wireless communication link to be established between them, for example Wi-Fi, Bluetooth, cellular or Ethernet.

[0414] The INT user interface allows a user to enter data or commands in order to be able to interact with the programs according to the invention.

[0415] The INT user interface includes, for example, an INTS output interface and an INTE input interface.

[0416] The input interface includes, for example, a keyboard or a pointing interface, such as a mouse, a light pen, a touchpad, a remote control, a speech recognition device, a haptic device.

[0417] The INTS output interface is designed to provide information to a user, either sensorially or electrically, such as, for example, visually or audibly. The output interface includes, for example, a display. The AFFD display stage may be a step for providing information by means other than a display.

[0418] The output interface INTS can be the input device INTE, for example, in the case of a touch tablet.

Claims

1. Demands Device for imaging a patient's heart, the heart comprising a myocardium delimiting a chamber of the heart, including: • At least one input interface configured for: • receive, during a pair of inter-beats comprising two consecutive inter-beats, initial signals in dark blood by magnetic resonance imaging (MRI) and in dark blood by late gadolinium enhancement at a first echo duration, • receive, during the inter-beat couple, the first signals in white blood by magnetic resonance imaging (MRI) in white blood by late gadolinium enhancement at the first echo duration, • receive, during the inter-beat couple, second signals of a type chosen from black blood and white blood by magnetic resonance imaging with late gadolinium enhancement at at least one other echo duration different from the first echo duration, • At least one processing unit (PU) configured to implement a computer-based processing step comprising: • generation of a first black blood image (BBI) from the first black blood signals, • generation of a first white blood cell image (WBC) from the first white blood cell signals, • generation of a second image of said type from the second signals of said type, • determination (DET), from the first black blood image (BBI), the second image of said type and possibly the first white blood image (WBBI), of initial localization data for a myocardial lesion and second localization data for adipose tissue, • At least one output interface configured to return said first myocardial lesion localization data and said second adipose tissue localization data.

2. Device according to claim 1, wherein at least one processing unit (PC) is configured to, during the processing step: • Generate a first fused image from a first reference white blood image derived from the first white blood image and the first myocardial lesion localization data.

3. Device according to the preceding claim, wherein: • At least one input interface is configured to receive, for a number N greater than 1 of inter-beat pairs, a plurality of N first black blood signals by black blood magnetic resonance and a plurality of N first white blood signals by white blood magnetic resonance, • At least one processing unit (PC) is configured to generate N first black blood images and N first white blood images, • the first reference white blood image is a combination of the N first white blood images.

4. A device according to the preceding claim, wherein at least one processing unit (PC) is configured to: • Generate a first phase image and a first magnitude image from a first reference image derived from the first image of said type, • Generate a second phase image and a second magnitude image from a second reference image derived from the second image of said type, • Generate a fat image from the first and second phase images and the first and second magnitude images, to determine, from the fat image, the said second data for the localization of adipose tissue.

5. A device according to any one of the preceding claims, wherein at least one processing unit (PC) is configured to: • during the determination of said first heart lesion localization data and said second adipose tissue localization data, segment a first reference image from the first white blood image, so as to generate positioning data for a set of at least one wall delimiting the myocardium, • implement a lesion and / or fatty geometric characterization (CAR) step of the heart, from said positioning data and said second adipose tissue localization data.

6. Device according to any one of the preceding claims in that they depend on claim 2, wherein at least one processing unit (PC) is configured to: • Generate a second fused image from the first fused image and the second adipose tissue localization data, in which the heart lesions are differentiated from the adipose tissue.

7. Device according to claim 5 or claim 6 as it depends on claim 5, wherein: • the assembly of at least one wall comprises a first wall delimiting and surrounding the myocardium, • the at least one processing unit (PC) is configured to implement, during lesion and / or fat geometric characterization, a preliminary segmentation step (PS) to localize an area likely to contain myocardial lesions and / or adipose tissue on a first reference black blood image from the first black blood image using data from positioning data of the first wall.

8. Device according to the preceding claim, wherein said first myocardial lesion localization data are obtained from data obtained during the preliminary segmentation (PS) stage and said second data on the localization of adipose tissue.

9. Device according to any one of the preceding claims, wherein at least one output interface comprises a display unit configured to display a first representation of the first myocardial lesion localization data in a first color and a second representation of the second adipose tissue localization data in a second color different from the first color.

10. Magnetic resonance imaging system comprising a magnetic resonance system (A) configured to implement an acquisition step and a device according to any one of the preceding claims, said acquisition step comprising: • During said inter-beat couple, acquisition of said first black blood signals by late gadolinium-enhanced black blood magnetic resonance at said first echo duration, • During said inter-beat couple, acquisition of said first white blood signals by late gadolinium-enhanced white blood magnetic resonance at said first echo duration, • During said inter-beat couple, acquisition of said second signals of a type taken from black blood and white blood by late gadolinium-enhanced magnetic resonance at said at least one other echo duration different from the first echo duration.

11. Magnetic resonance imaging system according to the preceding claim, wherein the magnetic resonance system (A) is configured to implement: • a step of acquiring second white blood signals by magnetic resonance by late gadolinium enhancement at the second echo duration, and in which at least one processing unit (TC) is configured to: • generate a second white blood image from the second white blood signals.

12. Magnetic resonance imaging system according to any one of claims 10 or 11, wherein the magnetic resonance system (A) is configured to implement: • a second black blood signal acquisition step by magnetic resonance late gadolinium enhancement at the second echo duration, and wherein at least one processing unit is configured to: • generate a second black blood image from the second black blood signals.

Images