Method and system for perfusion imaging

Abstract

A method for perfusion imaging in a region of interest of an individual, the method comprising a computer-implemented processing method, the processing method comprising: ■ Generating intensity-time curves from pixel / voxel intensities of N blocks of J elementary ultrasound images of the region of interest generated from raw ultrasound data, ■ Identifying a reference intensity-time curve of a predetermined reference zone of the region of interest, ■ Generating at least one perfusion indicator from the reference intensity-time curve.

Description

[0001]PERFUSION IMAGING METHOD AND SYSTEM Technical Scope The technical scope is that of parametric ultrasound perfusion imaging. Perfusion is an essential physiological parameter in medicine due to its fundamental role in maintaining tissue health and function. In medical practice, perfusion assessment is an essential tool in various specialties such as cardiology, neurology, oncology, and intensive care. Monitoring perfusion levels allows for the identification of tissues that are at risk of damage or dysfunction. State of the Art Methods for generating perfusion indicators using contrast-enhanced ultrasound imaging are known.Ultrasound imaging refers to a type of ultrasound imaging in which ultrasound data from a region of interest are acquired in the presence of a contrast agent, for example, in the form of microbubbles, previously injected into the blood vessels. From the article “Imaging of perfusion using ultrasound”, Cosgrove et al., Eur J Nucl Med Mol Imaging (2010) 37 (Suppl 1):S65–S85, we find a method for generating perfusion indicators such as arrival times or times to peak, classically called “time to peak” or TTP in English terminology. These indicators are calculated from curves representing the intensity of the pixels in the ultrasound images as a function of time. These curves are called time-intensity curves or TICs, referring to the English expression “Time Intensity Curve”. However, the measurements obtained are not very precise.One aim is to propose a solution for generating relatively accurate perfusion indicators. In summary, the invention relates to a perfusion imaging method for a region of interest in an individual. The method comprises a computer-implemented processing method, the processing method comprising: Generating intensity-time curves from pixel / voxel intensities of N blocks of J elementary ultrasound images of the region of interest generated from raw ultrasound data; Determining, from the intensity-time curves, a reference intensity-time curve for a predetermined reference area of ​​the region of interest; Generating at least one perfusion indicator from the reference intensity-time curve. In one embodiment, the processing method comprises: Displaying information representative of a perfusion indicator on a screen.In one embodiment, J is greater than 3, greater than or equal to 8, or greater than or equal to 10. In another embodiment, the J elementary images of each block are generated from J sub-blocks of raw ultrasonic data acquired sequentially, each elementary image being generated from a single sub-block of raw ultrasonic data, and the N blocks being generated from N blocks of J sub-blocks of raw data, the N blocks of J sub-blocks of raw data being acquired sequentially, and each block of J elementary images being generated from a single block of sub-blocks of raw data. In another embodiment, generating the intensity-time curves comprises, for each of a plurality of the N blocks of J elementary images, generating a combined image of the region of interest by combining J images representative of the J elementary images of said block.In one embodiment, the combination includes calculating a temporal and / or spatial cross-combination of the J representative images of the J elementary images. In one embodiment, generating at least one perfusion indicator includes deconvolving at least one of the intensity-time curves from the reference intensity-time curve. In one embodiment, generating the intensity-time curves includes: enhancing the signals from the contrast agent relative to the signals from the surrounding tissues so as to generate N blocks of J enhanced images from the N blocks of J elementary images.According to one embodiment, generating the intensity-time curves includes:^ improving the signals from the contrast agent relative to the signals from the surrounding tissues so as to generate N blocks of J enhanced images from the N blocks of J elementary images such that the images representing the J elementary images are representative of the J enhanced images.According to one embodiment, the generation of intensity-time curves includes generating N blocks of J refined images from N blocks of initial J images such that the refined images have a less broad dot spread function than the initial images, the N blocks of J initial images being the N blocks of elementary J images or the N blocks of J initial images being N blocks of J improved images obtained from the N blocks of J elementary images by improving the signals from the contrast agent relative to the signals from the surrounding tissues so as to generate the N blocks of J improved images.In one embodiment, generating the intensity-time curves comprises: generating N blocks of J refined images from the N blocks of J enhanced images such that the refined images have a narrower point spread function than the enhanced images, and so that the representative images of the J elementary images are representative of the J refined images. In one embodiment, the method comprises: acquiring the raw ultrasonic data using an acquisition system comprising a transducer array; and generating the N blocks of J elementary images from the raw ultrasonic data. In one embodiment, the blocks of J sub-blocks of raw data are acquired at a frequency between 5 Hz and 500 Hz, and the sub-blocks of raw data are acquired at a frequency between 200 Hz and 20,000 Hz.In one aspect, the invention relates to a perfusion imaging system configured to implement the method according to the invention, the system comprising a processing system including a processing unit comprising at least one processor, the processing unit being configured to implement the processing method. In one embodiment, the system comprises an acquisition system including a transducer array, the acquisition system being configured to acquire raw ultrasound data and to generate elementary images from the raw ultrasound data. In another embodiment, the acquisition system is configured to acquire N blocks of J sub-blocks of raw ultrasound data, the processing system being configured to generate N blocks of elementary images from the N blocks of J sub-blocks of raw ultrasound data.The invention also relates to a computer program product comprising instructions that lead the system according to the invention to execute the steps of the process according to the invention. The invention also relates to a computer-readable medium on which the computer program is recorded.Brief description of the figures: Other features and advantages of the invention will become apparent from the detailed description that follows, with reference to the attached figures, which illustrate: - Figure 1: a flowchart of the steps of an example of a processing method according to the invention, - Figure 2: N blocks of J elementary images, - Figure 3: examples of intensity-time curves, - Figure 4: in (a), an AIF and a TIC and in (b) and in (c) examples of perfusion indicators, - Figure 5: on the left an example of an elementary image in the middle an example of an enhanced image, on the right an example of a combined image, - Figure 6: an example of an individual acquisition sequence, - Figure 7: an example of elementary acquisition, - Figure 8: an example of the acquisition of a sub-block of data, - Figure 9: in block form, the hardware elements of an example of a system configured to implement the method according to the invention.Detailed Description: A parametric ultrasound perfusion imaging method is proposed. This imaging method comprises generating a blood perfusion indicator or parameter in a region of interest R of an individual from ultrasound images of the region of interest R generated from raw data acquired in the presence of an ultrasound contrast agent in the region of interest R. The individual may be human or animal. The region of interest is, for example, an area of ​​the individual's brain, but it may be any area of ​​the body, preferably vascularized. The invention also relates to an ultrasound imaging system configured to acquire the raw ultrasound data and to generate said at least one perfusion indicator. The method comprises, as shown in Figure 1, the acquisition of raw ultrasound data of the region of interest R by an ultrasound probe S, visible in Figure 6, comprising an array RE of transducers TR.Acquisition 10 of the raw ultrasound data is performed in the presence of contrast agents in the region of interest R. In one embodiment, acquisition 10 of the raw ultrasound data is performed after intravenous injection into the individual of a contrast agent comprising individual contrast agents, for example, in the form of gas microbubbles, such that contrast agents are present in the region of interest R during acquisition 10. These contrast agents injected into the body are exogenous. Alternatively, acquisition 10 is performed without prior injection of a contrast agent. The contrast agents present in the region of interest are then endogenous, for example, red blood cells. In the following text, the term microbubbles will be used, given that the procedure applies to any type of contrast agent.The process then includes a processing step T comprising in particular the generation 20 of ultrasound images of the region of interest R from the raw ultrasound data acquired during the acquisition 10 and the generation 100 of at least one perfusion indicator from the ultrasound images generated during step 20. During the generation step 20, N blocks Bi are generated, with i = 1 to N where i is an integer and N is an integer greater than 1, each formed of J elementary ultrasound images IEi,j with j = 1 to J, j being an integer and J being an integer greater than 1, of a predetermined region of interest of an individual. Blocks Bi of J elementary images IEi,j are represented in figure 2. Thus, during generation 20, we generate N*J elementary images IEi,j (with i= 1 to N and j = 1 to J) of the region of interest R. According to a non-limiting realization mode, the elementary images IEi,j are B-mode images.Other types of elementary ultrasonic images can, of course, be used. By elementary image IEi,j, we mean a grid of pixels or voxels whose respective intensities represent acoustic properties of the medium surrounding the probe at their respective coordinates. The coordinates of a pixel or voxel represent a position relative to the probe of a point in the region of interest. In the following text, each image is considered to be a grid of pixels, but the invention is of course applicable where the images are composed of voxels. The processing T then comprises the generation of at least one perfusion indicator from the elementary images IEi,j.Generation of Time-Intensity Curves: Step 100 of the perfusion indicator generation process includes the generation of time-intensity curves (TICs), also known as time-intensity curves, from the elementary images IEi,j, where i = 1 to N and j = 1 to J. This step is implemented automatically, i.e., by computer. The TIC (x, y, t) time-intensity curves represent the dynamics of the microbubbles. The shapes of these curves provide valuable information about the health of the tissues, particularly the different vessels. For example, the intensity can take on a set of M values ​​(M being a finite integer greater than 1) corresponding to M gray levels ranging from 0 to M-1. For example, this value can take 256 values ​​between 0 and 255, but M is not limited to 256.Intensity-time curves generated for pixels with different coordinates TIC(x1, y1, t), TIC(x2, y2, t), TIC(x3, y3, t), and TIC(x4, y4, t) are shown in Figure 3. This step 110 involves generating, for each pixel with predetermined spatial coordinates (x, y), an intensity-time curve TIC(x, y, t) representing the evolution of the pixel's intensity over time. This curve is generated from the elementary images IEi,j, with i = 1 to N and j = 1 to J. The time t is defined as a function of i and the acquisition frequency, denoted FB, of the blocks Bi, such that for each i, t = FB*(i-1). This yields Z TIC(x,y,t), where Z is the number of pixels in the elementary images IEi,j.Identification of the Reference Intensity-Time Curve: According to the invention, the method then comprises a step 120 of automatic determination, i.e., by computer, of a reference intensity-time curve corresponding to an intensity-time curve of a predetermined reference zone of the reference region R. This step is implemented using the Z TIC (x, y, t). In one embodiment, the reference intensity-time curve is the arterial input function (AIF). Advantageously, the reference zone is chosen as the main artery of the region, i.e., the one that supplies blood to the other vessels (veins / arteries) of the region of interest R. In another embodiment, the reference intensity-time curve is the venous output function (VOF).The reference area is then, for example, chosen to be the largest vein in the region of interest. In one embodiment, determining the arterial input function (AIF) involves automatically selecting the intensity-time curve that, among the Z TIC(x,y,t), has the fastest increasing intensity. To this end, the TIC that, among the Z TIC(x,y,t), has the steepest positive slope when its intensity level reaches a predetermined intensity ID is determined. For example, the predetermined intensity ID is given by: ID = c*IB, where c is a predetermined positive coefficient and IB is an average intensity of the image noise. For example, c is between 3 and 10. c is, for example, equal to 5. Alternatively, the intensity-time curve with the fastest increasing intensity is identified by selecting the intensity-time curve with the earliest absolute maximum among the Z TIC(x,y,t).In one embodiment, the VOF is determined by automatically selecting the intensity-time curve that, among the Z TIC(x,y,t), has the latest peak with an intensity exceeding a predetermined threshold. In another embodiment, the method includes, prior to the determination step 120, a step for selecting the predetermined reference area on a representative image of at least one elementary image IEi,j. According to one embodiment, the representative image of one or more elementary images IEi,j is a combined image ICi, generated from elementary images IEi,j as described later in the text. According to one embodiment, the reference area selection step is, for example, implemented automatically by computer. It is, for example, performed by morphological segmentation of the image based on prior knowledge of the morphology or anatomy of the area.Segmentation is implemented, for example, using a machine learning model trained to make this selection on a representative image of at least one elementary image of the region of interest R. Alternatively, segmentation is implemented by correlating the image with a predetermined model function. Alternatively, the selection of the reference region is performed manually by a user via an input user interface INTE of the system. Without limitation, the selection is performed by a pointing device on the input interface INTE. In this variant, step 100 includes displaying a representative image of one or more of the elementary images IEi,j, for example, displaying a combined image ICi, on a screen of an output interface INTS of the system's user interface. The user then manually segments the reference region using the pointing device on the image displayed on the INTS screen.In one embodiment, the data relating to the reference zone are then stored in the memory of a processing system so that the processing system can implement the identification step. This data is, for example, a set of pixels forming the reference zone. The process then includes the step of determining the arterial input function, or more generally the reference intensity-time curve, from the TIC(x,y,t) values ​​of the pixels within the reference zone. This step consists, for example, of automatically selecting the intensity-time curve that, among the TIC(x,y,t) values ​​of the pixels in the reference zone ZR, has the fastest increasing intensity, as explained previously. Alternatively, an average of the TIC(x,y,t) values ​​of the pixels in the reference zone ZR is calculated.Generation of a Perfusion Indicator According to the invention, step 100 comprises the generation 130 of at least one perfusion indicator from the reference intensity-time curve. This step is automatic. In other words, it is implemented by computer. Using the reference intensity-time curve to generate the perfusion indicator(s) allows for the normalization of the intensity-time curves of the different pixels, which makes it possible to generate perfusion indicators with absolute and therefore comparable values. Since the AIF is relative to a large artery, it is supplied directly by the heart. This makes it a good representation of the general microbubble concentration entering the tissues. Normalizing the intensity-time curves from the AIF thus allows for the characterization of only the behavior of the microbubbles in the tissues relative to their arrival in the artery corresponding to the AIF.Thus, arrival times and transit times in tissues are calculated from the arrival time in the artery, not from the start of acquisition. Furthermore, fluctuations due to systemic effects are removed from the signal, so that only the contributions of the tissues of interest are considered. In addition, the generation of indicators obtained by normalizing the intensity-time curves from the reference intensity-time curve allows for comparison of these indicators between individuals and for monitoring these indicators for the same individual over several days. Indeed, these indicators are independent of the image acquisition conditions. In particular, they do not depend on the time between the start of recording and the arrival time of microbubbles in the region of interest. Nor do they depend on potentially confounding physiological parameters, such as heart rate and cardiac output.Thus, medical analyses can be based on the generated perfusion indicators. In the following text, the reference intensity-time curve is assumed to be the arterial input function (AIF), but the description also applies to cases where the reference intensity-time curve is the flow rate (VOF). Alternatively, several reference intensity-time curves are determined. For example, the AIF and the VOF are determined. At least one indicator can then be determined from several reference intensity-time curves. In one embodiment, at least one normalized perfusion indicator is determined from the AIF. For example, the generation step includes a deconvolution step, using the AIF, of the intensity-time curves (TIC) associated with the different pixels, that is, associated with the spatial positions (x,y) of the different pixels.Thus the blood flow curve representing the evolution of blood flow in mL of blood per 100 g of tissue per minute at the spatial position (x,y) as a function of time is given by:. L es courbes ^^ model the relationship between the time-domain intensity curves of the pixels and the tissue perfusion dynamics. An example of such a curve is shown in Figure 4b. Other indicators can be generated during step 130. In the case where several reference intensity-time curves are determined, the process can include deconvolving the intensity-time curves by each or more of these reference curves to calculate several blood flow curves and a step of calculating an indicator from several of these blood flow curves. In one embodiment, at least one indicator is generated from the blood flow curves. .Step 130 may include the measurement or determination, for each of the spatial positions (x,y), that is, for each pixel, of a time to maximum, also called Tmax and denoted Tmax(x,y), referred to as "Time to Maximum" in Anglo-Saxon terminology. This is the instant associated with the maximum concentration of microbubbles (maximum flow?) for the pixel with spatial coordinates (x, y). Each time to maximum Tmax(x,y) is calculated from the corresponding blood flow curve. The time to maximum provides a standardized local measurement of tissue perfusion independent of temporal variations. Step 130 may include the measurement or determination, for each of the spatial positions (x,y), that is to say for each of the pixels, of a maximum flow rate. For example, in the brain, this maximum, for the pixel (x, y) is called cerebral blood flow or Cerebral Blood Flow (CBF) denoted CBF(x,y).The Tmax (x, y) and the CBF (x, y) are, for example, determined from the corresponding blood flow curve ^^ (^, ^, ^) by a maximum detection algorithm of the curve ^^ (^, ^, ^) or by curve fitting. Step 130 may include the measurement or determination, for each of the spatial positions (x,y), i.e. for each of the pixels, of a cerebral blood volume also called CBV and denoted CBV (x,y) called "cerebral blood volume" in Anglo-Saxon terminology. Each CBV (x,y) is measured from the corresponding blood flow curve ^^ (^, ^, ^), that is, measured for the same spatial position (x; y). The CBV (x,y) is the area under the curve ^^ (^, ^, ^). It should be noted that CBF and CBV are measured or determined when the region of interest is the brain or a vascularized part of the brain.More generally applicable to different vascularized areas, a blood volume (BV) and / or a blood flow rate (BF) are determined for different pixels. Step 130 can thus include the measurement, for each pixel, of a time to peak (TTP), also called Time to Peak (TTP), and denoted TTP(x, y). This indicator is, for example, the time between the microbubble injection time (t=0) and the maximum of the TIC corresponding to the pixel, denoted TIC(x, y). Step 130 can also include the measurement, for each pixel, of the mean transit time (MTT), also called Mean Transit Time (MTT), and denoted MTT(x, y), which is the average time for blood to cross the tissue, that is, for it to pass from the arterial side to the venous side. This average time is given for example by the full width at half maximum of the corresponding blood flow curve ^^ (^, ^, ^).In other words, ^^^(^, ^) = ^^^^(^^ (^, ^, ^)) Where ^^^^(^^ (^, ^, ^)) is the full width at half maximum (FWHM) of the blood flow curve ^^ (^, ^, ^). The list of perfusion indicators above is not exhaustive. Many other perfusion indicators can be generated, such as a volume (in pixels or voxels) whose Tmax is greater than a predetermined value. Generally, an indicator is determined for one pixel, for a plurality of pixels, or for each pixel as described above. Display According to one embodiment, the method includes displaying visual information representing at least one perfusion indicator on a screen of an INTS output interface of the system. The visual information of a perfusion indicator may include, but is not limited to, one or more numerical values, and / or a curve, and / or an image representing the perfusion indicator.Numerical information includes, for example, a numerical value of an indicator or a combination of indicators. For instance, one can display a numerical value of a pixel's Tmax and / or numerical values ​​of the Tmax values ​​calculated for different pixels. One or more curves, for example, of the TIC or BF type, can be displayed. A color or grayscale image can be represented with pixels / voxels whose intensities or colors depend on, or are the values ​​of, a parameter calculated for those pixels / voxels. For example, a Tmax map can be displayed as shown in Figure 4 in c. This map is a color or grayscale image with pixels whose intensities or colors depend on the Tmax(x,y) values ​​calculated for the (x,y) positions of the pixels. In one embodiment, information representing several indicators is displayed.In one embodiment, this information is displayed simultaneously on one or more screens of the processing system. In another embodiment, information is displayed superimposed. For example, the Tmax map can be displayed superimposed with a combined image ICi or an image reconstructed from the combined images. Enhancement: In one embodiment, step 110 includes an enhancement step 111. In the initial images IEi,j, the contribution of signals from microbubbles is mixed with signals from surrounding tissues. Enhancement 111 strengthens the signals from microbubbles and weakens the signals from surrounding tissues.Thus, enhancement 111 is implemented to improve the signals from microbubbles relative to the signals from surrounding tissues so as to generate N BAi blocks of J enhanced images IAi,j from the N Bi blocks of J elementary images IEi,j. The surrounding tissues include, for example, at least one tissue taken from the following: a blood vessel wall in which microbubbles circulate, blood circulating in the vessel, bone, fat, or tissue forming a neighboring organ. For each i, the J enhanced images IAi,j of the BAi block are generated from the J elementary images IEi,j of the Bi block.Enhancement 111 is implemented such that the contrast between the pixels of the J enhanced images IAi,j of block BAi representing microbubbles and the pixels of the J enhanced images IAi,j of block BAi representing the surrounding tissue is higher than the contrast between the pixels / voxels of the J elementary images IEi,j of block Bi representing the microbubbles and the pixels of the J elementary images IEi,j of block Bi representing the surrounding tissue. In one embodiment, each enhanced image IAi,j of block Bi is generated from an elementary image IEi,j of block Bi. Alternatively, each enhanced image IAi,j of block BAi is generated from several elementary images IEi,j of block Bi. For example, at least one or every enhanced image IAi,j of block BAi is generated from all the elementary images IEi,j of block Bi.Enhancement step 111 includes, for example, filtering each elementary image IEi,j to remove signals from the tissues in the elementary image and retain only the signals from the microbubbles. This step is known, for example, as "clutter filtering" or background removal. In one embodiment, enhancement step 111 includes applying temporal or spatiotemporal filters of the Singular Value Decomposition (SVD) type to separate the echoes from the microbubbles from the echoes from the tissues. Alternatively, each enhanced image IAi,j is generated by amplitude modulation of elementary images IEi,j from block Bi.For example, the intensity of each pixel with predetermined spatial coordinates in the enhanced image IAi,j is equal to a linear combination of the intensities of the pixels with the same spatial coordinates in a first elementary image of block Bi and a second elementary image of block Bi, the first and second elementary images having different order numbers in block Bi. For example, the intensity ^^^,^(^, ^) of the pixel with spatial coordinates (^, ^) in the enhanced image IAi,j is given by the following formula: ^^^,^(^, ^) = ^ ∗ ^^^,^−1(^, ^) + ^ ∗ ^^^,^+1(^, ^) Where ^^^,^(^, ^) is the intensity of the pixel with spatial coordinates (^, ^) in the elementary image ^^. ^,^and where a and b are coefficients. Alternatively, the intensity of each pixel of the enhanced image IAi,j is obtained by linear combination of the intensities of pixels having the same spatial coordinates of a first elementary image of order number j of block Bi and a second elementary image of order number j-1 of block Bi. The linear combination is, for example, a subtraction. Alternatively, the intensity of each pixel of the enhanced image^^^,^(^, ^) is obtained by linear combination of the intensities of pixels having the same spatial coordinates of a first elementary image of order numbers i and j and an average image being the average of the J elementary images IAi,j of block Bi. According to a particular embodiment, the acquisition 10 of the raw data is carried out in such a way as to be able to recombine the elementary images advantageously.For example, one can perform a phase inversion known as "Pulse Inversion" in Anglo-Saxon terminology, amplitude modulation, for example, amplitude modulation over long ensembles (AMLE), or beams emitted using CHIRP (Compressed High Intensity Radar Pulse), emission using a Golay code, or other equivalent strategies. Thus, the enhancement step, using amplitude modulation, employs a nonlinear strategy. At the end of the enhancement step 111, N BAi blocks are obtained, each consisting of J enhanced images IAi,j, where j = 1 to J. The enhancement step 111 is a computer-implemented step. In other words, the blocks N BAi, each having J enhanced images IAi,j with j= 1 to J, are automatically generated from the elementary images IEi,j with i= 1 to N and j= 1 to J. Alternatively, the process is devoid of an enhancement step.Refinement According to one embodiment, step 110 includes a refinement step 112 which is a generation step of refined images IAFi,j with j=1 to J and i=1 to N from the improved images IAi,j with j=1 to J and i=1 to N such that the refined images IAFi,j with j=1 to J and i=1 to N have a point spread function also called PSF, acronym for the Anglo-Saxon expression "Point Spread Function" less wide than that of the improved images IAi,j with j=1 to J and i=1 to N. During this step, a refined image IAFi,j is generated from each improved image IAi,j. Each refined image is generated from a single enhanced image IAi,j of the same order numbers i and j. Alternatively, each refined image IAFi,j is generated from a plurality of images IAi,j of the same block, i.e. presenting the same order number i.For example, each refined image IAFi,j is generated from the J images IAi,j of the same block, that is, those with the same sequence number i. This yields N blocks, denoted BAFi, each consisting of J refined images IAFi,j, where j = 1 to J. The point spread function (PSF) of an image refers to the intensity distribution of the image of a point object within the image. For example, the width of the point spread function of an image is the width or diameter of the image of the contrast agent within the image. Indeed, the contrast agent is considered to be a point. In the case where the contrast agent is a microbubble, the width of the point spread function (PSF) of an image is the width or diameter of the image of a microbubble in the image. Alternatively, the width of the PSF of an image can be measured from the intensity profile of the image of a point object present in the image.For example, the PSF width is the full width at half maximum (FWHM) of this intensity profile or any other equivalent indicator, such as the standard deviation of the intensity profile or of a Gaussian function fitted to this intensity profile. Alternatively, the PSF of an image can be estimated by known methods of analysis or blind deconvolution, allowing an approximation of the impulse response of the system that generated the image. Thus, the generation of each refined image IAFi,j from an initial image, for example, an enhanced image, results in a refined image IAFi,j that is sharper than the initial image, i.e., one with a higher spatial resolution. The refined image and the initial image have the same pixel or voxel density. In other words, the initial and refined images have the same number of pixels or voxels, and these pixels or voxels have the same dimensions.In other words, step 110 includes a refinement step 112, which is a generation step of refined images IAFi,j, where j = 1 to J and i = 1 to N, from initial images, for example, enhanced images IAi,j, where j = 1 to J and i = 1 to N, such that the refined images IAFi,j, where j = 1 to J and i = 1 to N, have a better spatial resolution than the initial images, for example, the enhanced images IAi,j, where j = 1 to J and i = 1 to N. The initial and refined images have the same resolution in pixels or voxels, that is, the same pixel or voxel density. In other words, the initial and refined images have the same number of pixels or voxels, and these pixels or voxels have the same dimensions.The generation of each refined image IAFi,j can be carried out using several known and equivalent methods, such as spatial deconvolution of the enhanced image IAi,j of the same order numbers i and j with a predetermined model PSF, by applying a binary mask corresponding to the local intensity maxima to the enhanced image IAi,j of the same order numbers i and j so as to generate a binary refined image, or by filtering the enhanced image IAi,j of the same order numbers i and j by applying a Jerman morphological filter. Alternatively, the refined images are generated from the enhanced images by implementing a fluctuation imaging processing method also called SOFI in reference to the Anglo-Saxon expression "Super-resolution Optical Fluctuation Imaging".For example, the spatial and / or temporal variations of the J enhanced images of a block of order number i are exploited to process the enhanced image IAi,j in such a way as to generate the refined image of the same order numbers i and j with a narrower PSF than the enhanced image. Alternatively, the refined images are generated from the enhanced images by applying an autocorrelation function to the enhanced images. For example, for each i = 1 in N and each j = 1 in J, a spatial and / or temporal autocorrelation function is applied to the enhanced image IAi,j to generate the refined image IAFi,j of the same order numbers i and j with a narrower PSF than the enhanced image. These examples are not limiting and can be used alone or in combination. When the process lacks an enhancement step, the refined images can be generated from the elementary images IEi,j instead of the enhanced images IAi,j.The refinement step 112 improves image resolution and refines microbubbles and, ultimately, the microbubble trajectories visible in the combined images, which will be described later. It also reduces the contribution of residual noise to the image. Refining the images results in combined images with particularly high spatial resolution, leading to high-resolution ICTs. This allows for reliable perfusion indicators. It also enables the detection of smaller abnormal areas from the generated perfusion indicators and better differentiation between signals related to large vessels and signals related to surrounding tissues. Furthermore, this refinement step is most effective when performed before the combination step 113. The refinement step is a computer-implemented, or automated, process.In another embodiment, the process lacks a refinement step. Combination According to one embodiment, step 110 includes a step 113 of combining J basic images IMi,j representing the J elementary images IEi,j of each of the blocks Bi, with i = 1 to N, so as to generate a combined image ICi for each block Bi, that is, so as to generate N combined images ICi with i = 1 to N. When step 110 includes the enhancement step 111 without the refinement step 112, the basic images IMi,j are the enhanced images IAi,j of the blocks BAi. When step 110 includes the refinement step, the basic images IMi,j are the refined images IAi,j of the blocks BAFi. When step 110 lacks the improvement and refinement steps, the basic images IMi,j are the elementary images IEi,j. The combination step 113 comprises N elementary combination steps 113i, with i = 1 to N.Each elementary combination step 113i consists of combining the J basic images IMi,j, with j = 1 to J, representing the elementary images IEi,j, with j = 1 to J of block Bi, to generate a single combined image ICi. This yields N combined images ICi, with i = 1 to N. These combined images ICi correspond to a sequence of N images ICi, where the index i corresponds to the image's position in the sequence. The elementary combination 113i is implemented so that the combined image ICi represents the trajectories of the microbubbles during the acquisition time of block Bi in the region of interest R. Thus, the combined image ICi represents the vasculature within the region of interest R. In other words, this step makes the blood vessels more distinct and visible by combining all the positions of the microbubbles that have explored the space.This allows for the reliable and precise detection of a target vascular zone based on morphological criteria in at least one combined image (ICi), which is not possible with a single image, an enhanced image, or a refined image representing not the trajectories of microbubbles during the Bi block acquisition but the "fixed" positions of the different microbubbles during a single acquisition. Furthermore, generating a combined image allows for the generation of spatially resolved and high-contrast TICs, i.e., images with a low noise level, enabling precise determination of the AIF and thus generating accurate and reliable perfusion indicators. Detection of the main artery, i.e., the largest artery within a region of interest, allows for the reconstruction of arterial entry function. The combined image (ICi) is generated by combining the basic images (IMi,j).In one embodiment, the combined image ICi is created by coherently summing the basis images IMi,j for j= 1 to J. For example, the coherent sum is a quadratic sum. Thus the intensity ^^^(^, ^) of each pixel of the combined image ICi is given by the following formula:. Where ^^^(^, ^, ^) is the intensity of the pixel with spatial coordinates (x,y) of the base image IM i,j More generally, we can generate the combined IC image i by performing a temporal and / or spatial cross-combination of the basic IM images i,j with j = 1 to J. This combination is, for example, a summation. Temporal cross-combination consists of combining or merging basic IM images i,j having different j. An example is the cross summation of basic images having the same order number i, but different order numbers j such that the difference between these order numbers is ^ ^The intensity ^^^(^, ^) of each pixel with coordinates (x,y) in the combined image ICi is given by: where ^^^^^, ^, ^ − ^^^ is the intensity of the pixel with spatial coordinates (x,y) of the base image IMi,j-^ ^ Spatial cross-combination involves combining or merging different spatial areas of base IM images i,j An example is the cross summation of a pixel with coordinates (x,y) with areas spaced two at a distance of a number ^ ^ pixels, called the time shift, of a spatial window centered on the pixel (x,y) and defined along a direction, for example the x-axis (x being the pixel's x-coordinate), and having a length equal to a number equal to ∆^ pixels. The intensity ^^^(^, ^) of each pixel with coordinates (x,y) of the combined image ICi is then given by: Where ^^^(^ − ^^ , ^, ^) is the intensity of the pixel with spatial coordinates ^^^(^ − ^^ , ^, ^) of the base image IMi,j. It is also possible to perform a cross summation with an offset along the y-axis (y being the y-coordinate) and / or with an offset along a third axis perpendicular to the x-axis and y-axis in the three-dimensional case. In general, this step can be generalized to any spatial or temporal cross combination on the prepared images. An example of this type of combination is a cross cumulant of order 1 or higher with a zero or non-zero spatial offset in a spatial window of predetermined size in each direction and with a zero or non-zero temporal offset. When all offsets are zero, the cross cumulant is a quadratic sum. At the end of the combination step,N combined images ICi with i = 1 to N were generated. The order number of each combined image ICi is the order number i of the block Bi of images from which the combined image ICi was generated in step 110. Generation of intensity-time curves from the combined images According to one embodiment, the Z intensity-time curves TIC (x,y,t) are then generated in step 114 from the N combined images ICi with i = 1 to N or from a plurality of these combined images. More precisely, each TIC (x,y,t) is generated from the intensities of the pixel with coordinates (x,y) of each of the N combined images ICi with i=1 to N or of a plurality of these combined images. Thus, the TIC (x,y,t) generated for the pixel with coordinates (x,y) represents the evolution of the intensity of this pixel on the different combined images ICi with i=1 to N as a function of time. Alternatively, the process includes a step of determining the Z intensity-time curves TIC (x,y,t) from the N*J elementary images IEi,j with i= 1 to N and j= 1 to J or from a plurality of these elementary images. Thus, the TIC (x,y,t) generated for each pixel with coordinates (x,y) represents the evolution of the intensity of this pixel on the different elementary images IEi,j with i=1 to N as a function of time and j= 1 to J. This is, for example, the case when the process is devoid of steps 111, 112, 113. Alternatively, the process includes a step of determining the Z intensity-time curves TIC (x,y,t) from the N*J improved images IAi,j with i= 1 to N and j= 1 to J or from the refined improved images IAFi,j with i= 1 to N and j= 1 to J or from a plurality of the improved images or a plurality of the refined images. Acquisition The process includes a sequential acquisition step of 10 raw ultrasonic data. More specifically,We acquire BDi blocks of raw ultrasonic data. This step is implemented by an acquisition system SA comprising a probe S with an array RE of transducers TR referenced in Figure 6. It is possible to define sub-apertures of the array RE, each sub-aperture being composed of one or more TR transducers of the array RE. This step is thus implemented by physical interaction between the probe and the region of interest R. Each sub-aperture is, for example, composed of all the TR transducers of the array RE. As seen in Figure 6, we define a transmit sub-aperture SOo and a receive sub-aperture SOo' which define a transmit / receive configuration Ch characterized by a pair (SOo, SOo'). The receive sub-aperture SOo' can be different from or the same as the transmit sub-aperture SOo. The pairs of sub-apertures for different configurations Ch are different. In the non-limiting example of Figure 6,The transmitting sub-aperture SOo is the receiving sub-aperture SOo'. The acquisition of a block of raw ultrasonic data BDi includes, for example, the repetition, for example a large number of times, of an individual acquisition sequence sihk comprising the following steps represented in Figure 6: - Emission Ehk of an ultrasonic beam Wk by the transmitting sub-aperture SOo of the RE array of TR transducers, - Reception Rhk of echoes emitted by the surrounding medium under the effect of the ultrasonic beam Wk by the receiving sub-aperture SOo' of the RE array of TR transducers so as to generate electrical signals, - Computer-implemented preprocessing PTkh, by the acquisition system SA, including the digitization of the electrical signals so as to generate an elementary data set RFijhk, and possibly (but not necessarily) one or more other operations performed on the received echoes.such as a filtering stage and / or a demodulation (carrier suppression) stage and / or a time gain compensation (TGC) stage, and / or resampling of the digitized signals. Among the TR transducers of the RE array, only each of the transmitting sub-aperture transducer(s) emits a signal to generate the ultrasonic beam Wk emitted during the transmitting stage Ehk. The electrical signals processed, including digitized signals, during preprocessing include only electrical signals generated by each of the receiving sub-aperture transducer(s) during the receiving stage Rhk, among the electrical signals generated by the TR transducers of the RE array during this receiving stage Rhk. The received echoes are represented in Figure 6 as a plurality of time-domain signals corresponding to the evolution, over time,The signal received by each of the transducers of the receiving sub-aperture SOo'. The elementary dataset RFijhk is represented in Figure 6 as a table. It comprises a temporal sampling of the signal received by each of the transducers of the receiving sub-aperture. Each cell corresponds to a temporal sample from one of the transducers of the receiving sub-aperture. The cell containing the black circle corresponds to the temporal sample marked by a black circle. According to one embodiment, each individual sequence sihk is performed such that the ultrasonic beam emitted during the reception step insonifies the region of interest R of the individual containing the contrast agents, for example, in the form of microbubbles, and so as to receive, during reception, echoes from the region of interest R. Thus,The raw ultrasound data originates from a region of interest R of an individual containing contrast agents at the time of acquisition. In one embodiment, the region of interest R is a region of the individual containing blood vessels in which endogenous or exogenous contrast agents are present at the time of acquisition. In another embodiment, the contrast agents R move within the region of interest. In the case where the region of interest includes blood vessels, the contrast agents move within the blood vessels due to blood flow. The following text considers a non-limiting example in which the acquisition concerns a region containing contrast agents in the form of microbubbles. As before, the steps described below relating to these microbubbles are valid for other types of contrast agents. As shown in Figure 7,Acquisition step 10 comprises an elementary acquisition sequence seh, in which K individual acquisition sequences sihk, where K is an integer greater than or equal to 1, are implemented using the same transmit / receive configuration Ch so as to acquire K elementary datasets RFijhk. In one embodiment, K is between 1 and 100. The individual acquisition sequences sihk differ from each other in that the ultrasonic beams Wk exhibit distinct spatial and / or temporal (frequency) characteristics. For example, in one embodiment,The K ultrasonic beams can be a family of plane waves emitted at different angles relative to the RE grating. Another embodiment consists of sending K beams corresponding to waves with respective amplitudes differing from each other by predetermined scalar factors. The method includes an acquisition step sbj of a data subblock SBij, which comprises the implementation of H elementary acquisition step(s) seh, with h = 1 to H, where H is an integer greater than or equal to 1, carried out with distinct respective transmit / receive configurations Ch. For example, H equals 1, which corresponds to a single transmit / receive configuration Ch = 1. According to one embodiment,H is between 1 and 100. h is an integer. The elementary acquisition 10i of a data block BDi comprises a sequence of J acquisitions sbj of data subblocks so as to acquire J data subblocks SBij. J is an integer greater than or equal to 1. In other words, the raw data subblocks SBij are acquired sequentially so that all the raw data of subblock SBij is acquired before any raw data of subblock SBij+1. Typically, J is between 1 and 2000. In one embodiment, J can be the same for each data block Bi, but data blocks Bi can, alternatively, have distinct J values. In one embodiment, J is greater than or equal to 2. In another embodiment, J is greater than or equal to 8. A J greater than or equal to 8 ultimately allowsto generate a reliable perfusion index because it takes into account slowly evolving events. This is especially useful when the region of interest is located inside a patient's skull. Indeed, the skull distorts the pressure field and degrades the signal-to-noise ratio of the images. In one embodiment, J is greater than or equal to 10, for example, greater than or equal to 15 or 20. In a particular embodiment, J is less than or equal to 1000 or 2000. In one embodiment, the product of J*N is greater than or equal to 4. In a particular embodiment, the product of J*N is between 50,000 and 150,000. The SBij data subblocks are acquired successively in the temporal order defined by the index j representing the jth emeAcquisition of a subblock, denoted sbj. In other words, the index j is an integer. When the acquisition sbj of a subblock of data SBij has been implemented—that is, when the elementary acquisition sequences seh have been implemented for the different transmit / receive configurations Ch (with h = 1 to H)—we obtain an elementary data subblock SBij composed of H*K elementary data sets RFijhk, as shown in Figure 8. The sequential acquisition 10 of the N raw data blocks BDi comprises a sequence of N elementary acquisitions 10i to acquire N ultrasonic data blocks BDi of order i. The data blocks BDi are acquired successively in the temporal order defined by the index i, representing the i emeImplementation of elementary acquisition step 10i. N is an integer greater than 1. In one embodiment, N is determined such that N*J*H*K is between 1000 and 10,000,000. In a typical embodiment, N = 300, J = 1000, H = 4, and K = 5, i.e., N*J*H*K = 6,000,000. In one embodiment, N is greater than or equal to 2, 10, 15, or 20. In a particular embodiment, N is greater than or equal to 100. In another embodiment, N is less than or equal to 1000 or to 10000. In one embodiment, N is greater than J. Alternatively, N is less than or equal to J. To generate N raw data blocks BDi, the acquisition 10i of a raw data block BDi is repeated N-1 times. This yields a set denoted BD, such that BD = {BD1, …, BDi, …, BDN}, of N raw data blocks BDi. The step 10i with index i corresponds to the implementation, for the i emeThis process involves the acquisition step 10i of a raw data block and allows the acquisition of the i-th order data block BDi. Typically, the acquisition frequency of the raw data blocks BDi is between 5 Hz and 500 Hz. The acquisition frequency, denoted FB, of the blocks Bi of elementary images IEi,j is considered to be the acquisition frequency of the raw data blocks BDi. Typically, the acquisition frequency of the sub-blocks SBij of raw data is between 200 and 20,000 Hz. In one embodiment, this frequency is between 500 and 4,000 Hz. The acquisition frequency of the elementary images IEi,j is considered to be the acquisition frequency of the sub-blocks of raw data SBij. The frequency FB is, for example, equal to the quotient of the acquisition frequency of the sub-blocks by the number of sub-blocks per block. Generation of elementary images E i, jThe generation 20 of elementary images Ei,j comprises, for each subblock SBij with j = 1 to J, the primary combination 20ij of the raw data from the data subblock SBij, to generate an elementary image IEi,j. The elementary image IEi,j is generated solely from the raw data of the data subblock SBij among all the data subblocks with i = 1 to N and j = 1 to J. The implementation of the J primary combinations 20ij for a given i with j = 1 to J allows the generation of the block Bi of images IEi,j. The implementation of the N sets, for i = 1 to i, of J primary combinations 20ij for a given i with j = 1 to J allows the generation of the N blocks Bi of images IEi,j with i = 1 to N and j = 1 to J. The primary combination 20ij allows the transition from time space to distance space. The elementary image is, for example, two-dimensional or three-dimensional.In one embodiment, the primary combination 20ij is implemented to generate elementary images IEi,j (with i = 1 to i and j = 1 to J) in B-Mode. Non-limiting examples of this process are listed below. In one embodiment, the primary combination 20ij of the data from the sub-data block SBij to generate an elementary image IEi,j comprises the combination of data from at least one elementary dataset RFijhk acquired for index i and index j of the sub-data block SBij. In another embodiment, the primary combination 20ij of the data from the sub-data block SBij to generate an elementary image IEi,j comprises the combination of data from several elementary datasets RFijhk acquired for index i and index j of the sub-data block SBij.In one embodiment, the primary combination 20ij of the data from the sub-data block SBij to generate an elementary image IEi,j comprises the data combination from each elementary dataset RFijhk acquired for index i and index j of the sub-data block SBij. In one embodiment, the primary combination 20ij includes a beamforming step. In one embodiment, the beamforming step is performed using the Delay & Sum method. For example, in the case of 2D plane-wave imaging where there is only one sub-aperture defined by transducers aligned along a probe axis, the different beams Wk are plane waves emitted at respective angles ^k with respect to a normal direction.The elementary images IEi,j are generated by combining signals acquired by the transducers delayed by delays ^(^, ^^, ^^) :. Where y is a coordinate of a point along the probe axis in a frame attached to the probe; x is a coordinate of a point along the normal to the probe axis in the frame attached to the probe; and y' is a coordinate of the transducer along the probe axis in the frame attached to the probe. The summation is performed on the y' coordinates of the different transducers and on the respective angles ^k with k = 1 to K. The summation is performed only on the y' coordinates in the case of a single angle ^^. In the case of several transmit / receive configurations, the elementary images IEi,j are, for example, generated by combining signals acquired by the transducers delayed by the delays ^(^, ^^, ^^) in the following way: The summation is performed over the y' coordinates of the different transducers, over the respective angles ^k, and over the different transmit / receive configurations with h = 1 to H. Alternatively, the combination of the data from the subblock SBi,j to generate the elementary image IEi,j can be achieved by other methods based on mathematical models, for example, by Fourier channel formation, using adaptive methods, or multivariate methods. Alternatively, it can be achieved by a learning method using, for example, a neural network trained to reconstruct the elementary images from the subblock data. In one particular embodiment, the elementary acquisition step seh comprises the implementation of several individual acquisition sequences sihk with k = 1 to K and K is an integer greater than 1.These individual acquisition sequences (sihk) differ in that the spatial and / or temporal characteristics of the beams emitted during their respective emission steps (Ehk) differ, as explained previously. The primary combination (20ij) of the data from the subblock (SBi,j) to generate the elementary image (IEi,j) comprises, for example, several channel-forming steps. These respective channel-forming steps use data acquired during the acquisition of the data subblock (SBij) to form several intermediate images, and a combination step, such as averaging, of these intermediate images to form the elementary image. This process is known as "compounding" or "coherent compounding." It allows for increased image contrast and resolution.Combining multiple images reduces noise and thus increases contrast. Combining images acquired using separate beams increases resolution. Alternatively, K is equal to 1. Alternatively and / or additionally, the acquisition step sbj of a subblock SBij includes the implementation of several elementary acquisition sequences seh, where h = 1 to H and H is an integer greater than 1. These elementary acquisition sequences seh differ in the transmit / receive configurations Ch used. This type of process allows addressing probes with a greater number of transducers than the number of available electronic processing channels. In this case, the primary combination 20ij of the data from subblock SBi,j to generate the elementary image IEi,j may include a known step of concatenating the acquired data with the different transmit / receive configurations to reconstruct the entire field of view.This step is implemented prior to the channel formation step(s). The primary combination 20ij of the data from the subblock SBi,j to generate the elementary image IEi,j includes, for example, a step of deconcatenating the data acquired by means of the different elementary sequences. Alternatively, H = 1. System In Figure 9, a block diagram of the SYS system is shown according to an embodiment of the invention configured to implement the method according to the invention. The SYS system includes a processing system, which is a processing device DT in the non-limiting example of Figure 9. Alternatively, the processing system includes various elements capable of communicating by wired or wireless means. In one embodiment, the SYS system includes an acquisition system SA. The acquisition system SA includes a probe S and an acquisition device DA. The probe S includes an array RE of transducers TR.The array of TR transducers can be one-dimensional. In this case, the TR transducers are arranged in a line. The line is, for example, a straight line, in which case the array is linear, or curved. Alternatively, as in the example in Figure 9, the TR transducers are arranged in rows and columns on a flat or curved surface. In one embodiment, the TR transducers are regularly distributed in space. Another variant consists of distributing the TR transducers randomly in space to form a sparse probe. Another variant consists of using probes called RCAs (Raw-Column Arrays), where the elements of the same row and column are connected together. The DA acquisition device includes an EM transmitter, a CTR controller, an optional MUX multiplexer, an MPR preprocessing module including an analog-to-digital converter (ADC), a first MT memory, and a CO1 probe communication system.The DA acquisition system is configured to implement acquisition step 10 of the process according to the invention when the latter includes acquisition step A. The MUX multiplexer allows selective addressing of the transmitting and receiving sub-apertures. When excited by the EM transmitter, via the MUX multiplexer, each TR transducer emits an ultrasonic pulse. The CTR controller is capable of controlling the other elements of the DA acquisition device. The CTR controller is configured to control, at each emission step Ehk: - the EM transmitter so that it generates, for the TR transducers, a defined excitation such that the transmitting sub-aperture SOo, receiving the excitation, emits an ultrasonic beam Wk defined by its temporal and spatial properties, - the MUX multiplexer so that the excitation is transmitted to the transducers of the transmitting sub-aperture SOo.For example, the EM transmitter generates an excitation signal defining a waveform and a predetermined wave frequency and applies different delays to this signal to generate respective elementary excitation signals intended to excite the respective TR transducers of the transmitting sub-aperture SOo, thus defining a direction for the beam Wk intended to be emitted by the sub-aperture SOo. The CTR controller commands the switch configuration of the multiplexer MUX so that the multiplexer transmits the respective elementary excitation signals to the respective TR transducers of the transmitting sub-aperture SOo, so that the array of TR transducers emits the transmitting beam Wk.The CTR controller is configured to control the MUX multiplexer at each Rhk reception stage so that only the TR transducers of the SOo' receiving sub-aperture transmit the electrical signals generated by the Wk ultrasonic beam emission to the MPR preprocessing device. The MPR preprocessing device may further include, for example, at least one filter and / or demodulator and / or time-gain compensator and / or sampler. The CTR controller is programmed to control elements of the DA acquisition device so that the latter implements acquisition stage 10. In one example embodiment, the CTR controller includes, for example, a set of at least one processor operationally coupled to a memory in which a program executed by the CTR controller is stored so that the DA acquisition system implements acquisition stage 10.The multiplexer MUX transmits the signals received by the TR transducers of the receiving sub-aperture SOo' to the pre-processing unit MPR, which includes the analog-to-digital converter (ADC), to generate the raw digital RFijhk data sets from the echo signals. These data sets are stored in the MT memory of the ACQ acquisition unit during a storage step. The DA acquisition unit includes at least one CO1 communication system enabling the DA acquisition unit to communicate with the probe S, and at least one CO2 communication system of the DT processing unit or system, so as to allow data transmission from the MT memory of the ACQ acquisition unit to the DT processing unit, for example, to a second MDT memory of the DT processing unit.The CTR controller is capable of controlling the CO1 communication system of the DA acquisition device so that it transmits data to the CO2 communication system of the DT processing device or system. The processing system, for example, the DT processing device, includes the second MDT memory, a UT processing unit, and an INT user interface comprising an INTE input interface and an INTS output interface. The UT processing unit is configured to implement the T processing step or at least the step of generating at least one perfusion indicator for the process.In one embodiment, the processing unit UT comprises, for example, a set of at least one processor operationally coupled to a memory in which a program is stored. This program is executed by the processing unit UT so that the processing unit UT implements the processing step T, or at least one step of the processing step, for example, the step of generating at least one perfusion indicator of the process. According to one embodiment, the processing unit UT and / or the controller CTR are configured to synchronize the operations performed by the acquisition system SA and the processing system so as to implement the process according to the invention. The data generated during the implementation of the processing is advantageously stored in the second memory MDT or in one or more other memories of the processing system, which may be a processing device DT.This data can then be transferred to internal memory of the DT processing system, for example, the DT processing device. In one embodiment, the DT processing system is configured to implement the display step of at least one generator, and the processing unit UT is configured to generate perfusion indicators and to display representative information for at least one indicator on at least one screen of the INTI output interface. The same applies to the generation of data to be displayed during the optional image display step on a screen of the INTS output interface to allow manual selection of the reference area via the INTE input interface.According to one embodiment, the processing system DT is configured to implement the display step of at least one generator, and the processing unit UT is configured to generate the data to be displayed during this step and display this data on the output interface INTI. The system, including the memories, the processing unit(s), and the communication systems, are configured, in particular sized, to allow implementation of the method according to the invention. This configuration is, for example, implemented experimentally. Hardware: From a hardware perspective, the processing system and the controller CTR can be viewed as computers interacting with computer programs. The processing system DT and the acquisition device DA comprise at least one computer, for example, a microcomputer, a computer network, an electronic component, a tablet, a smartphone, or a personal digital assistant (PDA).The data processing unit (DPU) and the controller (CTR) each comprise, for example, a computer, including at least one processor, and possibly a memory operationally coupled to the computer. The memory may include, for example, a computer-readable medium. The computer-readable medium is a tangible device readable by a reader on the processing unit, capable of storing electronic instructions and being coupled to the CO1, CO2 communication system. 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. 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 and programmable read-only memory (EEPROM), a random access memory (RAM), a magnetic card, or an optical card. Each of the first memory (MT) and second memory (MDT) can be of one of the aforementioned types. 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. Alternatively, the program instructions are obtained from an external source and downloaded via a network. This is particularly the case for applications. The data processing unit (DPU) and the controller (CTR) each include a computer, that is, at least one electronic data processing circuit designed to manipulate and / or transform data represented by electronic or physical quantities in registers of the evaluation system and / or memories into other similar data corresponding to physical data in register memories or other types of display, transmission, or storage devices.The data processing unit (DPU) and / or the controller (CTR) include, for example, memory for storing data, operationally coupled to the data processing circuitry, and a reader adapted to read computer-readable media. The steps of the method according to the invention are, for example, carried out by causing the processing circuits of the DPU and the controller to read predetermined programs stored on hardware such as memory, such that their data processing circuits perform calculations, control communications, and read and / or write data to memory. The processing step is, for example, carried out on a processing device, such as a single computer, or on a system distributed across several computers (particularly via the use of cloud computing).The data processing unit (DPU) and the controller (CTR) each comprise at least one computer including 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)) capable of interpreting instructions in the form of a computer program and / or a hardware element, such as an electronic board, in which steps of the method according to the invention are implemented in hardware elements. In a particular embodiment of the invention, the DPU comprises a graphics processing unit (GPU). Alternatively, the DPU comprises a central processing unit (CPU).The invention relates to a computer program product comprising a computer-readable storage medium containing instructions which, when executed by the processing circuit, cause the SYS system or the DT processing system to implement the steps of the process according to the invention, that is, to execute the functional components of the system according to the invention. The program product may include the computer-readable storage medium. Alternatively, the program instructions are obtained from an external source and downloaded via a network. This is particularly the case for applications. In this case, the computer program product comprises a computer-readable data storage medium on which the program instructions are stored or a data storage signal on which the program instructions are encoded.The form of program instructions is, for example, a source code form, 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 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), or a procedural programming language (e.g., C).The CO1 and CO2 communication systems enable communication between system elements and, optionally, between at least one system element and a device external to the system. These 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 wireless communication link between system elements and / or between a system element and a device external to the system. The communication systems may include any appropriate hardware, firmware, and / or software for communicating information between elements of the device to which the communication system belongs, for example, via a data bus, or to an element external to the device.To enable data communication between different devices that comprise the communication systems, these systems include firmware and / or software hardware that establish a wired or wireless communication link between them, such as Wi-Fi, Bluetooth, cellular, or Ethernet. The INT user interface allows a user to enter data or commands in order to interact with the programs according to the invention. The INT user interface includes, for example, an INTS output interface and an INTE input interface. The INTE input interface includes, for example, a keyboard or a pointing device, such as a mouse, a light pen, a touchpad, a remote control, a speech recognition device, or a haptic device.The input interface INTE allows, for example, a user to enter data or commands in order to interact with the programs according to the invention. For example, the input interface INTE allows a user to select a reference area ZR on an image displayed on a screen of the output interface INTS. The output interface INTS is designed to provide information to a user, either sensorially or electrically, such as visually or audibly. The output interface includes, for example, a display or screen. The display stage 200 can then be a step for providing information from data generated during the processing of order block i by the output interface INTS, by means other than a display. The output interface INTS can be the input device INTE, for example, in the case of a touchscreen tablet.

Claims

CLAIMS 1. A perfusion imaging method in a region of interest (R) of an individual, the method comprising a computer-implemented processing method, the processing method comprising:^ Generating intensity-time curves from pixel / voxel intensities of N blocks (Bi) of J elementary (IEi,j) ultrasound images of the region of interest (R) generated from raw ultrasound data,^ Determining, from the intensity-time curves, a reference intensity-time curve of a predetermined reference area of ​​the region of interest (R),^ Generating at least one perfusion indicator from the reference intensity-time curve, the generation of intensity-time curves comprising: generating N blocks of J refined images from N blocks of J initial images such that the refined images have a narrower dot spread function than the initial images,The N blocks of J initial images being the N blocks of J elementary images or being N blocks of J enhanced images obtained from the N blocks of J elementary images by enhancing the signals from the contrast agent relative to the signals from the surrounding tissues so as to generate N blocks of J enhanced images.

2. Method according to the preceding claim, wherein the processing method comprises: displaying information representative of a perfusion indicator on a screen.

3. Method according to any one of the preceding claims, wherein the J elementary images of each block (Bi) are generated from J sub-blocks of raw ultrasound data acquired sequentially, each elementary image (IEi,j) being generated from a unique sub-block of raw ultrasound data and the N blocks being generated from N blocks of J sub-blocks of data, raw, the N blocks of J sub-blocks of raw data being acquired sequentially and each block of J elementary images being generated from a single block of sub-blocks of raw data.

4. Method according to any one of the preceding claims wherein J is greater than 3 or greater than or equal to 8 or 10.

5. Method according to claim 3, wherein, generating the intensity-time curves comprises, for each of a plurality of the N blocks of J elementary images, generating a combined image (ICi) of the region of interest (R) by combining J representative images of the J elementary images of said block.

6. Method according to any one of the preceding claims, wherein the combination comprises calculating a temporal and / or spatial cross-combination of the J representative images of the J elementary images of the block. 7.A method according to any one of the preceding claims, wherein generating at least one perfusion indicator comprises deconvolving at least one of the intensity-time curves from the reference intensity-time curve.

8. A method according to any one of the preceding claims, wherein generating the intensity-time curves comprises: enhancing the signals from the contrast agent relative to the signals from the surrounding tissues so as to generate N blocks of J enhanced images from the N blocks of J elementary images.

9. A method according to any one of the preceding claims, comprising:^ Acquiring, by an acquisition system comprising a transducer array, the raw ultrasound data,^ Generating the N blocks of J elementary images from the raw ultrasound data.A method according to the preceding claim, wherein the raw data blocks are acquired at a frequency between 5 Hz and 500 Hz and wherein the raw data sub-blocks are acquired at a frequency between 200 and 20000 Hz. A perfusion imaging system configured to implement the method according to any one of the preceding claims, the system comprising a processing system (DT) including a processing unit comprising at least one processor, the processing unit being configured to implement the processing method (T). A system according to the preceding claim, comprising an acquisition system (SA) including a transducer array (RE) (TR), the acquisition system being configured to acquire the raw ultrasound data, the processing system being configured to generate the N elementary image blocks from the raw ultrasound data.Product computer program comprising instructions which lead the system according to any one of claims 11 to 12, to execute the steps of the process according to claim 1. Computer-readable medium on which the computer program according to claim 13 is recorded.

Images