Detection of immobilized contrast agent based on spatio-temporal correlation

Abstract

A solution is provided for imaging a body-part of a patient. A corresponding method (600) comprises providing (610-614;624-628) a contrast sequence of images representative of the body-part comprising a contrast agent and a background sequence 5 of images representative of the body-part substantially without the contrast agent. A combined sequence is generated (640) comprising the images of the contrast sequence and the images of the background sequence. The contrast agent being immobilized on a corresponding target is detected (642-692) according to a spatio-temporal correlation of image values in the combined sequence comprised between a low threshold and a 0 high threshold. A computer program (500) and a corresponding computer program product for implementing the method (600) are also proposed. Moreover, a corresponding computing system (112) and an imaging system (100) comprising it are proposed. A medical method based on the same imaging method (600) is further proposed.

Description

[0001] DETECTION OF IMMOBILIZED CONTRAST AGENT BASED ON SPATIOTEMPORAL CORRELATION

[0002] DESCRIPTION Technical field

[0003] The present disclosure relates to the field of medical imaging. More specifically, this disclosure relates to medical imaging based on contrast agent.

[0004] Background

[0005] The background of the present disclosure is introduced hereinafter with the discussion of techniques relating to its context. However, even when this discussion refers to documents, acts, artifacts and the like, it does not suggest or represent that the discussed techniques are part of the prior art or are common general knowledge in the field relevant to the present disclosure. Particularly, it is expressly understood that possible drawbacks mentioned herein should not be considered to have been previously recognized in the prior art.

[0006] Medical imaging is a well-established technique (in the field of equipment for medical applications) that is used to inspect body-parts of patients by physicians through images providing visual representations thereof (typically, in a substantially non-invasive manner even if the body-parts are not visible directly). Particularly, in ultrasound (medical) imaging, ultrasound waves are applied to each body-part; echo signals recorded in response to the ultrasound waves are then used to create corresponding images of the body-part. Generally, in a technique known as Contrast- Enhanced Ultrasound Imaging (CEUS), an (ultrasound) contrast agent is administered to a corresponding patient. The contrast agent comprises contrast particles that enhance the echo signals, so as to make the portions of the body-part containing them more conspicuous in its images.

[0007] Particularly, a technique known as Ultrasound Molecular Imaging (USMI) is based on the use of a (molecularly) targeted contrast agent that is formulated for attaching to a corresponding (biological) target (for example, a lesion). Once the targeted contrast agent has reached the target in the body -part remaining immobilized thereon, its representation in the corresponding images facilitates an identification of the target (especially when it would be otherwise difficult to detect).

[0008] However, only a small fraction of the targeted contrast agent actually reaches the target and remains immobilized thereon, whereas most of the targeted contrast agent instead continues to circulate for a relatively long time (until it is filtered out by the patient). Therefore, the images of the body-part comprise the representation of both the targeted contrast agent that is immobilized (immobilized contrast agent) and the targeted contrast agent that is still circulating (circulating contrast agent), so that it is not possible to differentiate them. A technique known as Differential Targeted Enhancement (dTE) is then commonly used for detecting the immobilized contrast agent, without waiting until the circulating contrast agent has disappeared. In this case, a destruction flash of ultrasound waves is applied to the body -part so as to destroy the (immobilized and circulating) targeted contrast agent. Images acquired after the application of the destruction flash (when the targeted contrast agent has started again to circulate in the body-part but has not immobilized yet) are subtracted from images acquired before the application of the destruction flash (when the body -part comprises both the immobilized contrast agent and the circulating contrast agent), so as to mainly preserve the contribution of the immobilized contrast only. However, this requires using a relatively low amount of the targeted contrast agent (to have the circulating contrast agent comparable in the images acquired before and after the application of the destruction flash), and in any case it does not allow detecting the immobilized contrast agent with an acceptable degree of accuracy (because of inherent differences between the circulating contrast agent before and after the application of the destruction flash, and because of possible immobilized contrast agent after the application of the destruction flash). Alternatively, WO-A-2007 / 054544 proposes reducing a contribution of the circulating contrast agent in the images of the body-part; for this purpose, the values of each image are filtered into corresponding filtered values, each representative of the lowest response of the corresponding portion of the body -part in multiple images (including the image in question) consisting of a number of images lower than a total number thereof.

[0009] In any case, the images of the body -part have a (spatial) resolution limit due to a wavelength of the echo signals used to acquire them. This does not allow distinguishing points in the images that are closer than the resolution limit; particularly, contrast particles being closer than the resolution limit appear overlapped in the images.

[0010] A technique known as Ultrasound Localization Microscopy (ULM) has been proposed to provide a super-resolution imaging surpassing the resolution limit. This is based on a localization at the super-resolution of (free-moving) contrast particles of a non-targeted contrast agent that are isolated, for example, as defined by their centroids. The localizations of the contrast particles so obtained are tracked, so as to determine their density and / or speed at the super-resolution that define corresponding morphological information (structure) and hemodynamical information (flow) of a vasculature of the body-part, including its micro-vasculature; this provides useful anatomical information that facilitates a characterization of lesions in the body -part.

[0011] Document “Zhao, Unnikrishnan et al., “A Targeted Molecular Localization Imaging Method Applied to Tumor Microvasculature”, Invest Radiol. 2021 April 01; 56(4): 197-206” proposes performing co-localization between the super-resolved micro-vasculature and the immobilized contrast agent, so as to provide simultaneous anatomical and biological information. For this purpose, a high-dose of a non-targeted contrast agent is administered to the patient and then a sequence of images of the bodypart is acquired for determining its super-resolved micro-vasculature. Once the nontargeted contrast agent has disappeared, the following operations are repeated three times: a low-dose of a targeted contrast agent is administered to the patient, a sequence of (pre-burst) images of the body-part is acquired, a destruction burst is applied for destroying the contrast agent and a sequence of (post-burst) images of the body-part is acquired; the sequences of pre-burst images and post-burst images are then used to determine the immobilized contrast agent with the differential targeted enhancement technique.

[0012] However, the above-mentioned imaging procedure requires a relatively long time (uncomfortable for the patient and increasing misalignments among the images). Moreover, it requires two different non-targeted / targeted contrast agents (with the risk of mixing them) and four administrations thereof (hindering clinical applications). In this respect, this document simply hypothesizes the possibility of using a low-dose of the targeted contrast agent for determining both the super-resolved micro-vasculature and the immobilized contrast agent (however, recognizing that this is beyond its scope).

[0013] Summary

[0014] The present invention is defined in the appended claims; particularly, one or more aspects of the present invention are defined in the independent claims and advantageous features thereof are defined in the dependent claims.

[0015] A simplified summary of the present disclosure is herein presented in order to provide a basic understanding thereof; however, the sole purpose of this summary is to introduce some concepts of the disclosure in a simplified form as a prelude to its following more detailed description, and it is not to be interpreted as an identification of its key elements nor as a delineation of its scope.

[0016] In general terms, the present disclosure is based on the idea of detecting the immobilized contrast agent according to spatio-temporal correlation.

[0017] Particularly, an aspect provides a method for imaging a body -part of a patient. The method comprises providing a contrast sequence of images representative of the body-part comprising a contrast agent and a background sequence of images representative of the body-part substantially without the contrast agent. A combined sequence is generated comprising the images of the contrast sequence and the images of the background sequence. The contrast agent being immobilized on a corresponding target is detected according to a spatio-temporal correlation of image values in the combined sequence comprised between a low threshold and a high threshold.

[0018] A further aspect provides a computer program for implementing the method.

[0019] A further aspect provides a corresponding computer program product.

[0020] A further aspect provides a computing system for implementing the method.

[0021] A further aspect provides an imaging system comprising the computing system.

[0022] A further aspect provides a corresponding medical method.

[0023] Brief description of the drawings The solution of the present disclosure, as well as further features and the respective advantages, will be better understood with reference to the following detailed description thereof, to be read in conjunction with the accompanying drawings (wherein, for the sake of simplicity, corresponding elements are denoted with equal or similar references and their explanation is not repeated, and the name of each entity is generally used to denote both its type and its attributes, like value, content and representation).

[0024] In any case, it should be expressly understood that features described in each complete sentence may be implemented independently of the features described in the other sentences, except for those strictly necessary functionally. Whenever an example of implementation is provided for a general feature, for the sake of clarity, it is possible that reference is made subsequently to that example of implementation; in such a case, it should be expressly understood that what is described in relation to the example of implementation also applies to the general feature and to any other implementation thereof. Moreover, any explanation provided in relation to an aspect of the present invention applies mutatis mutandis to any other aspect thereof.

[0025] Particularly:

[0026] FIG.l shows a schematic block diagram of an imaging system wherein the solution according to an embodiment of the present disclosure may be practiced,

[0027] FIG.2-FIG.3 show the general principles of the solution according to an embodiment of the present disclosure,

[0028] FIG.4 shows different imaging procedures for detecting both circulating contrast agent and immobilized contrast agent,

[0029] FIG.5 shows the main software components that may be used to implement the solution according to an embodiment of the present disclosure,

[0030] FIG.6A-FIG-6C show an activity diagram describing the flow of activities relating to the solution according to an embodiment of the present disclosure, and

[0031] FIG.7-FIG.8 show representative examples of experimental results relating to the solution according to an embodiment of the present disclosure.

[0032] Detailed description With reference in particular to FIG.l, a pictorial representation is shown of an imaging system 100 that may be used to practice the solution according to an embodiment of the present disclosure.

[0033] The imaging system 100 is used to image a body -part 103 of a patient 106 for its inspection in an imaging procedure (for example, for discovering a lesion, such as a tumor, in diagnostic applications). Particularly, in the following reference is made to a specific example of the imaging system 100 consisting of an ultrasound scanner.

[0034] The ultrasound scanner 100 comprises an imaging probe 109 (for example, of hand-held type) for acquiring (ultrasound) images of its field of view, comprising the body-part 103. The imaging probe 109 comprises a plurality of ultrasound transducers (for example, in the form of an array) for transmitting pulses of ultrasound waves to the body -part 103 and for receiving echo signals resulting from their reflection, which echo signals are then used to generate the images representative of the body -part 103 (for example, with the ultrasound transducers that are operated alternatively in a pulseecho mode). The imaging probe 103 has a resolution, defined by a maximum number of distinct points that may be imaged per space unit and then a minimum distance at which the points are distinguishable, which is equal to about half a wavelength of the ultrasound waves (for example, 50-500 pm).

[0035] The imaging probe 109 is coupled with a central unit 112 for controlling its operation, for example, via a flexible cable 115. The central unit 112 is provided with one or more display units 118, for example, a monitor (for displaying the images of the imaging procedure that is in progress) and one or more input units, for example, a keyboard 121 with a trackball 124 (for controlling operation of the ultrasound scanner 100). Particularly, the control unit 112 is used to implement the solution according to an embodiment of the present disclosure. In a specific implementation, this may require triggering corresponding actions (as described in detail in the following). For this purpose, the control unit 112 is optionally provided with one or more triggering commands 127, such as in the form of corresponding push buttons.

[0036] For example, the central unit 112 comprises several components that are connected among them through a bus structure 130. Particularly, a microprocessor (pP), or more, 133 provides a logic capability of the central unit 112. A non-volatile memory (ROM) 136 stores basic code for a bootstrap of the central unit 112 and a volatile memory (RAM) 139 is used as a working memory by the microprocessor 133. The central unit 112 has a mass-memory 142 for storing programs and data, for example, a Solid-State-Disk (SSD). Moreover, the central unit 112 comprises a number of controllers 145 for peripherals, or Input / Output (I / O) units, comprising the imaging probe 109, the monitor 118, the keyboard 121, the trackball 124 and the triggering commands 127; moreover, the peripherals may also comprise a network adapter for connecting the ultrasound scanner 100 to a network, a drive for reading / writing removable storage units (such as USB keys) and so on.

[0037] With reference now to FIG.2-FIG.3, the general principles are shown of the solution according to an embodiment of the present disclosure.

[0038] Starting from FIG.2, images of the body-part are provided. Each image comprises a plurality of image elements representative of corresponding locations of the body-part; particularly, each image element has an image value that is a function of the echo signal that has been received from the location of the body -part in response to the ultrasound wave applied thereto. For example, the images are in 2-Dimensions (2-D), with the image elements being pixels and then the image values being pixel values representative of corresponding 2-D locations of the body-part.

[0039] Particularly, a (contrast) sequence 205a of a plurality of (contrast) images 205ai (with i=l ...Na is provided. The contrast images 205ai are representative of the bodypart comprising a contrast agent. The contrast agent comprises contrast particles that act as efficient ultrasound reflector (thereby increasing the echo signals that are returned in response to the ultrasound waves). The contrast agent is substantially free to circulate within the patient, so as to perfuse the body-part. However, the contrast agent is of targeted type, being adapted to immobilizing on a corresponding (biological) target, for example, a lesion. In this case, the contrast agent is formulated so as to attach to the target and then remain in a substantially fixed position along a plurality of contrast images 205a; acquired over a significant period (for example, at least 10-30 s). Generally, the echo signals and then the corresponding pixel values result from the superimposition of different contributions, which are generated by a background (consisting of the body-part, such as its tissue, and possible noise and artifacts), the (circulating) contrast agent that is circulating through the body-part and the (immobilized) targeted contrast agent that is immobilized on the target. Therefore, each contrast image 205ai generally comprises a representation (shown very schematically in the figure) of the background 210ai, of the circulating contrast agent 215a; and of the immobilized contrast agent 220ai. Substantially, the background 210a, and the immobilized contrast agent 220a; remain in the same position along the contrast sequence 205a, whereas the circulating contrast agent 215a; changes its position along the contrast sequence 205a. Therefore, in the contrast sequence 205a the background 210a; and the immobilized contrast agent 220a; have a high spatiotemporal correlation and the circulating contrast agent 215a; has a low spatio-temporal correlation that would allow discriminating them; however, this does not allow discriminating the background 210a; and the immobilized contrast agent 220a; (because of their substantially equal spatio-temporal correlation).

[0040] Moreover, a (background) sequence 205b of a plurality of (background) images 205bj (with j=l ...Nb, for example, Nb=Nd) is provided. The background images 205bi are again representative of the body-part, but now substantially without the contrast agent (for example, being acquired immediately after its destruction). In this case, the echo signals and then the corresponding pixel values result from the contribution of the background only. Therefore, each background image 205bi only comprises a representation (very schematic in the figure) of the background 21 Obi, which substantially remains in the same position of above (as the background 210ai in the contrast sequence 205a).

[0041] In the solution according to an embodiment of the present disclosure, a combined sequence 205c is generated comprising the contrast images 205ai and the background images 205bi (for example, by interweaving them). As a consequence, in the combined sequence 205c the background 210ai,210bi appears in the same position in all the contrast / background images 205a;,205bi, the immobilized contrast agent 220a; appears in the same position only in the contrast images 205a; and the circulating contrast agent 215a; appears in difference positions only in the contrast images 205a; (with the contrast images 205a; equal to half the contrast / background images 205ai,205bi when Na=Nb). Therefore, in the combined sequence 205c the spatiotemporal correlation of the background 210ai,210bi remains the same, whereas the spatio-temporal correlation of the circulating contrast agent 215ai and of the immobilized contrast agent 220ai decreases. As a consequence, in the combined sequence 205c the background 210ai,210bi has a high spatio-temporal correlation (the same as above), the circulating contrast agent 215ai has a low spatio-temporal correlation (lower than above) and the immobilized contrast agent 220ai has an intermediate spatio-temporal correlation (between them), thereby allowing discriminating the immobilized contrast agent 220ai from both the circulating contrast agent 215ai and the background 210ai,210bi.

[0042] For this purpose, an indication of the spatio-temporal correlation is determined of the pixel values in the combined sequence 205c. This measures a degree of dependency of the pixel values on the locations of the body-part (spatial correlation) and on acquisition times of the contrast / b ackground images 205ai,205bi (temporal correlation). The immobilized contrast agent is detected according to the pixel values that have the corresponding spatio-temporal correlation comprised between a low threshold, being (strictly) higher than zero, and a high threshold, being (strictly) higher than the low threshold. The low threshold is selected so as to discriminate the immobilized contrast agent from the circulating contrast agent (whose corresponding pixel values have a spatio-temporal correlation lower than it) and the high threshold is selected so as to discriminate the immobilized contrast agent from the background (whose corresponding pixel values have a spatio-temporal correlation higher than it). A representation of the immobilized contrast agent in the body-part is then output (for example, in colors superimposed on a black-and-white representation of the bodypart).

[0043] The above-mentioned solution facilitates the detection of the immobilized contrast agent. Particularly, this allows discriminating the immobilized contrast agent from both the background and the circulating contrast agent (without waiting until the circulating contrast agent has disappeared).

[0044] The proposed solution provides a good accuracy, even when the amount of the circulating contrast agent is relatively high. This allows spatially delineating the immobilized contrast agent and possibly quantifying its amount, thereby providing accurate information about localization and / or concentration of the target in the bodypart, with beneficial effects on a quality of corresponding imaging applications (for example, facilitating correct diagnosis of several pathologies that would otherwise be difficult to detect).

[0045] At the same time, it is also possible to determine a representation of the circulating contrast agent, and then of a vasculature of the body part; particularly, a super-resolution representation of the circulating contrast agent, and then of a microvasculature of the body part, may be determined (in a manner being conventional per se). This allows providing simultaneous anatomical information (morphological and / or hemodynamical features) and biological information (localization and / or concentration of the target) about the body-part. The above-mentioned result may be achieved with a single (targeted) contrast agent and a single administration thereof (for example, by determining the super-resolved representation of the vasculature from the same contrast sequence or from another sequence of images that is acquired in advance thereto), thereby facilitating its clinical applicability.

[0046] In this case, it is also possible to obtain a similar super-resolution representation of the immobilized contrast agent, so as to provide co-localization between superresolution representation of both the vasculature of the body -part and the target being present therein.

[0047] In any case, the solution according to an embodiment of the present disclosure allows imaging the body-part in a relatively short time (substantially equal only to the time required by the contrast agent to immobilize on the target), thereby improving patient comfort and reducing misalignments among the images.

[0048] Moving to FIG.3, the spatio-temporal correlation of the pixel values in the combined sequence may be determined by applying a Singular Value Decomposition (SVD) technique. In this case, as described in detail in the following, the spatiotemporal correlation of the pixel values in the combined sequence is inversely related to an order of singular values of its decomposition (with the lower the order of the singular values, the higher the spatio-temporal correlation of the pixel values). Particularly, the figure shows two (singular value energy) maps 305a and 305c of the pixel values in an exemplary contrast sequence and in an exemplary combined sequence comprising it, respectively. The maps 305a and 305c show corresponding curves 310a and 310c, respectively, plotting a relative energy, in terms of number of pixel values higher than a minimum threshold (on the ordinate axis) corresponding to the order of the singular values (on the abscissa axis).

[0049] As can be seen, in the map 305a (of the contrast sequence) the curve 310a has two (high) peaks 315a, 320a at low orders of the singular values, corresponding to the background and the immobilized contrast agent having high spatio-temporal correlation in the contrast sequence. The peaks 315a, 320a rise up over a (substantially flat) bottom 325a at higher orders of the singular values, corresponding to the circulating contrast agent having low spatio-temporal correlation in the contrast sequence. In the map 305c (of the combined sequence), the curve 310c again has two (high) peaks 315c, 320c substantially at the same low orders of the singular values of the high peaks 315a, 320a in the curve 305a; the peaks 315c, 320c are lower than the peaks 315a, 320a, since they now correspond only to the background retaining the same high spatio-temporal correlation in the combined sequence. Moreover, the curve 310c again has a (substantially flat) bottom 325c at higher orders of the singular values, corresponding to the circulating contrast agent; the bottom 325c is slightly lower than the bottom 325a in the curve 310a, since the spatio-temporal correlation of the circulating contrast agent is further reduced in the combined sequence. Instead, the curve 310c has a (new) peak 330c at an intermediate order of the singular values; the peak 330c corresponds to the immobilized contrast agent, which has an intermediate spatio-temporal correlation (lower than the one of the background and higher than the one of the circulating contrast agent) in the combined sequence.

[0050] With reference now to FIG.4, different imaging procedures are shown for detecting both circulating contrast agent and immobilized contrast agent.

[0051] Particularly, a time diagram 405 provides a qualitative representation of an imaging procedure being known in the art (corresponding to the one described in the above-mentioned document by Zhao et al.). The imaging procedure starts at a time t(An), when a high-dose of a non-targeted contrast agent is administered to the patient. After a period P(CL) required by the non-targeted contrast agent to fully perfuse the body-part (3-5 minutes), a sequence of images of the body-part is acquired at a time t(Cn) for determining a super-resolution representation of its vasculature. Once the non-targeted contrast agent has disappeared, the following operations are repeated three times. At a time t(Ati) , with i=1..3, a low-dose of a targeted contrast agent is administered to the patient. After a period P(hi) required by the targeted contrast agent to immobilize on its target (6 minutes), at a time t(Ini) a sequence of (pre-burst) images of the body-part is acquired, a destruction burst is applied for destroying the targeted contrast agent and a sequence of (post-burst) images of the body-part is acquired (2 minutes). The post-burst images are subtracted from the pre-burst images to determine the immobilized targeted contrast agent that is immobilized on its target (accumulating the results being obtained for each repetition). Therefore, even considering the shortest value of the period P(CL) (3 minutes) and disregarding the time required for the nontargeted contrast agent to disappear (however not negligible in practice), a duration of the imaging procedure is at least 3+(6+2)-3 = 27 minutes.

[0052] The above-mentioned (known) imaging procedure is compared in the following with two exemplary operation modes of the solution according to corresponding embodiments of the present disclosure.

[0053] Particularly, a time diagram 410 provides a qualitative representation of an imaging procedure according to an embodiment of the present disclosure in a single operation mode. In this case, at a time t(At) a (targeted) contrast agent is administered to the patient (for example, at a relatively high-dose). At the beginning, in a period P(CL-IL) an amount of both the circulating contrast agent and the immobilized contrast agent is too low and in a next period P(CH-IL) the amount of the circulating contrast agent is enough but the amount of the immobilized contrast agent is still too low for their detection. At a time t(CH-In) following the period P(CH-IL), when the amount of both the circulating contrast agent and the immobilized contrast agent is enough for their detection (for example, 3-4 minutes after the administration of the contrast agent), the contrast sequence is acquired, a destruction flash is applied for destructing the contrast agent and then the background sequence is acquired. In this way, a duration of the imaging procedure (3-4 minutes) is far shorter than it is in the prior art. Moreover, the contrast sequence and the background sequence are acquired in short succession (apart for the time required to apply the destruction flash), thereby limiting (down to avoiding at all) any misalignment of their (contrast / b ackground) images. For example, the acquisition of the contrast sequence, the application of the destruction flash and the acquisition of the background sequence may be triggered by actuating a single triggering command, thereby significantly simplifying the imaging procedure.

[0054] A time diagram 415 instead provides a qualitative representation of an imaging procedure according to an embodiment of the present disclosure in a twofold operation mode. In this case, an (advance) sequence of (advance) images of the body-part is acquired at a time t(CH-h) after the period P(CL-IL), when the amount of the circulating contrast agent is enough but the amount of the immobilized contrast agent is too low for their detection (for example, 1.5-2.5 minutes after the administration of the contrast agent at the time t(At) the advance sequence is then used for detecting the circulating contrast agent. After the period P(CH-IL), in a period P(CH-IH) the amount of both the circulating contrast agent and the immobilized contrast agent is enough for their detection. At a time t(CL-In) following the period P(CH-IH), when the amount of the immobilized contrast agent is still enough but the amount of the circulating contrast agent has become too low for their detection (for example, 4.5-4.5 minutes after the administration of the contrast agent), the contrast sequence is acquired, the destruction flash is applied and then the background sequence is acquired as above. In this way, the advance sequence and the contrast sequence are acquired in conditions that are optimized for detecting the circulating contrast agent (low amount of the immobilized contrast agent) and the immobilized contrast agent (low amount of the circulating contrast agent), respectively. This is achieved with a small increase of a duration of the imaging procedure (4.5-5.5 minutes), which however remains far shorter than it is in the prior art. For example, the acquisition of the advance sequence is triggered by actuating a (further) triggering command, and the acquisition of the contrast sequence, the application of the destruction flash and the acquisition of the background sequence are triggered by actuating the same triggering command of above (with a slight increase of complexity of the imaging procedure, which however remains far simpler than it is in the prior art).

[0055] With reference now to FIG.5, the main software components are shown that may be used to implement the solution according to an embodiment of the present disclosure.

[0056] Particularly, all the software components (programs and data) are denoted as a whole with the reference 500. The software components are typically stored in the mass memory and loaded (at least partially) into the working memory of the central unit of the ultrasound scanner when the programs are running, in addition to an operating system and to other application programs not directly relevant to the solution of the present disclosure (thus omitted in the figure for the sake of simplicity). The programs are initially installed into the mass memory, for example, from removable storage units or from a network. In this respect, each program may be a module, segment or portion of code, which comprises one or more executable instructions for implementing the specified logical function.

[0057] A probe driver 503 drives the imaging probe of the ultrasound scanner (for acquiring the images of the body-part and possibly applying the destruction flash thereto during each imaging procedure). In the presence of the triggering commands, a triggering driver 506 drives them. The triggering driver 506 then controls the probe driver 503. The probe driver 503 writes a (possible) display images repository 509, a contrast images repository 512, a background images repository 515 and a (possible) advance images repository 518.

[0058] The display images repository 509 contains (display) images of the body-part, optionally to be displayed in real-time (as soon as they are acquired). For example, the display images have been acquired in a (contrast) specific-mode (and then they are referred to hereinafter as specific display images), so as to substantially reduce the (dominant) contribution of the tissue in the echo signals with respect to the contribution of the contrast agent (such as in an Amplitude Modulation-mode, or AMmode). Optionally, the display images have also been acquired at the same time in a standard-mode (and then they are referred to hereinafter as standard display images), so as to take into account all the contributions in the echo signals (such as in Brightness-mode, or B-mode). The display images repository 509 stores a current version of the display image(s), z.e., the specific display image and the optional standard display image in the example at issue. Each (specific / standard) display image is defined by a bitmap comprising a matrix of cells (for example, with 512-1,024 rows and 512-1,024 columns), each containing a pixel value representing a corresponding location of the body-part (with a size defined by a spatial density of the ultrasound actuators of the imaging probe); the pixel value defines a brightness of its pixel as a function of an intensity of the corresponding echo signal (for example, from 0 to 256).

[0059] The contrast images repository 512 contains the contrast sequence (of contrast images of the body-part) of the imaging procedure that is in progress. For example, the contrast images have been acquired in a contrast-specific mode (and then they are referred to hereinafter as specific contrast images) and optionally in a standard-mode (and then they are referred to hereinafter as standard contrast images) as above. The contrast images repository 509 has an entry for each acquisition instant of the (specific / standard) contrast images (for example, 500-1000, such as 750). The entry stores the corresponding contrast image(s), z.e., the specific contrast image and the possible standard contrast image in the example at issue (in the same format as above). The background images repository 515 contains the background sequence (of background images of the body -part) of the imaging procedure. For example, the background images have been acquired in a contrast-specific mode (and then they are referred to hereinafter as specific background images) and optionally in a standard mode (and then they are referred to hereinafter as standard background images) as above. The background images repository 515 has an entry for each acquisition instant of the (specific / standard) background images (for example, in the same number as in the contrast images repository 512). The entry stores the corresponding background image(s), z.e., the specific background image and the possible standard background image in the example at issue (in the same format as above). The advance images repository 518 contains the optional advance sequence (of advance images of the bodypart) of the imaging procedure. For example, the advance images have been acquired in a contrast-specific imaging mode (and then they are referred to hereinafter as specific advance images) and optionally in a standard imaging mode (and then they are referred to hereinafter as standard advance images) as above. The advance images repository 518 has an entry for each acquisition instant of the (specific / standard) advance images (for example, 500-1000, such as 750). The entry stores the corresponding advance image(s), z.e., the specific advance image and the possible standard advance image in the example at issue (in the same format as above). Optionally, a registration module 521 registers the (contrast / background / advance) images to reduce their misalignment. The registration module 521 reads / writes the contrast images repository 512, the background images repository 515 and the (possible) circulation images repository 518 for updating the corresponding (contrast / background / advance) images accordingly.

[0060] Optionally, a circulation module 524 determines a super-resolution representation of the circulation contrast agent, and then of the vasculature of the bodypart including its micro-vasculature. The circulation module 524 reads the contrast images repository 512 or the advance images repository 518, and it writes a vasculature image repository 527. The vasculature image repository 527 contains a (vasculature) image that provides the super-resolution representation of the vasculature of the bodypart. The vasculature image has a (spatial) resolution, defined by a number of the pixels per space unit, such as Pixel Per Inch (PPI), and then a size of the corresponding locations of the body-part, which is higher than the resolution of the imaging probe (for example, 20-40 pm). The vasculature image is defined by a bitmap comprising a matrix of cells (for example, with 1,024-2,048 rows and 1,024-2,048 columns), each containing a corresponding pixel value that defines a brightness of its pixel as a function of a characteristic of (circulating) contrast particles of the contrast agent that have been localized at the corresponding location of the body -part, for example, their density and / or speed, or a null value (for example, 0) otherwise.

[0061] In the solution according to an embodiment of the present disclosure, a combination module 530 combines the contrast sequence and the background sequence into the combined sequence. The combination module 530 reads the contrast images repository 512 and the background images repository 515, and it writes a combined images repository 533. The combined images repository 533 contains the combined sequence, comprising the (specific) contrast images of the contrast sequence and the (specific) background images of the background sequence (simply referred to as combined images in the following). An immobilization module 536 determines the representation of the immobilized contrast agent in the combined sequence (according to the spatio-temporal correlation of the pixel values of its combined images). The immobilization module 536 reads the combined images repository 533, and it writes an immobilization images repository 539. The immobilization images repository 539 contains an (immobilization) sequence of (immobilization) images corresponding to the contrast images (with the same size), mainly containing the contribution of the immobilized contrast agent only. An analysis module 542 detects the immobilized contrast agent from the immobilization sequence, and then the corresponding target in the body-part. The analysis module 542 reads the immobilization images repository 539, and it writes a target image repository 545 (described in the following).

[0062] Preferably, a super-resolution representation of the immobilized contrast agent, and then of the target, is determined. For example, in this case a localization module 548 localizes any (isolated) contrast particles in the immobilization images with the desired super-resolution. The localization module 548 reads the immobilization images repository 539, and it writes a localization maps repository 551. The localization maps repository 551 contains a (localization) sequence of (localization) maps corresponding to the immobilization images. The localization maps indicate corresponding localizations of the contrast particles in the body -part. Each localization map has a super-resolution that is higher than the resolution of the imaging probe (for example, equal to the resolution of the vasculature image). The localization map is defined by a bitmap comprising a matrix of cells (for example, with the same size of the vasculature image), each containing a corresponding localization flag; the localization flag is asserted (such as at the logic value 1) when a contrast particle has been localized in the corresponding location of the body-part and it is deasserted (such as at the logic value 0) otherwise. In this way, the localization map indicates the localizations of the contrast particles with its super-resolution. A tracking module 554 tracks the (localizations of the) contrast particles along the localization maps to reconstruct their trajectories. The tracking module 554 reads the localization maps repository 551, and it writes a tracking maps repository 557. The tracking maps repository 557 contains a (tracking) sequence of (tracking) maps corresponding to the localization maps. The tracking maps indicate the trajectories of the contrast particles in the body-part (with the same super-resolution). For example, each tracking map is defined by a matrix of cells (with the same size as the localization maps), with each cell containing a cell value. The cell value consists of a tracking value (when the trajectory of a contrast particle has been detected in the corresponding pixel of the corresponding localization map) or of a null value (for example, 0) otherwise; the tracking value contains an indication of the trajectory (for example, a unique identifier thereof) and an interruption counter (indicative of a period without the localization of the corresponding contrast particle, such as in terms of number of localization maps). Optionally, a filtering module 560 filters the tracking maps to remove trajectories incompatible with the immobilized contrast particles. The filtering module 560 reads / writes the tracking maps repository 557. A cumulation module 563 cumulates the trajectories of the (immobilized) contrast particles into their positions. The cumulation module 563 reads the tracking maps repository 557 and it writes the target image repository 545. The target image repository 545 contains a target image that provides the representation of the immobilized contrast agent, and then of the target, in the body-part (with the same super-resolution). The target image is defined by a bitmap comprising a matrix of cells (with the same size as the tracking maps), each containing a corresponding pixel value that defines a brightness of its pixel as a function of a characteristic of the immobilized contrast particles of the contrast agent that have been detected at the corresponding location of the body-part (for example, their presence or amount) or a null value (for example, 0) otherwise.

[0063] A display driver 566 drives the monitor of the ultrasound scanner for displaying the target image and optionally the display images and / or the vasculature image. The display driver 566 reads the (possible) display images repository 509, the (possible) vasculature image repository 527 and the target image repository 545.

[0064] With reference now to FIG.6A-FIG-6C, an activity diagram is shown describing the flow of activities relating to the solution according to an embodiment of the present disclosure.

[0065] Particularly, the diagram represents an exemplary process that may be used to image a body-part of a patient with a method 600. In this respect, each block may correspond to one or more executable instructions for implementing the specified logical function on the central unit of the ultrasound scanner.

[0066] Before a corresponding imaging procedure is started, a healthcare operator (for example, a nurse) administers the contrast agent to the patient. For example, the contrast agent is a suspension in a liquid carrier of contrast particles being stabilized gas-filled bubbles, generally referred to as microvesicles and particularly as microbubbles when dispersed in an aqueous medium and bounded at the gas / liquid interface by a very thin envelope involving a surfactant. A commercial contrast agent comprising microvesicles is SonoVue by Bracco International BV (trademarks thereof). In this way, the contrast particles may be destroyed by applying them ultrasound waves at relatively high energy. The contrast agent is formulated for attaching to its target, for example, by means of a specific interaction therewith. This behavior may be achieved by incorporating a target-specific ligand capable of selectively binding (such as through biochemical affinity and / or electrostatic interaction) to the desired target. Examples of target-specific ligands (which may be inserted into a membrane of the microbubbles) are monoclonal antibodies, peptides, or polysaccharides. The target may be a tissue (including individual cells as well as aggregates of cells, such as membranes or organs, like myocardial, membranous and connective tissue), which tissue may be either in a normal (healthy) or abnormal (pathological) condition (such as tumoral tissue, infarcted heart tissue, blood clots, atherosclerotic plaques or inflammatory tissue), or receptors located on the tissue (such as within the cells or on their surfaces) that are capable of selectively binding to a specific substance (such as VEGFR2 located in tumoral tissue, glycoprotein GPIIbllla located in blood clots or P-Selectin located on inflamed tissue). For example, the contrast agent is administered to the patient intravenously as a bolus (z.e., a single dose provided with a syringe over a short period of time, of the order of 2-20 s); as a consequence, the contrast agent circulates within a circulatory system of the patient, so as to perfuse the body-part.

[0067] The imaging procedure begins by passing from black start circle 602 to block 604 when the physician switches on the ultrasound scanner and places its imaging probe in contact with the patient in the area of the body-part. In response thereto, in the example at issue the display images of the body-part are acquired (and saved temporarily into the corresponding repository) and displayed onto the monitor of the ultrasound scanner in real-time. For example, the following acquisition protocol is used. The images are acquired with a frequency, or Frame Rate (FR), of 60 Hz. At each acquisition instant, the imaging probe transmits ultrasound waves with low acoustic energy (for example, with mechanical index MI=0.01-0.4), so as to involve a negligible destruction of the contrast agent. Particularly, the imaging probe transmits pulses of ultrasound waves at a center frequency of 7.24 MHz (and then wavelength of about 0.2117 mm for a typical propagation speed in the body-part of about 1,540 m / s), with a pulse repetition frequency (PRF) of 15.4 kHz (set to the inverse of a ray round trip time, which is obtained dividing twice an imaging depth that provides sufficient attenuation, such as 50 mm, by the propagation speed - so as to take into account a time required for saving the corresponding echo signals, thereby avoiding that (non-attenuated) echo signals originating from larger depths manifest as echo signals obtained from smaller depths corresponding to later pulses of the ultrasound waves). The pulses are kept active for 67% of each half-cycle period, so as to generate a digital sinusoidal waveform. The ultrasound waves are transmitted with a plane wavefront (allowing a relatively high frame rate, especially useful for performing super-resolution), using a plane wave compounding (PWC) technique (improving resolution of the images). For this purpose, at each acquisition instant a group of pulses of the ultrasound waves are transmitted with different steering angles being centered around zero degrees (with the corresponding echo-signals that are then summed up). The number and the maximum value of the steering angles are set so as to limit a generation of corresponding artefacts. Preferably, the steering angles areA / =3-7being equally spaced up to a maximum value of 1-3°, and more preferably -2°, -1°, 0°, 1° and 2°. Therefore, a delay between a last pulse of each acquisition instant (3-M111) and a first pulse of a next acquisition instant is (l / FR)-(3-M / PRF). In the example at issue, at each acquisition instant a specific image (such as in AM-mode) and possibly a standard image (such as in B-mode) are acquired concurrently. For this purpose, ultrasound waves of opposite polarity are transmitted, so that in the sum of the corresponding echo signals the linear contribution of the tissue removes, or at least substantially reduces, instead leaving the non-linear contribution of the contrast agent. Particularly, in the example at issue for each steering angle three (first, second and third) pulses of the ultrasound waves are transmitted (with equal delays to avoid interference): the second pulse has a certain polarity (for example, positive) and a full amplitude (obtained by using all the ultrasound transducers of the imaging probe), whereas the first / third pulses have an opposite polarity (negative in this case) and a halved amplitude (for example, obtained by using only a half of the ultrasound transducers, such as the odd / even ones). The echo signals received in response to all the pulses are summed up and used to generate the specific display image, whereas the second pulses only are used to generate the standard display image. For example, a loop is performed for each (current) steering angle M=5 in this case). At each iteration of the loop, the first pulse (negative / half) is transmitted and the corresponding echo signal is saved into a first frame of a display receive buffer, the second pulse (positive / full) is transmitted and the corresponding echo signal is saved into the first frame of the display receive buffer, and the third pulse (negative / half) is transmitted and the corresponding echo signal is saved into the first frame of the display receive buffer (3-M echo signals in total). The In-phase / Quadrature (I / Q) components of all the 3-M echo signals are accumulated in a first frame of a display inter buffer, and the I / Q components of the (M) echo signals corresponding to the second pulses only (3z+2) are accumulated in a second frame of the display inter buffer. The first frame and the second frame of the display inter buffer are then converted to video form (for example, into intensity form, log-compressed and so on) and transferred to the display images repository (replacing its content) for their display in real-time.

[0068] The solution according to an embodiment of the present disclosure then involves the execution of different branches for corresponding implementations thereof (for example, set manually during a configuration of the ultrasound scanner, selected dynamically with a possible default value or the only one available). Particularly, in the following reference will be made to the determination of the superresolution representation of both the vasculature and the target in the single operation mode and in the twofold operation mode. Therefore, the flow of activity branches at block 606 accordingly; blocks 608-614 are executed in case of the single operation mode and blocks 616-628 are executed in case of the twofold operation mode, with the flow of activity that then merges again at block 630.

[0069] With reference in particular to block 608 (single operation mode), the process passes to block 610 as soon as the physician actuates the corresponding triggering command (once the display images show that the amount of both the circulating contrast agent and the immobilized contrast agent is enough for their detection). In response thereto, the acquisition and display of the display images is stopped, and the contrast sequence (of contrast images of the body-part) is acquired and saved into the corresponding repository. For example, the contrast images are acquired with the same acquisition protocol of above, but with a higher frame rate (such as 500 Hz). In the example at issue, at each acquisition instant a specific contrast image and a standard contrast image are again acquired concurrently. For example, an outer loop is performed for each (current) acquisition instant (j=0...N-l, N=750 in this case). At each iteration of the outer loop, an inner loop is performed for each (current) steering angle (i=0...M-l, M=5 in this case). At each iteration of the inner loop, the first pulse (negative / half) is transmitted and the corresponding echo signal is saved into a first frame of a contrast receive buffer, the second pulse (positive / full) is transmitted and the corresponding echo signal is saved into the first frame of the contrast receive buffer, and the third pulse (negative / half) is transmitted and the corresponding echo signal is saved into the first frame of the contrast receive buffer (3 MN echo signals in total). Afterwards, the In-phase / Quadrature (I / Q) components of all the (3- AT) echo signals are accumulated in a corresponding (j-th) page of a first frame of a contrast inter buffer, and the I / Q components of the (M) echo signals corresponding to the second pulses only (3z+2) are accumulated in the corresponding (j-th) page of a second frame of the contrast inter buffer. The pages of the first frame and of the second frame of the contrast inter buffer are then converted to intensity form and transferred to the contrast images repository.

[0070] The process then continues automatically to block 612, wherein the destruction flash is applied to the body-part. For example, the destruction flash comprises the repetition of a destruction burst (such as 20-30 times, like 25 times). Each destruction burst involves the transmission of the ultrasound waves with high acoustic energy (for example, with mechanical index MI=l-2), so as to cause the destruction of a significant amount of the contrast agent. Particularly, the imaging probe transmits a series of pulses of ultrasound waves (such as 40 pulses at a frequency of 6 MHz), which are kept active for 67% of each half-cycle period. The destruction flash is applied as soon as possible after the acquisition of the contrast sequence, z.e., in direct succession apart from a time (such as 300-500 ms) required for changing a power profile of the ultrasound scanner (such as its Transmit Power Controller (TPC) profile from 1 to 5 at 50 V).

[0071] The process then continues automatically to block 614, wherein the background sequence (of background images of the body -part) is acquired and saved into the corresponding repository. For example, the background images are acquired with the same acquisition protocol and frame rate of the contrast sequence. In the example at issue, at each acquisition instant a specific background image and a standard background image are likewise acquired. For example, an outer loop is performed for each (current) acquisition instant (j=0...N-l, N=750 in this case). At each iteration of the outer loop, an inner loop is performed for each (current) steering angle (i=0...M-l, M=5 in this case). At each iteration of the inner loop, the first pulse (negative / half) is transmitted and the corresponding echo signal is saved into a first frame of a background receive buffer, the second pulse (positive / full) is transmitted and the corresponding echo signal is saved into the first frame of the background receive buffer, and the third pulse (negative / half) is transmitted and the corresponding echo signal is saved into the first frame of the background receive buffer (3-MN echo signals in total). Afterwards, the In-phase / Quadrature (I / Q) components of all the (3 A ) echo signals are accumulated in a corresponding (j-th) page of a first frame of a background inter buffer, and the I / Q components of the (M) echo signals corresponding to the second pulses only (3z+2) are accumulated in the corresponding (j-th) page of a second frame of the background inter buffer. The pages of the first frame and of the second frame of the background inter buffer are then converted to intensity form and transferred to the background repository. The background sequence of images is acquired as soon as possible after the application of the destruction flash, z.e., in direct succession apart from a time (such as 300-500 ms) required for changing the power profile of the ultrasound scanner (TPC profile from 5 to 1 in the example at issue).

[0072] With reference instead to block 616 (twofold operation mode), the process passes to block 618 as soon as the physician actuates the triggering command of the advance sequence (once the display images show that the amount of the circulating contrast agent is enough for its detection). In response thereto, the acquisition and display of the display images is stopped, and the advance sequence (of advance images of the body-part) is acquired and saved into the corresponding repository. For example, the advance images are acquired with the same acquisition protocol and frame rate of the contrast sequence, with a specific advance image and a standard advance image that are again acquired concurrently at each acquisition instant. Continuing to block 620, the acquisition and display of the display images is resumed. The process then passes from block 622 to block 624 as soon as the physician actuates the triggering command of the contrast / background sequences (once the display images show that the amount of the circulation contrast agent has decreased). In response thereto, the contrast sequence is acquired and saved into the corresponding repository as above. The process continues automatically to block 626 as soon as possible, wherein the destruction flash is applied to the body-part as above. The process continues automatically to block 628 as soon as possible, wherein the background sequence is acquired and saved into the corresponding repository as above.

[0073] With reference now to block 630, optionally all the (contrast / background / advance) images of the contrast sequence, the background sequence and the possible advance sequence are registered (in the corresponding repositories) with respect to a reference image thereof (for example, the first one), so as to reduce their misalignment due to movements and / or deformations of the bodypart during their acquisition. The registration is based on the images without the representation of the contrast agent (or with its contribution substantially reduced), so as to substantially contain the representation of tissue only; in fact, the tissue represents the content of the images that is to be maintained steady (whereas the contrast agent may change when it is circulating). Particularly, in the example at issue the registration is based on the standard images, optionally filtered to further reduce any residual contribution of the circulating contrast agent (for example, by removing its content having a spatio-temporal correlation lower than a threshold value, such as with the same singular value decomposition technique). For this purpose, a transformation required for registering the (standard) images with respect to a reference image thereof is determined. For example, the registration is performed in an iterative mode, wherein each (current) image is registered with respect to an adjacent image (leveraging the fact that the closer the images the less misaligned, especially when their frame rate is far higher than a misalignment rate of the body-part like in this case). Particularly, the current image is registered with respect to the adjacent image that has already being registered, z.e., with the images adjacent to the reference image that are registered with respect thereto, the images adjacent to these images that are registered with respect to them as already registered and so on (so as to limit an error propagation due to the fact that each image is registered only once). The registration of each image involves determining a transformation to be applied thereto for bringing it into spatial correspondence to the (registered) adjacent image (z.e., referring them to a common reference space), such as translation, rotation, scaling, deformation and so on (for example, with a feature-matching technique). The specific images are then registered by applying thereto the transformation of the corresponding standard images (acquired substantially at the same time and then subjected substantially to the same misalignment).

[0074] Optionally, the super-resolution representation of the vasculature is then determined (as soon as the corresponding images are available). In the example at issue, for this purpose the flow of activity again branches at block 632 according to the operation mode of the ultrasound scanner. Particularly, in case of the single operation mode the process continues to block 634, wherein the contrast sequence is taken into account (from the corresponding repository). Instead, in case of the twofold operation mode the process continues to block 636, wherein the advance sequence is taken into account (from the corresponding repository). The flow of activity merges again at block 638 from block 634 or block 636. At this point, the vasculature image of the body-part is generated (and saved into the corresponding repository) from the contrast / advance sequence (with any technique being known per se). In the solution according to an embodiment of the present disclosure, the contrast sequence and the background sequence are combined into the combined sequence (from / to the corresponding repositories) at block 640. Particularly, in the example at issue the combined sequence comprises the specific contrast images and the specific background images. For example, the combined sequence is obtained by interweaving the (specific) contrast images and the (specific) background images (in the same number), so as to alternate them in the resulting combined images. The immobilization sequence (of immobilization images of the body-part) is now generated (and saved into the corresponding repository) according to the spatiotemporal correlation of the pixel values of the combined sequence (from the corresponding repository). For example, this result is obtained by applying the singular value decomposition technique.

[0075] Generally, the singular value decomposition of a (real) matrix / with r rows and c columns (rxc) factorizes it into three matrices as follow:

[0076] I = U-S-VT, wherein U is an orthogonal matrix with r rows and r columns (rxr), 27 is a rectangular diagonal matrix with r rows and c columns (rxc), Kis an orthogonal matrix with c rows and c columns (cxc) and VTis the transpose of the matrix V. The columns of the matrix U (referred to as left-singular vectors) are the eigenvectors of IF, the columns of the matrix V (referred to as right-singular vectors) are the eigenvectors of Fl, and the nonzero elements of the matrix 27 (referred to as singular values) are the square roots of the non-zero eigenvalues of I-F o Fl. The left-singular vectors capture a variance of characteristics of the data along the rows of the matrix / , the right-singular vectors capture a variance of characteristics of the data along the columns of the matrix I and the singular values capture an importance of these characteristics; the singular values are normally arranged in descending order in the matrix 27, so that the lower the order (and then the higher the magnitude) of the singular values the higher the importance of the corresponding left / right-singular vectors in defining the data of the matrix / .

[0077] In the application at issue, at block 642 the combined images are reorganized into a (spatio-temporal) matrix (denoted with the same reference I). For example, the matrix / is a Casorati matrix, having a number of rows equal to the number of pixels of each combined image and a number of column equal to a number of the combined images, which is constructed by inserting the pixel values of each combined image into the corresponding column. The matrix I is factorized with the singular value decomposition technique as above. In this case, the left-singular vectors (also referred to as spatial singular vectors) capture the variance of spatial characteristics of the pixel values and the right-singular vectors (also referred to as temporal singular vectors) capture the variance of temporal characteristics of the pixel values (whose importance is captured by the singular values); therefore, the lower the order (and then the higher the magnitude) of the singular values the higher the spatio-temporal correlation of the pixel values along the combined sequence. An isolation (spatio-temporal) matrix li (again a Casorati matrix in this case) is reconstructed at block 644 from the singular value decomposition of the matrix I by only retaining part of the singular values. Normally, this is used in super-resolution techniques to remove the representation of the background with respect to the circulating contrast agent; for this purpose, only high singular values are retained whose order is higher than a threshold. In the solution according to an embodiment of the present disclosure, instead, the isolation matrix li is reconstructed from the singular value decomposition of the matrix I by only retaining intermediate singular values whose order is comprised between a low threshold Tl (Tl>0) and a high threshold Th (with Th>TT). For example, the low / high threshold Tl, Th are set dynamically, such as by determining the new peak that appears in the energy map of the combined sequence with respect to the energy map of the contrast sequence and then setting them to define an interval around it (like ±1-20%); for example, the low threshold Tl and the high threshold Th are set to orders 20-25 and orders 35-40, respectively. An isolation matrix 27z is generated by resetting to zero all the other singular values, whose order is (possibly strictly) lower than the low threshold Tl or higher than the high threshold Th, and the isolation matrix li is then calculated as: li = UTA-VT

[0078] (with the same result that may also be obtained directly by calculating the isolation The immobilization sequence (of immobilization images of the body-part) is then deconstructed from the isolation matrix li, by taking into account only its columns corresponding to the contrast images. In the example at issue, each immobilization image is obtained by inserting the pixel values of the column of the isolation matrix li of the corresponding contrast image, in inverse order with respect to the one used to construct the matrix I from the combined images (with the same result that may also be obtained by taking into account all the columns of the isolation matrix li and then discarding the obtained images corresponding to the background images).

[0079] In the example at issue, a super-resolution representation of the target is determined from the immobilization sequence. For this purpose, at block 646 the contrast particles being isolated are localized in the immobilization sequence (from the corresponding repository) and the localization sequence (of localization maps) is generated accordingly (and saved into the corresponding repository). In fact, the immobilization images provide a blurred representation of the contrast particles due to the resolution limit of the imaging probe. More specifically, each contrast particle appears as a blurred spot given by its convolution with a Point Spread Function (PSF) of the imaging probe (defining a response thereof to a point source). Therefore, each immobilization image is filtered by removing its content having a correlation to the point spread function being lower than a threshold (such as 0.3-0.5), so as to substantially leave only the contribution of the (isolated) contrast particles. A point representing the localization of each contrast particle is then determined. For example, maxima of the pixel values in the immobilization image are determined and then corresponding (intensity-weighted) centroids are calculated in a window corresponding to the point spread function (such as a square window with a side equal to 120% of a diameter of the point spread function) being centered around them. Each centroid is then defined by its row / column coordinates in the immobilization image (generally in terms of fractional numbers). If the immobilization images have the same size of the localization maps, the row / column coordinates of the centroids are simply rounded; otherwise, if the immobilization images are smaller than the localization maps (according to a corresponding scaling factor higher than 1), the row / column coordinates of the centroids are scaled (multiplied by the scaling factor) and then rounded. In both cases, the corresponding localization map is generated by asserting its localization flags corresponding to the (possibly scaled and rounded) row / column coordinates of the centroids and deasserting all the other localization flags.

[0080] The localization of the contrast particles along the localization sequence are then tracked to reconstruct their possible trajectories. For example, for this purpose a Kalman filter is leveraged. Generally, the Kalman filter is an algorithm that is used to estimate a state of a system in presence of noise. Particularly, a measure of a current state of the system (subjected to observation noise) is applied to the Kalman filter, which outputs an estimate of a future state of the system according to a dynamic model of the system (possibly subjected to process noise) and to a covariance matrix (indicative of an uncertainty of state variables due to the observation / process noise). The covariance matrix is updated iteratively. Particularly, at each iteration a difference between the corresponding state of the system as measured and as estimated (from a previous measure thereof) is used to refine the covariance matrix; the actual state of the system is then determined by applying its previous measure to the Kalman filter with the refined covariance matrix (which result is generally more preci se / accurate than the state of the system that is actually measured).

[0081] In the application at issue, a first localization map along the localization sequence is taken into account (from the corresponding repository) at block 648. The contrast particles in the first localization map potentially start corresponding (new) trajectories. The first tracking map in the tracking sequence is generated (and saved into the corresponding repository) at block 650 accordingly. Particularly, the cell values corresponding to the ones of the contrast particles in the first localization map are set to corresponding tracking values (with all the other cells values set to the null value) and corresponding Kalman filters, based on a dynamic model at constant velocity, are initialized to zero velocity; each tracking value contains a new (trajectory) identifier and the interruption counter that is initialized to 0.

[0082] A loop is entered at block 652, wherein a following tracking map along the tracking sequence is taken into account. An estimated localization of the contrast particle corresponding to each (preceding) trajectory in a preceding tracking map is estimated at block 654; this result is achieved by applying the localization of the contrast particle of the preceding trajectory to the corresponding Kalman filter. The contrast particles of the preceding trajectories are paired at block 656 with the contrast particles in the following localization map according to their estimated localizations. This result may be achieved by applying minimal-distance techniques, such as by using the James Munkres’ variant of the Hungarian assignment algorithm, which minimize a total cost of the pairings being determined by corresponding distances (such as the Euclidean distance) between the estimated localizations of the contrast particles of the preceding trajectories and the localizations of the contrast particles in the following localization map. The localizations of the paired contrast particles in the following localization map are then considered as corresponding measured localizations of the contrast particles of the preceding trajectories.

[0083] A further loop is then entered at block 658, wherein a (current) preceding trajectory is taken into account (starting from a first one in any arbitrary order). A distance is calculated at block 660 between the estimated position and the measured position of the contrast particle of the preceding trajectory (for example, again their Euclidian distance). The flow of activity branches at block 662 according to this distance. If the distance does not reach a (distance) threshold, z.e., it is (possibly strictly) lower than the distance threshold (for example, 3-5 pixels), this means that the preceding trajectory continues in the following localization map; therefore, the preceding trajectory is classified as continuous at block 664. Conversely (meaning that the preceding trajectory does not continue in the following localization map), the flow of activity further branches at block 666 according to the interruption counter of the preceding trajectory (in the corresponding tracking value of the preceding tracking map). If the interruption counter does not reach an (interruption) threshold, z.e., it is (possibly strictly) lower than the interruption threshold (for example, 2-5), this means that the preceding trajectory may still be considered interrupted because of a blinking of its contrast particle (for example, because of noise in the contrast images); in this case, the preceding trajectory is classified as interrupted at block 668. Conversely, if the interruption counter reaches the interruption threshold, this means that the contrast particle of the preceding trajectory has disappeared (for example, because it has resumed circulating after being immobilized). In this case, the preceding trajectory is classified as terminated at block 670; at the same time, the corresponding tracking values in the immediately preceding tracking maps with the interruption counter higher than 0 are reset to null (since it is now ascertained that the preceding trajectory was already terminated in them). The flow of activity merges again at block 672 from block 666, block 668 or block 670. If a last preceding trajectory has not been processed yet, the flow of activity returns to block 658 to repeat the same operations for a further preceding trajectory. Conversely (once all the preceding trajectories have been processed), the loop is exited by descending into block 674.

[0084] At this point, the Kalman filter of each preceding trajectory classified as continuous (continuous trajectory) is refined according to a difference between the estimated position and the measured position of its contrast particle. A new estimated localization of the contrast particle of each continuous trajectory is estimated again at block 676, by applying the localization of the contrast particle of the continuous trajectory in the preceding tracking map to the corresponding (refined) Kalman filter. The generation is started at block 678 of the tracking map corresponding to the following localization map (in its repository), with all the cell values being initialized to null. Particularly, the cell values corresponding to the (newly) estimated localizations of the contrast particles of the continuous trajectories are considered as their actual localizations. Therefore, these cell values are set to corresponding tracking values; each tracking value contains the corresponding trajectory identifier and the interruption counter that is reset to 0 (so as to restart a measure of the interruption of the trajectory if necessary). Continuing to block 680, the cell values corresponding to the (previously) estimated localizations of the contrast particles of the preceding trajectories classified as interrupted (interrupted trajectories) are considered as their possible localizations. Therefore, these cell values are set to corresponding tracking values; each tracking value contains the corresponding trajectory identifier and the interruption counter that is incremented with respect to the one of the corresponding tracking value in the preceding tracking map (z.e., by 1 in the example at issue). Passing to block 682, all the other contrast particles in the following tracking map, z.e., the ones that have not been paired with the preceding trajectories or that have been paired with preceding trajectories classified as terminated (terminated trajectories), potentially start corresponding (new) trajectories (because they may have just immobilized on the target). Therefore, as above the cell values corresponding to these contrast particles are set to corresponding tracking values (with each tracking value containing a new (trajectory) identifier and the interruption counter that is initialized to 0), and corresponding Kalman filters are initialized to zero velocity. A test is performed at block 684 to verify whether a last tracking map has been processed. If not, the flow of activity returns to block 652 to repeat the same operations for a next tracking map. Conversely (once all the tracking maps have been processed), the process descends into block 686. At this point, optionally the tracking maps are filtered to remove trajectories that exhibit characteristics incompatible with the corresponding contrast particles being immobilized. For example, the trajectories being too large are discarded. For this purpose, a footprint of each trajectory is determined as a maximum distance (for example, the Euclidean distance) between all the localizations of its contrast particle in the tracking maps where the trajectory exists (z.e., among its cells containing the tracking values with its identifier). The trajectory is considered too large when its footprint reaches a (footprint) threshold, z.e., it is (possibly strictly) higher than the footprint threshold (for example, 3-5 pixels), meaning that the corresponding contrast particle moves too much for being immobilized. In this case, the trajectory is discarded by resetting to the null value the corresponding cells in the tracking maps. In addition or in alternative, the trajectories being too fast are discarded at block 688. For this purpose, a speed of the contrast particle of each trajectory is determined according to a physical distance between the locations of the body -part corresponding to its localizations in the tracking maps where the trajectory exists (based on their resolution) and a time period between them (based on the frame rate of the contrast images), for example, its average. The trajectory is considered too fast when its speed reaches a (speed) threshold, z.e., it is (possibly strictly) higher than the speed threshold (for example, 1-2 mm / s), meaning that the corresponding contrast particle moves too often to be immobilized. In this case as well, the trajectory is discarded by resetting to the null value the corresponding cells in the tracking maps. In addition or in alternative, the trajectories being too short are discarded at block 690. For this purpose, a length of each trajectory is determined according to a number of the tracking maps where the trajectory exists. The trajectory is considered too short when its length reaches a (length) threshold, z.e., it is (possibly strictly) lower than the length threshold (for example, 30-50), meaning that the corresponding contrast particle has immobilized only temporarily (for example, because the contrast particle has attached in a weaker way to another biological element similar to its target or it has been modified, such as under the action of the immune system of the patient, so as to reduce its capability of attaching to the target or to increase its capability of attaching to other biological elements). In this case as well, the trajectory is discarded by resetting to the null value the corresponding cells in the tracking maps.

[0085] The (possibly filtered) tracking maps are cumulated at block 692 into the target image (initialized with all its pixel values to the null value). For this purpose, an immobilized contrast particle is detected for each trajectory, for example, at the most frequent localization of its contrast particle in the tracking maps. The corresponding pixel value of the target image is set accordingly (for example, by asserting it to indicate a presence of the immobilized contrast agent or incrementing it to indicate an amount of the immobilized contrast agent). A representation of the body -part based on the target image and the possible vasculature image is displayed at block 694 on the monitor of the ultrasound scanner (such as side-by-side). For example, the vasculature / target image is displayed by rendering in color the pixel values corresponding to the immobilized / circulating contrast agent (possibly with corresponding brightness), superimposed to a uniform (such as black) background or to a background image of the body-part. The process then ends at the concentric white / black stop circles 696.

[0086] Experimental results

[0087] With reference now to FIG.7-FIG.8, representative examples are shown of experimental results relating to the solution according to an embodiment of the present disclosure.

[0088] Starting from FIG.7, the following in-vitro experiment was performed. A phantom mimicking tissue based on gelatin #0 by Humimic Medical (trademarks thereof) was used. The phantom was prepared by initially heating about 20 grams of solid gelatin, mixing 0.15 grams of cellulose (to serve as ultrasound scatterers) and stirring a mixture so obtained until all particles were fully dissolved. The mixture was then transferred into a 6.5 cm><4 cm><2 cm cuboidal mold with a top tube and a bottom tube (each with a diameter of 1 mm) that lied in a vertical plane, perpendicularly to a vertical axis. The mixture was heated until all air bubbles were removed and finally allowed to cool down. A plug was installed at an end of the bottom tube for selectively stopping a flow through it. A contrast agent was infused into both the top tube and the bottom tube (with its plug being open). Particularly, a suspension of VEGF / KDR- targeting gas-filled microvesicles was prepared according to the procedure described in WO-A-2016 / 97130. Briefly, DSPC and palmitic acid (molar ratio of 95 / 5) were dissolved in cyclooctane (12.5 mg of lipid mixture per ml of solvent) at 70°C. Separately, DSPE-PEG 2000 (0.040 mg / mL), the VEGF / KDR-targeting lipopeptide of formula II (0.058 mg / mL) and histidine (3.10 mg(mL) were dissolved in a 10% PEG4000 aqueous solution. The cyclooctane preparation and the PEG4000 aqueous solution, in a volume ratio of 1 / 12, were emulsified by using a Megatron MT3000 emulsifier (5 minutes at 12,000 rpm). The resulting emulsion was heated at 80°C for 1 hour under agitation. After cooling at room temperature (for about 1 hour), the emulsion was diluted twice with a 10% PEG4000 aqueous solution and sampled in volumes of 1 ml in DIN8R vials. The vials were subjected to the freeze-drying procedure detailed in WO-A-2016 / 97130, Prep-01, step (vi). A 35 / 65 (by volume) mixture of C4F10 / N2 was then added to the headspace of the vials, which were finally stoppered and sealed. The freeze-dried product was then resuspended in 2 mL of a 5% glucose aqueous solution before use. The plug of the bottom tube was then closed. Therefore, the contrast agent continuously moved through the top tube (mimicking the circulating contrast agent) while the contrast agent started becoming stationary in the bottom tube (mimicking the immobilized contrast agent). A (contrast) sequence of (contrast) images of the phantom was acquired with the circulating / immobilized contrast agent. Thereafter, a destruction flash was applied to the phantom (to destroy the contrast agent in the top / bottom tubes) and then a (background) sequence of (background) images of the phantom was acquired without the contrast agent.

[0089] A (target) image 705 was generated as described above from the contrast sequence and the background sequence. As can be seen, the target image 705 correctly provides the (super-resolution) representation of the immobilized contrast agent 710 in the bottom tube. At the same time, a (vasculature) image 715 was generated as described above from the contrast sequence. As can be seen, the vasculature image 715 correctly provides the (super-resolution) representation of the top tube 720t and of the bottom tube 720b both containing the circulating contrast agent.

[0090] Moving to FIG.8, the following in-vivo experiment was performed. Female Fischer F344 / IcoCrl rats (Janvier Labs, Le Genest-Saint-Isle, France), weighing 180- 200 g (about 8 weeks old) were used. Tumor cells 13762 Mat B III (batch 70003533; ref: CRL- 1666 / American Tissue Culture Collection (ATCC), Manassas, VA) in suspension (5x 105 tumor cells in 100 pl culture medium, passage P5 to P8) were used to induce rat mammary adenocarcinoma via orthotopic injection into the fifth left (cranial inguinal) mammary fat pad of the rats (under anesthesia). Post implantation, a 5-8 mm diameter tumor was grown within a span of 7-8 days. Each rat was prepared for the experiment by first anaesthetizing it and then removing hair around a large area surrounding the tumor by using an animal clipper and a depilatory cream. Subsequently, ultrasound coupling gel was applied on area of the tumor and the imaging probe was positioned precisely to image a large cross section thereof. Finally, the same targeted contrast agent of above was administered via intravenous injection with a Hamilton syringe in the tail vein using a home-made butterfly 25G catheter. Just after the injection of the targeted contrast agent, a flush of approximately 250 pl saline was infused through the rat using a pump at a flow rate of 4 ml / min. A (contrast) sequence of (contrast) images of the area of the tumor was acquired, a destruction flash was applied to the area of the tumor and then a (background) sequence of (background) images was acquired of the area of the tumor.

[0091] A target image 805 was generated as described above from the contrast sequence and the background sequence. As can be seen, the target image 805 correctly provides the (super-resolution) representation of the immobilized contrast agent 810 on the VEGFR2-expressing sites in the neovasculature endothelial cells of the tumor (superimposed on a representation of the corresponding area of the rat). At the same time, a vasculature image 815 was generated as described above from the contrast sequence. As can be seen, the vasculature image 815 correctly provides the (superresolution) representation of the micro-vasculature 820 of the tumor.

[0092] Modifications

[0093] In order to satisfy local and specific requirements, a person skilled in the art may apply many logical and / or physical modifications to the present disclosure, provided that it remains within the scope of the claims. Particularly, the present disclosure may be practiced even without the specific details (such as the numerical values) set forth in the preceding description to provide a more thorough understanding thereof; conversely, well-known features may have been omitted or simplified in order not to obscure the description with unnecessary particulars. Specific features described in connection with any embodiment of the present disclosure may be incorporated in any other embodiment as a matter of general design choice. Moreover, items presented in a same group and different embodiments, examples or alternatives are not to be construed as de facto equivalent to each other (but they are separate and autonomous entities). In any case, each numerical value should be read as modified according to applicable tolerances; particularly, unless otherwise indicated, the terms “substantially”, “about”, “approximately” and the like should be intended as within 10%, preferably 5% and still more preferably 1%. Moreover, each range of numerical values should be intended as expressly specifying any possible number along the continuum within the range (comprising its end points). Ordinal or other qualifiers are merely used as labels to distinguish elements with the same name but do not by themselves connote any priority, precedence or order. The terms include, comprise, have, contain, involve and the like should be intended with an open, non-exhaustive meaning (z.e., not limited to the recited items), the terms based on, dependent on, in agreement with, according to, function of and the like should be intended as a non-exclusive relationship (z.e., with possible further variables involved), the term a / an should be intended as one or more items (unless expressly indicated otherwise), the terms and / or, at least one of, one or more of and the like with respect to a list of two or more entities should be understood comprising each one of such entities individually and any combination of any number of such entities (with the possible addition of other entities), and the term means for (or any similar functional formulation) should be intended as any structure adapted or configured for carrying out the relevant function.

[0094] More specifically, each of the following modifications may be applied (alone or in combination with any other modification) to the corresponding features mentioned above. Particularly, it is expressly understood that each feature mentioned above may be replaced by any of its alternatives or it may be generalized to the corresponding genus as set out in the following.

[0095] For example, an embodiment provides a method for imaging a body -part of a patient. However, the (imaging) method may be of any type (for example, based on ultrasound, magnetic resonance, computed tomography, fluorescence and the like techniques, and so on) and it may be used in any medical application for imaging a body-part of any type and in any condition that belongs to any patient (see below). In any case, although the imaging method may facilitate the task of a physician, it only provides intermediate results that may help him / her but with the medical activity stricto sensu that is always made by the physician himself / herself.

[0096] In an embodiment, the method is implemented under the control of a computing system. However, the computing system may be of any type (see below).

[0097] In an embodiment, the method comprises providing (to the computing system) a contrast sequence and a background sequence. However, the contrast sequence and the background sequence may be provided in any way, either the same or different to each other (for example, in real-time / off-line, acquired, downloaded from a network, read from a storage device and so).

[0098] In an embodiment, the contrast sequence and the background sequence comprise corresponding pluralities of images. However, the images of the contrast sequence and the images of the background sequence may be of any type, either the same or different to each other (for example, with any number, frame rate and so on).

[0099] In an embodiment, each of the images comprises a plurality of image values representative of corresponding locations of the body-part. However, the images may be of any type (for example, with any size, resolution and so on) and comprising any image values (for example, pixel / voxel values, with any chromaticity, bit depth and so on).

[0100] In an embodiment, the images of the contrast sequence are representative of the body-part comprising a contrast agent. However, the contrast agent may be of any type (for example, an exogen contrast agent, such as gas-filled bubbles or phasechange nanodroplets for ultrasound applications, gadolinium complex or paramagnetic lanthanide complex for magnetic resonance applications, iodinated agents or barium sulfate for computed tomography applications, indocyanine green or toluidine blue for fluorescence applications, an endogen contrast agent and so on).

[0101] In an embodiment, the contrast agent is adapted to immobilizing on a biological target. However, this result may be achieved in any way (for example, by using a target-specific contrast agent, a non-target-specific contrast agent, such as a contrast agent that is recognized as a foreign substance by the immune system of the patient and then transported to the liver for its metabolism and elimination, a mixture thereof and so on).

[0102] In an embodiment, the images of the background sequence are representative of the body-part substantially without the contrast agent. However, this result may be achieved in any way (for example, by acquiring the images of the background sequence after destroying the contrast agent, before administering the contrast agent, after the contrast agent has disappeared, such as filtered out by the lungs and / or in the liver of the patient, and so on). In any case, this encompasses either the case wherein the contrast agent is completely absent or the case wherein the contrast agent is present in a negligible amount (for example, not more than 1-5% of its amount corresponding to the contrast sequence).

[0103] In an embodiment, the method comprises generating (by the computing system) a combined sequence comprising the images of the contrast sequence and the images of the background sequence. However, the combined sequence may be generated by combining the contrast sequence and the background sequence in any way (for example, interweaving them with any pitch, concatenating them in any order and so on).

[0104] In an embodiment, the method comprises determining (by the computing system) an indication of a spatio-temporal correlation of the image values in the combined sequence. However, the spatio-temporal correlation may be determined in any way (for example, by applying a singular value decomposition, a Fourier transform (temporal one plus spatial one), a wavelet transform, and so on).

[0105] In an embodiment, the method comprises detecting (by the computing system) the contrast agent being immobilized on the biological target according to the spatiotemporal correlation of the image values in the combined sequence comprised between a low threshold being higher than zero and a high threshold being higher than the low threshold. However, the low / high thresholds may have any value and they may be determined in any way (for example, dynamically, such as based on peaks of the energy maps of the contrast sequence and the combined sequence, on spatial similarity of (left / right) singular vectors, on points of turning and / or points of linear decrease of a curve of the singular values, statically and so on).

[0106] In an embodiment, the method comprises outputting (by the computing system) a representation of the immobilized contrast agent in the body-part. However, the representation of the immobilized contrast agent in the body-part may be provided in any way (for example, with any resolution, by any highlighting of the immobilized contrast agent over the representation of the body -part, such as in color with respect to black-and-white, with higher brightness and so on) and it may be output in any way (for example, displayed on any display unit, such as a monitor, virtual-reality glasses and the like, printed, transmitted remotely and so on).

[0107] Further embodiments provide additional advantageous features, which may however be omitted at all in a basic implementation. In this respect, it is expressly understood that the features of each of the following embodiments may be combined with the above features either alone or in combination with the features of any number of the other following embodiments.

[0108] In an embodiment, the method comprises calculating (by the computing system) a singular value decomposition of the images of the combined sequence. However, the singular value decomposition may be calculated in any way (for example, in the full, thin, compact of truncated version, with the one-sided Jacobi algorithm, two-sided Jacobi algorithm, with a numerical approach, analytically and so on).

[0109] In an embodiment, the method comprises generating (by the computing system) an immobilization sequence of the images corresponding to the contrast sequence from the singular value decomposition being limited to singular values thereof comprised between the low threshold and the high threshold. However, the immobilization sequence may be generated in any way (for example, by reconstructing the isolation matrix after resetting to zero the singular values not comprised between the low threshold and the high threshold or by using only the singular values comprised between the low threshold and the high threshold, by deconstructing (from the isolation matrix) only the images of the combined sequence corresponding to the contrast sequence or all the images of the combined sequence and then discarding the obtained images corresponding to the background sequence, and so on).

[0110] In an embodiment, the method comprises detecting (by the computing system) the immobilized contrast agent from the immobilization sequence. However, the immobilized contrast agent may be detected from the immobilization sequence in any way (for example, by localizing, tracking and accumulating the contrast particles in the immobilization sequence, directly in the immobilization sequence and so on).

[0111] In an embodiment, the method comprises controlling (by the computing system) an imaging system for acquiring the contrast sequence and the background sequence. However, the imaging system may be of any type (see below) and it may be controlled for acquiring the contrast / b ackground sequences in any way (for example, manually, automatically and so on). In any case, this is a computer-implemented method only comprising steps performed by the computing system (which steps may be performed even independently of the acquisition of the corresponding images and then without requiring any interaction with the patient).

[0112] In an embodiment, the imaging system is an ultrasound imaging system. However, the ultrasound imaging system may be of any type (see below) and it may acquire the contrast / background sequences in any way (for example, with ultrasound waves of plane or focused wavefront, with plane wave compounding technique using any number and type of steering angles, with echo signals in I / Q or RF form, in any contrast-specific mode, such as harmonic imaging (HI), pulse inversion (PI), power modulation (PM) or contrast pulse sequencing (CPS) and the like, with or without corresponding images acquired in any standard mode, such as B-mode, TM-mode and the like, with or without any registration of the images, such as in iterative or constant mode, and so on).

[0113] In an embodiment, the method comprises controlling (by the computing system) the imaging system for applying a destruction flash to the body-part for substantially destroying the contrast agent. However, the destruction flash may be of any type (for example, with any energy, any number of destruction bursts each comprising any number of pulses of ultrasound waves, and so on) and applied in any way (for example, in response to the acquisition of the contrast sequence, in response to a corresponding triggering command, after a pre-defined delay from the administration of the contrast agent and so on). In any case, this encompasses either the case wherein the contrast agent is completely destroyed or the case wherein the contrast agent remains in a negligible amount (for example, by destroying at least 95- 99% of its amount).

[0114] In an embodiment, the method comprises controlling (by the computing system) the imaging system for acquiring the background sequence in response to said destroying the contrast agent. However, the background sequence may be acquired in response to the destruction of the contrast agent in any way (for example, immediately, with any delay and so on).

[0115] In an embodiment, the method comprises controlling (by the computing system) the imaging system for acquiring the contrast sequence. However, the contrast sequence may be acquired in any way (for example, in response to a triggering command, after a pre-defined delay from the administration of the contrast agent and so on).

[0116] In an embodiment, the method comprises controlling (by the computing system) the imaging system for applying the destruction flash in response to said acquiring the contrast sequence. However, the destruction flash may be applied in response to the acquisition of the contrast sequence in any way (for example, immediately, with any delay and so on).

[0117] In an embodiment, the method comprises controlling (by the computing system) the imaging system for acquiring the contrast sequence in response to a triggering command being entered manually on the imaging system. However, the triggering command may be of any type (for example, of hard type, such as a push button, a lever and the like, of soft type, such as an icon, a menu item and so on).

[0118] In an embodiment, the method comprises detecting (by the computing system) the contrast agent being circulating in the body-part from the contrast sequence. However, the circulating contrast agent may be detected from the contrast sequence in any way (for example, with any resolution, with any technique for localizing, tracking and accumulating the contrast particles, directly in the contrast sequence and so on).

[0119] In an embodiment, the method comprises outputting (by the computing system) a representation of the circulating contrast agent in the body-part. However, the representation of the circulating contrast agent may be provided in any way (for example, based on density, speed and the like, with or without the application of any processing for reducing discontinuities, such as based on interpolation, sparsitypromoting and the like technique, and so on) and it may be output in any way (for example, either the same or different with respect to the representation of the immobilized contrast agent).

[0120] In an embodiment, the method comprises providing (to the computing system) an advance sequence. However, the advance sequence may be provided in any way (for example, either the same or different with respect to the contrast / b ackground sequence).

[0121] In an embodiment, the advance sequence comprises a corresponding plurality of the images. However, the images of the advance sequence may be of any type (for example, either with the same or different number, frame rate with respect to the images of the contrast / background sequence).

[0122] In an embodiment, the images of the advance sequence are representative of the body-part comprising the contrast agent at a time preceding an acquisition of the images of the contrast sequence. However, the advance sequence may have been acquired in any way (for example, in response to a corresponding triggering command, after a pre-defined delay from the administration of the contrast agent and so on) and with any advance with respect to the acquisition of the contrast sequence.

[0123] In an embodiment, the method comprises detecting (by the computing system) the contrast agent being circulating in the body-part from the advance sequence. However, the circulating contrast agent may be detected from the advance sequence in any way (for example, either the same or different with respect to its detection from the contrast sequence).

[0124] In an embodiment, the method comprises controlling (by the computing system) the imaging system for acquiring the advance sequence in response to a further triggering command being entered manually on the imaging system. However, the further triggering command may be of any type (for example, either the same or different with respect to the triggering command).

[0125] In an embodiment, the method comprises detecting (by the computing system) the circulating contrast agent with a super-resolution higher than a resolution of an imaging probe being used for acquiring the images. However, the super-resolution may have any value (for example, in either absolute or relative terms, and so on).

[0126] In an embodiment, the method comprises detecting (by the computing system) the immobilized contrast agent with a further super-resolution higher than a resolution of an imaging probe being used for acquiring the images. However, the further superresolution may have any value (for example, in either absolute or relative terms, the same or different with respect to the super-resolution of the circulating contrast agent, and so on).

[0127] In an embodiment, the method comprises generating (by the computing system) a localization sequence comprising a plurality of maps corresponding to the images of the contrast sequence. However, the localization sequence may be generated in any way (for example, only for the images of the combined sequence corresponding to the contrast sequence, for all the images of the combined sequence and then discarding the obtained images corresponding to the background sequence, and so on).

[0128] In an embodiment, each of the maps is indicative of corresponding localizations of any contrast particles of the contrast agent (being detected according to the spatiotemporal correlation of the image values in the combined sequence comprised between the low threshold and the high threshold) with the further super-resolution. However, the contrast particles may be detected in any way according to the spatio-temporal correlation (see above), and their localizations may be determined in any way (for example, by detecting distinguishable contrast particles with PSF cross-correlation, deconvolution, entropy and the like techniques and then determining their localizations with peak detection, weighted average, curve fitting and the like techniques, and so on).

[0129] In an embodiment, the method comprises determining (by the computing system) one or more trajectories of corresponding ones of the contrast particles according to the localizations thereof along at least part of the localization sequence. However, the trajectories may be determined in any number and in any way (for example, with or without the use of any Kalman filter to refine the localizations, accepting any number of interruptions of the trajectories down to none, with or without any filtering of the trajectories according to their footprint, speed, length and the like, and so on).

[0130] In an embodiment, the method comprises detecting (by the computing system) the immobilized contrast agent with the further super-resolution from the trajectories. However, the immobilized contrast agent may be detected from the trajectories in any way (for example, based on average, interpolation, sparsity-promoting and the like techniques, and so on).

[0131] In an embodiment, the method comprises, for each following one of the maps following a first one of the maps along the localization sequence, calculating (by the computing system) corresponding estimated localizations in the following map of the contrast particles of any preceding ones of the trajectories being determined for a preceding one of the maps preceding the following map along the localization sequence, the estimated localization of each of the preceding trajectories being calculated by applying the localization in the preceding map of the contrast particle of the preceding trajectory to a corresponding estimation model. However, the estimation model may be of any type (for example, a Kalman filter, a Bayesian network, a Hidden Markov model and so on).

[0132] In an embodiment, the method comprises, for each following map, determining (by the computing system) corresponding measured localizations in the following map of the contrast particles of the preceding trajectories being equal to the localizations in the following map of paired ones of the contrast particles being paired with the corresponding estimated localizations. However, the paired contrast particles may be determined in any way (for example, by applying minimal-distance, Markov chain, cross-correlation, assignment and the like techniques, and so on).

[0133] In an embodiment, the method comprises, for each following map, classifying (by the computing system) each preceding trajectory according to a distance between the corresponding measured localization and estimated localization. However, the distance may be of any type (for example, the Euclidean, Wasserstein, Chebyshev, Manhattan, Minkowski and the like distance, and so on).

[0134] In an embodiment, if the distance does not reach a distance threshold the preceding trajectory is classified as continuous or otherwise the previous trajectory is classified as interrupted if a number of preceding ones of the maps, preceding the following map along the localization sequence, for which the preceding trajectory has been classified as interrupted does not reach an interruption threshold. However, the distance / interruption thresholds may have any value (for example, in either absolute or relative terms).

[0135] In an embodiment, the method comprises, for each following map, setting (by the computing system) the localization in the following map of the contrast particle of any continuous preceding trajectory to a further estimated localization in the following map of the corresponding contrast particle (being calculated by applying the localization in the preceding map of the corresponding contrast particle to the corresponding estimation model being refined according to a difference between the corresponding measured localization and estimated localization), and setting (by the computing system) the localization in the following map of the contrast particle of any interrupted previous trajectory to the corresponding estimated localization. However, the estimation model may be refined in any way (for example, with or without linearization steps for a Kalman filter, with structure / parameter refinement algorithms for a Bayesian network, with expectation-maximization or gradient-based algorithms for a Hidden Markov model, and so on).

[0136] In an embodiment, the estimation model is a Kalman filter. However, the Kalman filter may be of any type (for example, in original, factored, parallel and the like form, and so on).

[0137] In an embodiment, the method comprises starting (by the computing system) corresponding ones of the trajectories at the localizations in the first map of any contrast particles and at the localizations in each of the following maps of any contrast particles being not paired with any continuous preceding trajectories or interrupted preceding trajectories. However, the trajectories may be started in any way (for example, immediately, after the localization of the corresponding contrast particle in two or more maps, and so on).

[0138] In an embodiment, the method comprises terminating (by the computing) each of the preceding trajectories if the corresponding distance reaches the distance threshold and the number of preceding maps for which the preceding trajectory has been classified as interrupted reaches the interruption threshold. However, each trajectory may be terminated in any way (for example, with or without going back according to the interruption threshold, and so on).

[0139] In an embodiment, the method comprises discarding (by the computing system) each of the trajectories having a footprint reaching a footprint threshold. However, the footprint may be defined in any way (for example, maximum, average and the like value, defined in terms of number of cells, corresponding physical extent and so on) and it may be compared with any footprint threshold (for example, in either absolute or relative terms).

[0140] In an embodiment, the method comprises discarding (by the computing system) each of the trajectories having a speed of the corresponding contrast particle reaching a speed threshold. However, the speed may be defined in any way (for example, average, maximum and the like value, defined in terms of physical extent, number of cells and the like divided by number of images, frame rate and the like, and so on) and it may be compared with any speed threshold (for example, in either absolute or relative terms).

[0141] In an embodiment, the method comprises discarding (by the computing system) each of the trajectories having a length not reaching a length threshold. However, the length may be defined in any way (for example, in terms of number of cells, corresponding physical extent and so on) and it may be compared with any length threshold (for example, in either absolute or relative terms).

[0142] Generally, similar considerations apply if the same solution is implemented with an equivalent method, provided that it remains within the scope of the claims. In any case, the method may be performed by using similar steps with the same functions of more steps or portions thereof, removing some non-essential steps or adding further optional steps; moreover, the steps may be performed in a different order, concurrently or in an interleaved way (at least in part).

[0143] An embodiment provides a computer program configured for causing a computing system to perform the method of above when the computer program is executed on the computing system. An embodiment provides a computer program product, the computer program product comprising one or more non-transitory computer readable storage media having program instructions collectively stored on the readable storage media, the program instructions readable by a computing system to cause the computing system to perform the same method. However, the (computer) program may be of any type (for example, implemented as a stand-alone module, a plug-in for a pre-existing application, such as a control program of the imaging system, directly in the latter and so on) and it may be used on any computing system (see below).

[0144] Generally, similar considerations apply if the program is structured in a different way or it has different / additional modules or functions, provided that it remains within the scope of the claims. Likewise, the memory structures may be of other types, or they may be replaced with equivalent entities (not necessarily consisting of physical storage media). The program may take any form suitable to be used by the computing system, thereby configuring it to perform the desired operations; particularly, the program may be in the form of external or resident software, firmware, or microcode (either in object code or in source code), for example, to be compiled or interpreted. Moreover, it is possible to provide the program on any computer readable storage medium. The storage medium is any tangible medium (different from transitory signals per se) that may retain and store instructions for use by the computing system. For example, the storage medium may be of electronic, magnetic, optical, electromagnetic, infrared, or semiconductor type; examples of such storage medium are fixed disks (where the program may be pre-loaded), removable disks, memory keys (for example, USB), and the like. The program may be downloaded to the computing system from the storage medium or via a network (for example, the Internet, a wide area network and / or a local area network comprising transmission cables, optical fibers, wireless connections, network devices); one or more network adapters in the computing system receive the program from the network and forward it for storage into one or more storage devices of the computing system. In any case, the solution according to an embodiment of the present disclosure lends itself to be implemented even with a hardware structure (for example, by electronic circuits integrated on one or more chips of semiconductor material), or with a combination of software and hardware suitably programmed or otherwise configured.

[0145] An embodiment provides a computing system, which comprises means configured for performing the steps of the method of above. An embodiment provides a computing system comprising corresponding circuitries (z.e., any hardware suitably configured, for example, by software) for performing the steps of the same method. However, the computing system may be of any type (for example, a central unit of an imaging system, a separate computer and so on).

[0146] An embodiment provides an imaging system. However, the imaging system may be of any type (for example, of ultrasound, magnetic resonance, computed tomography, fluorescence and the like type, in the form of a scanner, an endoscope, a laparoscope and so on).

[0147] In an embodiment, the imaging system comprises the computing system of above. However, the computing system may be comprised in the imaging system in any way (for example, integrated therein, connected thereto in any way, and so on).

[0148] In an embodiment, the imaging system comprises an imaging probe for acquiring the images. However, the imaging probe may be of any type (for example, hand-held, fixed, of matrix or liner type, and so on).

[0149] In an embodiment, the computing system is coupled with the imaging probe for receiving the images of the contrast sequence and the images of the background sequence therefrom. However, the computing system and the imaging probe may be coupled in any way (for example, locally / remotely via any type of wired and / or wireless connection, and so on). Generally, similar considerations apply if the computing system and the imaging system each has a different structure or comprises equivalent components or has other operative characteristics, provided that it remains within the scope of the claims. In any case, every component thereof may be separated into more elements, or two or more components may be combined together into a single element; moreover, each component may be replicated to support the execution of the corresponding operations in parallel. Moreover, unless specified otherwise, any interaction between different components generally does not need to be continuous, and it may be either direct or indirect through one or more intermediaries.

[0150] An embodiment provides a medical method for imaging a body-part of a patient. However, the medical method may be used to image any body-part (for example, of any type, such as organs, regions thereof, tissues, bones, joints and the like, in any condition, such as healthy, pathological with any lesions and the like, and so on) of any patient (for example, human beings, animals and so on).

[0151] In an embodiment, the medical method comprises administering a contrast agent adapted to immobilizing on a biological target to the patient. However, the contrast agent may be of any type (see above) and it may be administered in any way (for example, as a bolus by a syringe, as a continuous infusion by a pump and so on) and at any time (for example, in advance, immediately before performing the medical method, continuously during it and so on); the contrast agent may also be administered to the patient in a non-invasive manner (for example, orally for imaging the gastrointestinal tract or via a nebulizer into the airways) or without any substantial physical intervention on the patient that would require professional medical expertise or entail any health risk (for example, intramuscularly).

[0152] In an embodiment, the medical method comprises acquiring the contrast sequence and the background sequence. However, the contrast / b ackground sequences may be acquired in any way (see above).

[0153] In an embodiment, the medical method comprises outputting the representation of the immobilized contrast agent according to the (imaging) method of above. However, the representation of the immobilized contrast agent may be outputted in any way (see above). In an embodiment, the medical method comprises performing a medical procedure relating to the body -part according to the representation of the immobilized contrast agent. However, the medical procedure may be of any type (for example, a diagnostic procedure, a therapeutic procedure, a surgical procedure and so on).

[0154] In an embodiment, the medical method is a diagnostic method comprising evaluating a health condition of the body-part according to the representation of the immobilized contrast agent. However, the proposed solution may find application in any kind of diagnostic method in the broadest meaning of the term (for example, aimed at discovering new lesions, monitoring known lesions and so on).

[0155] In an embodiment, the medical method is a therapeutic method comprising treating the body-part according to the representation of the immobilized contrast agent. However, the proposed solution may find application in any kind of therapeutic method in the broadest meaning of the term (for example, aimed at curing a pathological condition, avoiding its progress, preventing the occurrence of a pathological condition, ameliorating a comfort of the patient and so on).

[0156] In an embodiment, the medical method is a surgical method comprising operating the body-part according to the representation of the immobilized contrast agent. However, the proposed solution may find application in any kind of surgical method in the broadest meaning of the term (for example, for curative purposes, prevention purposes, aesthetic purposes and so on).

Claims

CLAIMS1. A method (600) for imaging a body -part of a patient, wherein the method (600) comprises, under the control of a computing system (112): providing (610-614;624-628), to the computing system (112), a contrast sequence and a background sequence comprising corresponding pluralities of images, each of the images comprising a plurality of image values representative of corresponding locations of the body -part, wherein the images of the contrast sequence are representative of the body-part comprising a contrast agent adapted to immobilizing on a biological target and wherein the images of the background sequence are representative of the body-part substantially without the contrast agent, generating (640), by the computing system (112), a combined sequence comprising the images of the contrast sequence and the images of the background sequence, determining (642), by the computing system (112), an indication of a spatiotemporal correlation of the image values in the combined sequence, detecting (644-692), by the computing system (112), the contrast agent being immobilized on the biological target according to the spatio-temporal correlation of the image values in the combined sequence comprised between a low threshold being higher than zero and a high threshold being higher than the low threshold, and outputting (694), by the computing system (112), a representation of the immobilized contrast agent in the body -part.

2. The method (600) according to claim 1, wherein the method (600) comprises: calculating (642), by the computing system (112), a singular value decomposition of the images of the combined sequence, generating (644), by the computing system (112), an immobilization sequence of the images corresponding to the contrast sequence from the singular value decomposition being limited to singular values thereof comprised between the low threshold and the high threshold, and detecting (646-692), by the computing system (112), the immobilized contrastagent from the immobilization sequence.

3. The method (600) according to claim 1 or 2, wherein the method (600) comprises: controlling (608-614;622-628), by the computing system (112), an imaging system (100) for acquiring the contrast sequence and the background sequence.

4. The method (600) according to claim 3, wherein the imaging system (100) is an ultrasound imaging system (100).

5. The method (600) according to claim 3 or 4, wherein the method (600) comprises: controlling (612;626), by the computing system (112), the imaging system (100) for applying a destruction flash to the body-part for substantially destroying the contrast agent, and controlling (614;628), by the computing system (112), the imaging system (100) for acquiring the background sequence in response to said destroying the contrast agent.

6. The method (600) according to claim 5, wherein the method (600) comprises: controlling (610;624), by the computing system (112), the imaging system (100) for acquiring the contrast sequence, and controlling (612;626), by the computing system (112), the imaging system (100) for applying the destruction flash in response to said acquiring the contrast sequence.

7. The method (600) according to claim 6, wherein the method (600) comprises: controlling (608-610;622-624), by the computing system (112), the imaging system (100) for acquiring the contrast sequence in response to a triggering command being entered manually on the imaging system (100).

8. The method (600) according to any claim from 1 to 7, wherein the method (600) comprises: detecting (634,638), by the computing system (112), the contrast agent beingcirculating in the body -part from the contrast sequence, and outputting (694), by the computing system (112), a representation of the circulating contrast agent in the body -part.

9. The method (600) according to any claim from 1 to 7, wherein the method (600) comprises: providing (618), to the computing system (112), an advance sequence comprising a corresponding plurality of the images, the images of the advance sequence being representative of the body-part comprising the contrast agent at a time preceding an acquisition of the images of the contrast sequence, detecting (636-638), by the computing system (112), the contrast agent being circulating in the body -part from the advance sequence, and outputting (694), by the computing system (112), a representation of the circulating contrast agent in the body -part.

10. The method (600) according to claim 9 when dependent directly or indirectly on claim 3, wherein the method (600) comprises: controlling (616-618), by the computing system (112), the imaging system (100) for acquiring the advance sequence in response to a further triggering command being entered manually on the imaging system (100).

11. The method (600) according to any claim from 8 to 10, wherein the method (600) comprises: detecting (632-638), by the computing system (112), the circulating contrast agent with a super-resolution higher than a resolution of an imaging probe (109) being used for acquiring the images.

12. The method (600) according to any claim from 1 to 11, wherein the method (600) comprises: detecting (646-692), by the computing system (112), the immobilized contrast agent with a further super-resolution higher than a resolution of an imaging probe (109) being used for acquiring the images.

13. The method (600) according to any claim from 1 to 12, wherein the method(600) comprises:generating (646), by the computing system (112), a localization sequence comprising a plurality of maps corresponding to the images of the contrast sequence, each of the maps being indicative of corresponding localizations of any contrast particles of the contrast agent, being detected according to the spatio-temporal correlation of the image values in the combined sequence comprised between the low threshold and the high threshold, with the further super-resolution, determining (648-690), by the computing system (112), one or more trajectories of corresponding ones of the contrast particles according to the localizations thereof along at least part of the localization sequence, and detecting (692), by the computing system (112), the immobilized contrast agent with the further super-resolution from the trajectories.

14. The method (600) according to claim 13, wherein the method (600) comprises, for each following one of the maps following a first one of the maps along the localization sequence: calculating (654), by the computing system (112), corresponding estimated localizations in the following map of the contrast particles of any preceding ones of the trajectories being determined for a preceding one of the maps preceding the following map along the localization sequence, the estimated localization of each of the preceding trajectories being calculated by applying the localization in the preceding map of the contrast particle of the preceding trajectory to a corresponding estimation model, determining (656), by the computing system (112), corresponding measured localizations in the following map of the contrast particles of the preceding trajectories being equal to the localizations in the following map of paired ones of the contrast particles being paired with the corresponding estimated localizations, classifying (660-668), by the computing system (112), each preceding trajectory according to a distance between the corresponding measured localization and estimated localization, if the distance does not reach a distance threshold the preceding trajectory being classified as continuous or otherwise the preceding trajectory being classified as interrupted if a number of preceding ones of the maps,preceding the following map along the localization sequence, for which the preceding trajectory has been classified as interrupted does not reach an interruption threshold, setting (674-680), by the computing system (112): the localization in the following map of the contrast particle of any continuous preceding trajectory to a further estimated localization in the following map of the corresponding contrast particle being calculated by applying the localization in the preceding map of the corresponding contrast particle to the corresponding estimation model being refined according to a difference between the corresponding measured localization and estimated localization, and the localization in the following map of the contrast particle of any interrupted previous trajectory to the corresponding estimated localization.

15. The method (600) according to claim 14, wherein the estimation model is a Kalman filter.

16. The method (600) according to claim 14 or 15, wherein the method (600) comprises: starting (650,682), by the computing system (112), corresponding ones of the trajectories at the localizations in the first map of any contrast particles and at the localizations in each of the following maps of any contrast particles being not paired with any continuous preceding trajectories or interrupted preceding trajectories.

17. The method (600) according to any claim from 14 to 16, wherein the method (600) comprises: terminating (670), by the computing system (112), each of the preceding trajectories if the corresponding distance reaches the distance threshold and the number of preceding maps for which the preceding trajectory has been classified as interrupted reaches the interruption threshold.

18. The method (600) according to any claim from 13 to 17, wherein the method (600) comprises: discarding (686), by the computing system (112), each of the trajectories having a footprint reaching a footprint threshold,discarding (688), by the computing system (112), each of the trajectories having a speed of the corresponding contrast particle reaching a speed threshold, and / or discarding (690), by the computing system (112), each of the trajectories having a length not reaching a length threshold.

19. A computer program (500) configured for causing a computing system (112) to perform the method (600) according to any claim from 1 to 18 when the computer program (500) is executed on the computing system (112).

20. A computer program product, the computer program product comprising one or more non-transitory computer readable storage media having program instructions collectively stored on the readable storage media, the program instructions readable by a computing system to cause the computing system to perform the method according to any claim from 1 to 18.

21. A computing system (112) comprising means (133) configured for performing the steps of the method (600) according to any claim from 1 to 18.

22. A computing system comprising corresponding circuitries for performing the steps of the method according to any claim from 1 to 18.

23. An imaging system (100) comprising the computing system (112) according to claim 21 or 22, an imaging probe (109) for acquiring the images, the computing system (112) being coupled with the imaging probe (109) for receiving the images of the contrast sequence and the images of the background sequence therefrom.

24. A medical method for imaging a body -part of a patient, wherein the medical method comprises: administering a contrast agent adapted to immobilizing on a biological target to the patient, acquiring a contrast sequence and a background sequence comprising corresponding pluralities of images, each of the images comprising a plurality of image values representative of corresponding locations of the body -part, wherein the images of the contrast sequence are representative of the body-part comprising the contrast agent and wherein the images of the background sequence are representative of thebody-part substantially without the contrast agent, outputting the representation of the immobilized contrast agent according to the method of any claim form 1 to 18, performing a medical procedure relating to the body-part according to the representation of the immobilized contrast agent.

25. The medical method according to claim 24, wherein the medical method is a diagnostic method comprising: evaluating a health condition of the body-part according to the representation of the immobilized contrast agent.

26. The medical method according to claim 24, wherein the medical method is a therapeutic method comprising: treating the body-part according to the representation of the immobilized contrast agent.

27. The medical method according to claim 24, wherein the medical method is a surgical method comprising: operating the body-part according to the representation of the immobilized contrast agent.

Images