Method for imaging coronary blood flow in a biological heart

Abstract

The present invention relates to the field of ultrasound and imaging of coronary blood flow in the heart. Patients with coronary microvascular dysfunction (CMD) have a poor prognosis and a significantly higher incidence of cardiovascular events, including heart failure, sudden cardiac death, and myocardial infarction (MI) hospitalization. Despite the urgent clinical need, there are no non-ionizing and non-invasive techniques available clinically for directly visualizing the coronary microvasculature and assessing the local coronary microvasculature. Blood flow imaging remains a difficult task in the heart due to the rapid motion of this organ. To overcome the limitations of practical imaging methods for coronary blood flow, the inventors have proposed an ultrasound ultrafast imaging method that automatically detects time periods of low myocardial velocity and estimates coronary blood flow velocity and tissue velocity from the same data acquisition.

Description

Technical Field Background Art

[0001] The coronary circulation is responsible for the perfusion of the heart, as observed in cases of stable angina or myocardial infarction, and alterations in coronary blood flow have serious consequences for the performance of the heart. The coronary vasculature consists of three compartments. The first consists of the epicardial coronary arteries, which extend along the surface of the heart and exhibit diameters ranging from a few millimeters to 500 μm. The second includes the anterior arteries, which penetrate the myocardium from the epicardium to the endocardium and exhibit diameters ranging from 500 μm to 100 μm. The third corresponds to the coronary microvasculature, which displays vessel diameters below 100 μm [1].

[0002] To date, the epicardial coronary vasculature is the only compartment that can be imaged in humans using current angiographic techniques [1, 2] (such as X-ray [3]), computed tomography angiography (CTA) [4] or cardiac magnetic resonance (CMR) imaging [5]. Therefore, cardiology practice has focused on focal macroscopic coronary artery disease. For example, invasive coronary angiography (ICA) with catheterization [3] is still the reference technique for studying coronary artery lesions in the setting of suspected ischemia. ICA provides both an anatomical analysis of the main epicardial stenosis and a global functional assessment based on the fractional flow reserve (FFR). The assessment of FFR is indeed the main tool for clinical decision-making in ischemic heart disease [2, 6] and subsequent pharmacological or invasive treatment by percutaneous coronary intervention or surgery [7].

[0003] In many patients, microvascular disease is an early manifestation of coronary artery disease (CAD), and coronary microvascular dysfunction (ie, including the anterior artery) is now recognized as an important marker of myocardial ischemia [1, 6], however, its assessment in clinical practice remains challenging.

[0004] In fact, clinical guidelines for the treatment of stable ischemic heart disease only consider coronary microvascular dysfunction after exclusion of signs of epicardial disease[8]. A large number of patients with signs of angina and ischemia on stress testing have normal coronary angiography[9]. Current evidence suggests that the majority of these patients have coronary microvascular dysfunction (CMD), also known as microvascular angina[9]. Patients with CMD have a poor prognosis, with a significantly higher incidence of cardiovascular events, including heart failure, sudden cardiac death, and hospitalization for myocardial infarction (MI).

[0005] Despite the urgent clinical need, no technology exists to directly visualize and assess the regional coronary microvasculature. To date, only global indirect measurements performed by functional tests (PET, CMR, and contrast echocardiography) provide hemodynamic information in response to the vasodilator adenosine, such as myocardial blood flow (MBF) and coronary flow reserve (CFR) [1].

[0006] However, despite improvements in radiation dose management, cumulative radiation exposure from ionizing modes remains associated with cancer risk.

[10] This risk is particularly important in pediatric patients, such as children with congenital or acquired heart disease, who may be exposed to relatively high lifetime cumulative doses of ionizing radiation from necessary medical imaging procedures,

[11] including radiography, fluoroscopic procedures (including diagnostic and interventional cardiac catheterization), electrophysiological studies, cardiac computed tomography (CT) studies, and nuclear cardiology examinations.

[0007] Blood flow imaging remains a difficult task for rapidly moving organs such as the heart. Conventional ultrasound Doppler imaging has long been limited in sensitivity for imaging small vessels at low flow velocities (<1 cm / s), and the overlap of tissue and blood motion in this velocity range makes separation of tissue and blood signals challenging. In recent years, ultrafast Doppler imaging has greatly increased the sensitivity of blood flow imaging. This technique has been shown to be able to detect small blood flow changes in the brain due to neurovascular coupling, and has therefore been used for functional brain imaging in anesthetized and awake small animals for neuroscience research

[12] . Sensitivity has been further increased by developing new clutter filters suitable for ultrafast imaging (e.g., spatiotemporal singular value decomposition)

[13] . However, in cardiac applications, ultrasound Doppler imaging of coronary blood flow remains limited due to the rapid motion of the heart.

[0008] It has been demonstrated that ultrafast Doppler imaging can limit some of the effects of this motion and allow for increased sensitivity of Doppler imaging

[14] , but Doppler imaging during the rapid motion phases of the heart is still not possible.

[0009] To overcome the limitations of practical imaging methods for coronary blood flow, the present inventors employed a recently proposed 4-D (3-D+time) ultrasound ultrafast imaging method

[15] to automatically detect time periods of low myocardial velocity and estimate flow and tissue velocity based on the same data acquisition.

[0010] Thus, a non-invasive, non-ionizing technique for imaging coronary blood flow at both macro- and microscales at the patient's bedside is provided. SUMMARY OF THE INVENTION

[0012] The scope of the invention is defined by the claims. Any subject matter falling outside the scope of the claims is provided for informational purposes only.

[0013] A non-invasive and non-ionizing imaging method is disclosed herein to enhance direct imaging of coronary blood flow and the anatomy and function of coronary vessels at both macro- and microscales, from epicardial to endocardial regions.

[0014] Thus, imaging methods, imaging apparatus, and computer-readable media are provided for non-ionizing, non-invasive anatomical and functional imaging of coronary vessels at both macroscopic and microscopic scales.

[0015] List of abbreviations

[0016] CAD = Coronary Artery Disease

[0017] CFR = coronary flow reserve

[0018] CMD = Coronary Microvascular Dysfunction

[0019] CMR = cardiac magnetic resonance

[0020] CT = computed tomography

[0021] CTA = computed tomography angiography

[0022] DSP = Digital Signal Processor

[0023] ECG = electrocardiogram

[0024] FFR = Fractional Flow Reserve

[0025] ICA = invasive coronary angiography

[0026] MBF = myocardial blood flow

[0027] MI = myocardial infarction

[0028] PET = Positron Emission Tomography

[0029] SVD = Singular Value Decomposition

[0030] TD = Transmission Delay

[0031] BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Other features and advantages of the present disclosure emerge from the following detailed description of a non-limiting embodiment thereof, with reference to the accompanying drawings.

[0033] In the attached figure:

[0034] Figure 1 is a schematic diagram showing an apparatus for 4D imaging of the heart;

[0035] Figure 2 It shows Figure 1 Block diagram of some equipment;

[0036] Figure 3 Is explained by Figure 1-2 A diagram of a virtual source of diverging ultrasonic waves generated by a device;

[0037] Figure 4 Shown by Figure 1-2 The device emits ultrasound waves scattered in the living heart;

[0038] Figure 5 The emission of two consecutive diverging ultrasonic waves with different propagation directions from two virtual sources is shown;

[0039] Figure 6 The myocardial wall motion during the cardiac cycle is shown. Blood flow can be reconstructed in two time windows with limited tissue velocity;

[0040] Figure 7 Examples of coronary flow velocity imaging at baseline and during reactive hyperemia are shown;

[0041] Figure 8 Microbubble imaging and localization (mean intensity projection) are shown;

[0042] Figure 9 : Microbubble localization and tracking are shown, where (A) the coronary artery network of a perfused isolated heart is imaged by ultrasound and (B) coronary flow velocity is mapped in the isolated perfused heart. Detailed Description of the Invention

[0044] In the drawings, the same reference numerals refer to the same or similar elements.

[0045] Figure 1 and Figure 2 The device shown is suitable for ultrafast 4D ultrasound imaging of the heart of a living being 1 , such as a mammal, in particular a human.

[0046] The apparatus may include, for example, at least a 2D array ultrasound probe 2 and a control system.

[0047] The 2D array ultrasound probe 2 may have, for example, several hundred to several thousand transducer elements T ij , with a spacing of less than 1 mm. The 2D array ultrasound probe 2 may have n*n transducer elements arranged in a matrix along two perpendicular axes X and Y, and transmit ultrasound along an axis Z perpendicular to the XY plane. In a specific embodiment, the 2D array ultrasound probe 2 may have 1024 transducer elements T ij(32*32) with a pitch of 0.3 mm. The transducer elements may transmit at a center frequency of, for example, 1-10 MHz, for example, 3 MHz.

[0048] The control system may, for example, include a specific control unit 3 and a computer 4. In this embodiment, the control unit 3 is used to control the 2D array ultrasound probe 2 and acquire signals therefrom, while the computer 4 is used to control the control unit 3, generate a 3D image sequence based on the signals acquired by the control unit 3, and determine quantitative parameters therefrom. In a variant, a single electronic device may implement all the functions of the control unit 3 and the computer 4.

[0049] like Figure 2 As shown above, the control unit 3 may include, for example:

[0050] - n transducers T connected to the 2D array ultrasound probe 2 ij n*n analog / digital converters 5(AD ij );

[0051] n*n buffer memories 6 (B ij );

[0052] - a central processing unit 7 (CPU) communicating with the buffer memory 6 and the computer 4;

[0053] - a memory 8 (MEM) connected to the central processing unit 7;

[0054] - a digital signal processor 9 (DSP) connected to the central processing unit 7.

[0055] The device can be operated as follows.

[0056] (a) Collection

[0057] The 2D array ultrasound probe 2 is placed on the chest 10 of the patient 1, usually between two ribs, in front of the patient's heart 12, as shown in FIG. Figure 4 shown.

[0058] Because the intercostal space between the ribs 11 is limited compared to the size of the heart 12 to be imaged, the 2D array ultrasound probe 2 is controlled to transmit divergent ultrasound waves, such as spherical ultrasound waves (i.e., having a spherical wavefront O1), into the chest 10. The control system can be programmed so that ultrasound waves are transmitted at a rate of several thousand ultrasound waves per second, for example, more than 10,000 unfocused ultrasound waves per second.

[0059] Spherical waves can be generated by a single transducer element (with low amplitude) or, more advantageously, by a large portion of a matrix array using one or more virtual point sources T' ij Produced with a higher amplitude, a virtual point source T'ij A virtual array 2' is formed and placed behind or in front of the 2D array ultrasound probe 2, as shown in FIG. Figure 3-4 As shown. The control system is applied to the The location associated with the virtual source v The transmission delay TD of the transducer element e is:

[0060]

[0061] where c is the speed of sound.

[0062] For each virtual source T' used ij , the control system can activate only a subset 2a of the 2D array ultrasound probe 2, which has a sub-aperture L that determines the aperture angle α of the diverging ultrasound waves. The aperture angle α can be, for example, 90°. The imaging depth along the axis Z can be about 12-15 cm.

[0063] Only one virtual source T' can be used ij , whereby one ultrasound wave is used for each 3D image of the heart, as will be explained later.

[0064] However, in order to improve image resolution and contrast, it is useful to transmit the unfocused ultrasound wave in a series of continuous unfocused ultrasound waves, each series of continuous unfocused ultrasound waves having a different propagation direction: in this case, each 3D image is synthesized by the signal acquired from one of the series of continuous unfocused ultrasound waves, as will be explained later. Each series of continuous ultrasound waves can be generated by placing the virtual source T ij By changing from one wave to another, thus changing the wavefronts O1, O2, etc., e.g. Figure 5 Each series may include, for example, 1-81 continuous ultrasonic waves in different directions, for example, 3-25 continuous ultrasonic waves in different directions, for example, 5-20 continuous ultrasonic waves in different directions, for example, 10-20 continuous ultrasonic waves in different directions.

[0065] In all cases, after each ultrasound wave is transmitted, the backscattered echo is collected by the 2D array ultrasound probe (eg, at a sampling rate of 12 MHz) and stored. This raw data (often also referred to as RF data or radio frequency data) is used to generate a 3D image sequence.

[0066] The duration of the acquisition can be from 10 ms to several cardiac cycles, for example, at least a portion of a cardiac cycle (e.g., diastole or systole, preferably diastole, or one cardiac cycle) and less than 10 cardiac cycles (e.g., less than 5 cardiac cycles). The duration can be, for example, 1 s-10 s (e.g., less than 5 s). In a specific embodiment, the duration is about 1.5 s.

[0067] An electrocardiogram (ECG) was co-recorded during the acquisition.

[0068] (b) Imaging:

[0069] After receiving the backscatter echo, parallel beamforming can be applied directly by the control system to reconstruct a 3D image from each single ultrasound wave. Delay and sum beamforming can be used in the time domain or the Fourier domain. In the time domain, delay and sum beamforming is applied to the signal received by each transducer element e to reconstruct the image placed on the The delay of a voxel in is the sum of the forward propagation time from the virtual source v to the voxel and the backscattered propagation time to the transducer element e:

[0070] Delay = forward delay + backscatter delay

[0071]

[0072] Another possibility is to use Fourier domain imaging (spatial frequency, k-space).

[0073] When ultrasound waves are emitted as a series of waves each having a different propagation direction, as described above, each image can be obtained by the control system through a process known as composite imaging. The voxels are beamformed using a delay-and-sum algorithm for each virtual source and then synergistically combined to form a final high-quality 3D image. Details of this composite imaging can be found, for example, in the following literature:

[0074] Montaldo,G.,Tanter,M.,Bercoff,J.,Benech,N.,Fink,M.,2009.Coherentplane-wave compounding for very high frame rate ultrasonography and transientelastography.IEEE Trans.Ultrason.Ferroelectr.Freq.Control 56,489–506.doi:10.1109 / TUFFC.2009.1067

[0075] Nikolov,SI,2001.Synthetic aperture tissue and flow ultrasoundimaging.Orsted-DTU,Technical University of Denmark,Lyngby,Denmark.

[0076] Nikolov,SI,Kortbek,J.,Jensen,JA,2010.Practical applications ofsynthetic aperture imaging,in:2010 IEEE Ultrasonics Symposium(IUS).Presented at the 2010IEEE Ultrasonics Symposium(IUS),pp.350–358.doi:10.1109 / ULTSYM.2010.5935627

[0077] Lockwood,GR,Talman,JR,Brunke,SS,1998.Real-time 3-Dultrasoundimaging using sparse synthetic aperture beamforming.IEEE Trans.Ultrason.Ferroelectr.Freq.Control 45,980–988.doi:10.1109 / 58.710573

[0078] Papadacci, C., Pernot, M., Couade, M., Fink, M. & Tanter, M. High-contrastultrafast imaging of the heart. IEEE transactions on ultrasonics, ferroelectrics, and frequency control 61, 288-301, doi:10.1109 / tuffc.2014.6722614(2014).

[0079] The frame rate, ie the rate of 3D images in the final animation sequence, may be several thousand 3D images per second, for example 3000-5000 3D images per second.

[0080] (c) Determine the time window:

[0081] The time window in which the tissue velocity reaches a minimum can be identified by the control system using known methods.

[0082] In a specific embodiment, the time window in which the tissue velocity reaches a minimum value can be defined as:

[0083] - a time window in which myocardial velocity is less than 5 cm / s, or

[0084] - The time windows corresponding to the start and end of diastole.

[0085] In a specific embodiment, the time window is determined by electrocardiography.

[0086] In another embodiment, the time window is determined by a tissue motion estimation performed by the control system using the following method.

[0087] (d) Calculation of blood and tissue velocity:

[0088] Blood flow and tissue motion estimation can be performed by the control system using known methods.

[0089] For example, the Kasai algorithm can be used to estimate motion in blood and tissue using half-wavelength spatial sampling (Kasai, C., Namekawa, K., Koyano, A., Omoto, R., 1985. Real-Time Two-Dimensional Blood Flow Imaging Using an Autocorrelation Technique. IEEE Trans. Sonics Ultrason. 32, 458–464. doi: 10.1109 / T-SU.1985.31615). Blood flow can be estimated by first applying a high-pass filter to the baseband data. Then, for each individual voxel, power Doppler can be obtained by integrating the power spectral density, pulsed Doppler can be obtained by calculating the short-time Fourier transform, and color Doppler images can be obtained by estimating the first moment of the voxel-specific pulsed Doppler spectrogram. A power-velocity integral map can be obtained by calculating the time integral of the power times the velocity to obtain an image of a parameter related to flow velocity. Advanced filtering, such as spatiotemporal filtering based on singular value decomposition, can also be used to better remove clutter signals (Demené, C. et al. Spatiotemporal Clutter Filtering of Ultrafast Ultrasound Data Highly Increases Doppler and Ultrasound Sensitivity. IEEE transactions on medical imaging 34, 2271-2285, doi: 10.1109 / tmi.2015.2428634 (2015)).

[0090] In a specific embodiment:

[0091] - 4D tissue velocity can be calculated by performing a 1D cross-correlation to obtain the volume of tissue volume-volume axial displacement. Then, a low-pass Butterworth filter with a 60 Hz cutoff frequency is applied to the displacement. A myocardial 3D mask (specific for myocardial tissue) can be applied to remove the signal outside the muscle. To display 4D tissue velocity, the Software. In each voxel, a tissue velocity curve can be derived.

[0092] 4D color Doppler can be calculated by performing SVD filtering to remove the signal from tissue and retain only the signal from blood flow, as done, for example, in the aforementioned publication by Demené et al. A per-pixel 1D axial cross-correlation on the SVD-filtered voxels can be performed to obtain the color Doppler volume.

[0093] The myocardium can be segmented using integrated tissue velocity over the cardiac cycle and manual selection of contours on two perpendicular 2D slices. Elliptical interpolation can be used to obtain a three-dimensional representation.

[0094] More generally, step (d) comprises automatically calculating a 3D mapping of at least one parameter related to blood velocity and / or tissue velocity in the imaging volume based on the sequence of 3D images. The 3D mapping may comprise an animated sequence of 3D images of the calculated parameter. The parameter may be blood and / or tissue velocity or a component thereof.

[0095] (e) Positioning focus:

[0096] Depending on the quantitative parameter sought, at least one point of interest having a predetermined characteristic is located in the 3D image sequence.The at least one point of interest having a predetermined characteristic may be located automatically by a control system or manually by an operator.

[0097] When the quantitative parameter relates to blood velocity in a certain anatomical region, the control system can automatically locate, or the operator can manually locate, the point of interest as a blood velocity point in the anatomical region and in at least a portion of the 3D image sequence. In a specific embodiment, a Fourier transform over time can be performed at each voxel using a 60-sample sliding window to retrieve spectra at various locations in the volume. Automatic dealiasing can be performed according to Demené et al. The location of the point of interest can then be automatically detected by detecting blood flow maxima.

[0098] When the quantified parameter relates to tissue velocity at a specific anatomical location in the heart, the control system may automatically locate, or the operator may manually locate, the anatomical location in the 3D image sequence. This automatic location may be performed based on an anatomical model of the heart stored in the computer 4, or by selecting a point in the tissue.

[0099] When the quantitative parameter relates to a minimum tissue velocity in a particular anatomical region, the control system can automatically locate, or the operator can manually locate, the anatomical region in the 3D image sequence, and the point of interest is used as the point of minimum tissue velocity in the anatomical region in the 3D image sequence. For example, when the minimum tissue velocity of the myocardium must be calculated, the system determines the point of myocardium with the minimum velocity in the myocardium in the image sequence.

[0100] (f) Quantification

[0101] The desired quantitative parameters may then be calculated by the control system (and in particular by the computer 4 ) based on the previously determined point of interest and based on the peak blood or tissue velocity of such point of interest.

[0102] Note:

[0103] - in the step of locating points of interest (step (e)), locating said at least one point of interest based solely on said 3D mapping and its temporal profile;

[0104] - and in a quantification step (step (f)), automatically determining said at least one velocity at said at least one point of interest based solely on said 3D mapping and its temporal profile.

[0105] More generally, in the present disclosure, coronary blood flow may be localized using only spatial and temporal velocity information without any additional anatomical information.

[0106] Due to the fact that the present method involves 3D mapping to determine velocities in the entire imaging volume, no anatomical images are required, and in particular no B-mode anatomical images are required to determine the point of interest and the velocity at the point of interest. Therefore, the entire method of the present disclosure does not require B-mode imaging, and more generally does not require anatomical imaging, which makes the results of the present method faster.

[0107] Therefore, a method for imaging coronary blood flow in a biological heart is provided, the method comprising at least the following steps:

[0108] Step a) an acquisition step, wherein unfocused ultrasonic waves are transmitted in the heart by a 2D array ultrasonic transducer, and raw data from backscattered ultrasonic waves are acquired by the 2D array ultrasonic transducer;

[0109] Step b) an imaging step, wherein a sequence of N 3D volumetric coronary blood flow images of the living heart is generated from the raw data, the sequence of 3D images forming an animation showing the motion of the imaging volume of the heart;

[0110] Step c) a determination step, wherein at least one time window in which the movement of the heart is minimal is determined;

[0111] step d) a calculation step, wherein a 3D mapping of at least one parameter related to coronary blood flow velocity is automatically calculated in said imaging volume based on a sequence of N 3D coronary blood flow images corresponding to at least one time window identified in step c);

[0112] step e) a positioning step, wherein at least one point of interest having predetermined characteristics is positioned in the sequence of N 3D coronary blood flow images corresponding to at least one time window identified in step c) based solely on the 3D mapping of step d);

[0113] Step f) a quantification step, wherein the coronary blood flow velocity is automatically determined at the at least one point of interest of step e) and a predetermined quantitative parameter related to the coronary blood flow velocity is automatically calculated; the coronary blood flow velocity is automatically determined at the at least one point of interest based solely on the 3D mapping of step d).

[0114] The method may further include one and / or another of the following features:

[0115] - determining at least one time window of step c) by means of electrocardiography;

[0116] - The determination step c) comprises the following steps:

[0117] step i) an imaging step, wherein a sequence of N 3D volume tissue images of the living heart is generated from the raw data of step a), the sequence of 3D images forming an animation showing the motion of the imaging volume of the heart,

[0118] step ii) a calculation step, wherein a 3D mapping of at least one parameter related to the velocity of cardiac tissue is automatically calculated in said imaging volume based on said sequence of N 3D volume tissue images showing the motion of the imaging volume of the heart,

[0119] step iii) a cardiac tissue motion estimation step, wherein at least one point of interest having predetermined characteristics is located in the sequence of N 3D volume tissue images based solely on the 3D mapping of step ii), and wherein the tissue velocity at the at least one point of interest is automatically determined; and

[0120] step iv) a determination step, wherein the time window in which the tissue velocity quantified in step iii) reaches a minimum velocity is determined;

[0121] - the tissue imaging step of step i) and the blood flow imaging step of step b) are performed simultaneously;

[0122] - the minimum speed in step iv) is less than 5 cm / s;

[0123] - repeating steps a)-f) for each cardiac cycle;

[0124] - said at least one time window of step c) corresponds to the beginning and the end of diastole;

[0125] - selecting the quantitative parameter of step f) from flow rate, maximum speed, average speed or time speed distribution;

[0126] - a tracking step in which microbubbles or ultrasound contrast agents are tracked in a patient to whom microbubbles or ultrasound contrast agents injected in their vascular system have previously been administered and their trajectory and speed are determined;

[0127] - performing said tissue motion estimation by means of a Doppler estimator or speckle tracking;

[0128] - performing said coronary blood flow 3D mapping by means of Doppler power imaging, Doppler color imaging or speckle tracking;

[0129] - estimating the tissue motion during the time window of step c) and applying motion correction, the estimation of the tissue motion comprising the following steps:

[0130] step 1) an imaging step, wherein a sequence of N 3D volume tissue images of the living heart is generated from the raw data of step a) corresponding to the time window of step c), the sequence of 3D images forming an animation showing the movement of the imaging volume of the heart during the time window of step c),

[0131] step 2) a calculation step, wherein a 3D mapping of at least one parameter related to cardiac tissue velocity is automatically calculated in said imaging volume based on said sequence of N 3D volume tissue images showing the motion of the imaging volume of the heart during the time window of step c), and

[0132] Step 3) a cardiac tissue motion estimation step, wherein at least one point of interest having predetermined characteristics is located in the sequence of N 3D volume tissue images based solely on the 3D mapping of step 2), and wherein the tissue velocity at the at least one point of interest is automatically determined;

[0133] - Automatic image registration using the calculated continuous coronary blood flow 3D images;

[0134] - Bubble or ultrasound contrast agent tracking steps include spatiotemporal filtering or machine learning;

[0135] - Automatically quantify the density of the coronary vessels;

[0136] - Automatically quantify the volume of blood perfused in volumetric units;

[0137] -Automatic detection of stenosis by accelerating blood flow velocity;

[0138] -Coronary flow reserve index is obtained by estimating the change in coronary blood flow velocity in patients who have previously been administered vasodilators;

[0139] - an automatic segmentation step of the central cavity;

[0140] - automatically locating or manually locating by an operator at least one point of interest having the predetermined characteristics of the locating step e);

[0141] - automatically locating or manually locating by an operator at least one point of interest having the predetermined characteristics of the motion estimation step iii);

[0142] - Automatically locating or manually locating by an operator at least one point of interest having the predetermined characteristics of the motion estimation step 3).

[0143] Furthermore, a device for performing 4D imaging of coronary artery blood flow of a biological heart according to the above method is disclosed, wherein the device comprises at least a 2D array ultrasound probe (2) and a control system (3, 4), wherein the control system (3, 4) is configured to:

[0144] (a) transmitting unfocused ultrasonic waves in the heart through a 2D array ultrasonic transducer, and acquiring raw data from backscattered ultrasonic waves through the 2D array ultrasonic transducer;

[0145] (b) generating a sequence of N 3D volumetric coronary blood flow images of the living heart from the raw data, the sequence of 3D images forming an animation showing motion of the imaging volume of the heart;

[0146] (c) identifying at least one time window in which cardiac motion is minimal;

[0147] (d) automatically calculating a 3D mapping of at least one parameter related to coronary blood flow velocity in the imaging volume based on a sequence of N 3D coronary blood flow images corresponding to at least one time window identified in (c);

[0148] (e) locating at least one point of interest having a predetermined characteristic in the sequence of N 3D coronary blood flow images corresponding to at least one time window identified in (c) based solely on the 3D mapping of (d);

[0149] (f) Automatically determining the coronary blood flow velocity at at least one point of interest of (e) based solely on the 3D mapping of (d), and automatically calculating a predetermined quantitative parameter related to the coronary blood flow velocity.

[0150] The device may further include one and / or another of the following features:

[0151] - In (c), the device is configured to:

[0152] (i) generating a sequence of N 3D volume tissue images of the living heart from the raw data of (a), the sequence of 3D images forming an animation showing the motion of the imaging volume of the heart,

[0153] (ii) automatically computing a 3D mapping of at least one parameter related to cardiac tissue velocity in an imaging volume based on the sequence of N 3D volumetric tissue images showing motion of the imaging volume of the heart,

[0154] (iii) based solely on the 3D mapping of (ii), locating at least one point of interest having predetermined characteristics in the sequence of N 3D volumetric tissue images, and automatically determining tissue velocity at the at least one point of interest; and

[0155] (iv) determining the time window in which the tissue velocity quantified in (iii) reaches a minimum velocity.

[0156] In (c), the device is configured to determine the at least one time window of (c) by electrocardiography.

[0157] Also disclosed is a computer-readable medium for performing 4D imaging of coronary blood flow in a biological heart according to the above method, the computer-readable medium comprising instructions that, when executed by a computer, cause the computer to perform the following steps:

[0158] Step a) transmitting unfocused ultrasonic waves in the heart through a 2D array ultrasonic transducer, and acquiring raw data from backscattered ultrasonic waves through the 2D array ultrasonic transducer;

[0159] Step b) generating a sequence of N 3D volumetric coronary blood flow images of the living heart from the raw data, wherein the sequence of 3D images forms an animation showing the motion of the imaging volume of the heart;

[0160] Step c) identifying at least one time window in which the motion of the heart is minimal;

[0161] step d) automatically calculating a 3D mapping of at least one parameter related to coronary blood flow velocity in the imaging volume based on a sequence of N 3D coronary blood flow images corresponding to at least one time window identified in step c);

[0162] step e) locating at least one point of interest having predetermined characteristics in the sequence of N 3D coronary blood flow images corresponding to at least one time window identified in step c) based solely on the 3D mapping of step d);

[0163] Step f) automatically determines the coronary blood flow velocity at at least one point of interest in step e) based solely on the 3D mapping in step d), and automatically calculates a predetermined quantitative parameter related to the coronary blood flow velocity.

[0164] The computer-readable medium may further include instructions that, when executed by a computer, cause the computer to perform the following steps:

[0165] - In step c), the computer is configured to:

[0166] (i) generating a sequence of N 3D volume tissue images of the living heart from the raw data of (a), the sequence of 3D images forming an animation showing the motion of the imaging volume of the heart,

[0167] (ii) automatically computing a 3D mapping of at least one parameter related to cardiac tissue velocity in an imaging volume based on the sequence of N 3D volumetric tissue images showing motion of the imaging volume of the heart,

[0168] (iii) based solely on the 3D mapping of (ii), locating at least one point of interest having predetermined characteristics in the sequence of N 3D volumetric tissue images, and automatically determining tissue velocity at the at least one point of interest; and

[0169] (iv) determining the time window in which the tissue velocity quantified in (iii) reaches a minimum velocity.

[0170] - the computer readable medium is arranged to perform step c) by means of electrocardiography. DETAILED DESCRIPTION

[0171] In a specific embodiment, a method for imaging coronary blood flow in a living being's heart comprises the following steps:

[0172] Step a) an acquisition step, wherein unfocused ultrasonic waves are transmitted in the heart by a 2D array ultrasonic transducer, and raw data from backscattered ultrasonic waves are acquired by the 2D array ultrasonic transducer;

[0173] Step b) an imaging step, wherein a sequence of N 3D volumetric coronary blood flow images of the living heart is generated from the raw data, the sequence of 3D images forming an animation showing the motion of the imaging volume of the heart;

[0174] Step c) determining step, wherein at least one time window in which the movement of the heart is minimal is determined, comprises the following steps:

[0175] (i) a calculation step, wherein a 3D mapping of at least one parameter related to the velocity of cardiac tissue is automatically calculated in the imaging volume based on the sequence of N 3D volume tissue images of step b) showing the motion of the imaging volume of the heart,

[0176] (ii) a cardiac tissue motion estimation step, wherein at least one point of interest having predetermined characteristics is located in the sequence of N 3D volumetric tissue images based solely on the 3D mapping of step (i), and wherein the tissue velocity at the at least one point of interest is automatically determined; and

[0177] (iii) a measuring step, wherein the time window in which the tissue velocity quantified in step (ii) reaches a minimum velocity is measured;

[0178] step d) a calculation step, wherein a 3D mapping of at least one parameter related to coronary blood flow velocity is automatically calculated in said imaging volume based on a sequence of N 3D coronary blood flow images corresponding to at least one time window identified in step c);

[0179] step e) a positioning step, wherein at least one point of interest having predetermined characteristics is positioned in the sequence of N 3D coronary blood flow images corresponding to at least one time window identified in step c) based solely on the 3D mapping of step d);

[0180] Step f) a quantification step, wherein the coronary blood flow velocity is automatically determined at the at least one point of interest of step e) and a predetermined quantitative parameter related to the coronary blood flow velocity is automatically calculated; the coronary blood flow velocity is automatically determined at the at least one point of interest based solely on the 3D mapping of step d).

[0181] In another specific embodiment, a method for imaging coronary blood flow in a heart of a living being comprises the following steps:

[0182] Step a) an acquisition step, wherein unfocused ultrasonic waves are transmitted in the heart by a 2D array ultrasonic transducer, and raw data from backscattered ultrasonic waves are acquired by the 2D array ultrasonic transducer;

[0183] Step b) an imaging step, wherein a sequence of N 3D volumetric coronary blood flow images of the living heart is generated from the raw data, the sequence of 3D images forming an animation showing the motion of the imaging volume of the heart;

[0184] Step c) a determination step, in which at least one time window is determined in which the movement of the heart is minimal, said time window being determined with the aid of an electrocardiogram, wherein said at least one time window preferably corresponds to the beginning and the end of diastole;

[0185] step d) a calculation step, wherein a 3D mapping of at least one parameter related to coronary blood flow velocity is automatically calculated in said imaging volume based on a sequence of N 3D coronary blood flow images corresponding to at least one time window identified in step c);

[0186] step e) a positioning step, wherein at least one point of interest having predetermined characteristics is positioned in the sequence of N 3D coronary blood flow images corresponding to at least one time window identified in step c) based solely on the 3D mapping of step d);

[0187] Step f) a quantification step, wherein the coronary blood flow velocity is automatically determined at the at least one point of interest of step e) and a predetermined quantitative parameter related to the coronary blood flow velocity is automatically calculated; the coronary blood flow velocity is automatically determined at the at least one point of interest based solely on the 3D mapping of step d).

[0188] The references in this disclosure are listed in the following order:

[0189] [1] PG Camici, G. d'Amati, O. Rimoldi, Nat. Rev. Cardiol. 12, 48–62 (2015).

[0190] [2]SDFihn et al., J. Thorac. Cardiovasc. Surg. 149, e5-23 (2015).

[0191] [3] JA Ambrose, DHIsrael, Curr. Opin. Cardiol. 5, 411–416 (1990).

[0192] [4] A. Sharma, A. Arbab-Zadeh, J. Nucl. Cardiol. 19, 796–806 (2012).

[0193] [5]WYKim et al., N.Engl.J.Med.345,1863–1869(2001).

[0194] [6] TPvan de Hoef, M. Siebes, JAESpaan, JJPiek, Eur. Heart J. 36, 3312–3319a (2015).

[0195] [7]EJVelazquez et al., N.Engl.J.Med.374,1511–1520(2016).

[0196] [8]Task Force Members et al.,Eur.Heart J.34,2949–3003(2013).

[0197] [9]AI J.Bourque,Curr.Cardiol.Rep.18,1(2016).

[0198]

[10] A.Berrington de Gonzalez et al.,Br.J.Cancer.114,388–394(2016).

[0199]

[11] N.Journy et al.,Circ.Cardiovasc.Interv.11,e006765(2018).

[0200]

[12] E.Macéet al.,Nat.Methods.8,662–664(2011).

[0201]

[13] C.Demenéet al.,IEEE Trans.Med.Imaging.34,2271–2285(2015).

[0202]

[14] D.Maresca et al.,JACC Cardiovasc.Imaging.11,798–808(2018).

[0203]

[15] WO 2019 / 158741 A1

Claims

1. A method for imaging coronary blood flow in a living heart, the method comprising at least the following steps: Step a) an acquisition step, wherein unfocused ultrasonic waves are transmitted in the heart by a 2D array ultrasonic transducer, and raw data from backscattered ultrasonic waves are acquired by the 2D array ultrasonic transducer; Step b) an imaging step, wherein a sequence of N 3D volumetric coronary blood flow images of the heart is generated from the raw data, the sequence of N 3D volumetric coronary blood flow images forming an animation showing the motion of the imaging volume of the heart; Step c) a determination step, wherein at least one time window in which the movement of the heart is minimal is determined; step d) a calculation step, wherein a 3D mapping of at least one parameter related to the coronary blood flow velocity is automatically calculated in said imaging volume based on a sequence of N 3D volumetric coronary blood flow images corresponding to at least one time window determined in step c); step e) a positioning step, wherein at least one point of interest having predetermined characteristics is positioned in the sequence of N 3D volumetric coronary blood flow images corresponding to at least one time window determined in step c) based solely on the 3D mapping of step d) and its temporal profile; Step f) a quantification step, wherein the coronary blood flow velocity is automatically determined at at least one point of interest in step e) and predetermined quantitative parameters related to the coronary blood flow velocity are automatically calculated; the coronary blood flow velocity is automatically determined at the at least one point of interest based solely on the 3D mapping of step d) and its time profile.

2. The method according to claim 1, wherein the determining step c) comprises the following steps: step i) an imaging step, wherein a sequence of N 3D volume tissue images of the heart is generated from the raw data of step a), the sequence of N 3D volume tissue images of the heart forming an animation showing the motion of the imaging volume of the heart, step ii) a calculation step, wherein a 3D mapping of at least one parameter related to the velocity of cardiac tissue is automatically calculated in said imaging volume based on said sequence of N 3D volume tissue images showing the motion of the imaging volume of the heart, step iii) a cardiac tissue motion estimation step, wherein at least one point of interest having predetermined characteristics is located in the sequence of N 3D volume tissue images based solely on the 3D mapping of step ii), and wherein the tissue velocity at the at least one point of interest is automatically determined; and Step iv) a determination step, wherein the time window in which the tissue velocity quantified in step iii) reaches a minimum velocity is determined.

3. The method according to claim 2, wherein the minimum speed of step iv) is less than 5 cm / s. The method according to claim 1 , wherein the at least one time window of step c) is determined by electrocardiography.

5. The method according to any one of claims 1 to 4, wherein the at least one time window of step c) corresponds to the beginning and the end of diastole, and the quantitative parameter of step f) is selected from the group consisting of maximum velocity, mean velocity or temporal velocity distribution of coronary blood flow.

6. The method according to claim 1, further comprising a tracking step, wherein ultrasound contrast agents are tracked in a patient to whom ultrasound contrast agents injected in their vascular system have been previously administered and their trajectory and velocity are determined. The method of claim 1 , further comprising performing tissue motion estimation by a Doppler estimator or speckle tracking. The method of claim 6 , wherein the ultrasound contrast agent tracking step comprises spatiotemporal filtering or machine learning.

9. The method of claim 1 , further comprising automatically quantifying the density of coronary vessels; and automatically quantifying the volume of blood perfused in volumetric units; automatically detecting stenosis by acceleration of blood flow velocity; obtaining a coronary flow reserve index by estimating changes in coronary blood flow velocity in patients who have previously been administered vasodilators; and further comprising the step of automatically segmenting the central lumen.

10. A device for performing 4D imaging of coronary blood flow in a biological heart, the device comprising at least a 2D array ultrasound probe (2) and a control system (3, 4), wherein the control system (3, 4) is configured to: (a) transmitting unfocused ultrasonic waves in the heart through a 2D array ultrasonic transducer, and acquiring raw data from backscattered ultrasonic waves through the 2D array ultrasonic transducer; (b) generating a sequence of N 3D volumetric coronary blood flow images of the heart from the raw data, the sequence of N 3D volumetric coronary blood flow images forming an animation showing motion of an imaging volume of the heart; (c) determining at least one time window during which cardiac motion is minimal; (d) automatically calculating a 3D mapping of at least one parameter related to coronary blood flow velocity in the imaging volume based on a sequence of N 3D volumetric coronary blood flow images corresponding to at least one time window determined in (c); (e) locating at least one point of interest having a predetermined characteristic in the sequence of N 3D volumetric coronary blood flow images corresponding to at least one time window determined in (c) based solely on the 3D mapping of (d) and its temporal profile; (f) Automatically determining the coronary blood flow velocity at at least one point of interest of (e) based solely on the 3D mapping of (d) and its time profile, and automatically calculating predetermined quantitative parameters related to the coronary blood flow velocity.

11. The apparatus of claim 10, wherein in (c), the apparatus is configured to: (i) generating a sequence of N 3D volume tissue images of the heart from the raw data of (a), the sequence of N 3D volume tissue images forming an animation showing the motion of the imaging volume of the heart, (ii) automatically computing a 3D mapping of at least one parameter related to cardiac tissue velocity in an imaging volume based on the sequence of N 3D volumetric tissue images showing motion of the imaging volume of the heart, (iii) based solely on the 3D mapping of (ii), locating at least one point of interest having predetermined characteristics in the sequence of N 3D volumetric tissue images, and automatically determining tissue velocity at the at least one point of interest; and (iv) determining the time window in which the tissue velocity quantified in (iii) reaches a minimum velocity.

12. The apparatus of claim 11, wherein in (c), the apparatus is configured to determine the at least one time window of (c) by electrocardiography.

13. A computer-readable medium comprising instructions which, when executed by a computer, cause the computer to: Step a) transmitting unfocused ultrasonic waves in the heart through a 2D array ultrasonic transducer, and acquiring raw data from backscattered ultrasonic waves through the 2D array ultrasonic transducer; Step b) generating a sequence of N 3D volumetric coronary blood flow images of the heart from the raw data, wherein the sequence of N 3D volumetric coronary blood flow images forms an animation showing the motion of the imaging volume of the heart; Step c) determining at least one time window in which the movement of the heart is minimal; step d) automatically calculating a 3D mapping of at least one parameter related to the coronary blood flow velocity in the imaging volume based on a sequence of N 3D volumetric coronary blood flow images corresponding to at least one time window determined in step c); step e) locating at least one point of interest having predetermined characteristics in the sequence of N 3D volumetric coronary blood flow images corresponding to at least one time window determined in step c) based solely on the 3D mapping of step d) and its time profile; Step f) automatically determines the coronary blood flow velocity at at least one point of interest in step e) based solely on the 3D mapping in step d) and its time profile, and automatically calculates predetermined quantitative parameters related to the coronary blood flow velocity.

14. The computer-readable medium of claim 13, further comprising instructions for performing step c), which, when executed by a computer, cause the computer to perform the following steps: (i) generating a sequence of N 3D volume tissue images of the heart from the raw data of (a), the sequence of N 3D volume tissue images forming an animation showing the motion of the imaging volume of the heart, (ii) automatically computing a 3D mapping of at least one parameter related to cardiac tissue velocity in an imaging volume based on the sequence of N 3D volumetric tissue images showing motion of the imaging volume of the heart, (iii) based solely on the 3D mapping of (ii), locating at least one point of interest having predetermined characteristics in the sequence of N 3D volumetric tissue images, and automatically determining tissue velocity at the at least one point of interest; and (iv) determining the time window in which the tissue velocity quantified in (iii) reaches a minimum velocity.

15. The computer-readable medium of claim 14, comprising instructions that, when executed by a computer, cause the computer to perform step c) via electrocardiography.

Images