Signal modulation based blood vessel selection arterial spin labeling encoding method and apparatus

Abstract

The application relates to the technical field of medical imaging, in particular to a blood vessel selection arterial spin labeling coding method and device based on signal modulation. The method comprises the following steps: generating a plurality of labeling signal modulation maps on a labeling plane; acquiring the number of blood vessels to be coded and the spatial position of the blood vessels to be coded on the labeling plane; determining a plurality of groups of alternative coding mode combinations; selecting the labeling signal modulation map of each coding mode of each group of alternative coding mode combinations; identifying the corresponding parameter combination and blood vessel labeling signal value in the labeling signal modulation map, so as to form a blood vessel labeling signal value matrix corresponding to each group of alternative coding mode combinations; calculating the loss value of each group of alternative coding mode combinations; selecting a target coding mode combination according to the loss value; feeding back to a magnetic resonance scanner; and performing blood vessel coding and scanning of the blood vessels to be coded by the magnetic resonance scanner. Thus, the problems of the related art, such as the requirement that the signal modulation curve is a cosine function, low universality and low coding accuracy, are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical imaging, in particular to a blood vessel selection arterial spin labeling encoding method and device based on signal modulation. BACKGROUND

[0002] ASL (Arterial Spin Labeling) MRI (Magnetic Resonance Imaging) is a blood perfusion imaging method, which plays an important role in CBF (Cerebral Blood Flow) quantification and collateral circulation evaluation. Due to its complete non-invasiveness and repeatable measurement, it has been widely used in clinical work such as brain tumors, strokes, and neurodegenerative diseases. PCASL (pseudo-continuous ASL) has good signal-to-noise ratio and high labeling efficiency, and is the most widely used ASL perfusion imaging method in clinical use. Its imaging principle is mainly to use a long pulse array, as shown in FIG. 1, to perform special phase control on the arterial blood downstream of the neck, so that the flowing blood signal is inverted. The arterial blood is used as an endogenous tracer, and after a PLD (Post Labeling Delay), the perfusion process of the arterial blood flowing into the static brain tissue can be directly observed. Figure 1

[0003] Vessel supply area is another important concept in blood perfusion evaluation. Obtaining the supply area of a blood vessel in brain tissue is very important information in tumor blood supply tracing, collateral circulation evaluation, and the like. VEASL (Vessel-Encoded ASL) is the earliest blood vessel selection imaging method based on PCASL to simultaneously divide the supply areas of multiple blood vessels. As shown in FIG. 2, this method applies a magnetic field gradient in the x-y plane, i.e., Gx and Gy in a, to generate a periodic "signal modulation" in the labeling plane, i.e., a periodic labeling stripe in a specific direction, as shown in b; to achieve specific labeling of blood vessels at different positions, as shown in c; thereby distinguishing the signal patterns of different blood vessel supply areas and obtaining information of different blood vessel supply areas. By changing the direction and size of the magnetic field gradient applied in the x-y plane and the phase of the radio frequency signal, different labeling patterns can be generated in the labeling plane. In order to simultaneously obtain the supply areas of multiple blood vessels, VEASL needs to acquire image data under multiple labeling patterns. Figure 2 Figure 2 Figure 2 Figure 2

[0004] ​​​​​Designing the blood vessel encoding mode is one of the most critical steps for blood supply division of VEASL. Current blood vessel encoding techniques all assume that the signal modulation generated by VEASL is approximately a cosine function, that is, the switching of the labeling band and the control band is relatively smooth, and the width is comparable. This assumption is only valid under specific PCASL parameters, thus limiting the selection range of PCASL parameters in the VEASL sequence, and the labeling mode of part of the blood vessels may deviate from the preset value, so that the accuracy of the encoding cannot be guaranteed. In addition, these methods are limited to considering the blood vessels as a single point on the labeling plane, without considering the size of the blood vessels on the labeling plane. SUMMARY

[0005] The application provides a blood vessel selection arterial spin labeling encoding method and device based on signal modulation to solve the problem of low encoding accuracy in related technologies.

[0006] The first aspect of the application provides a blood vessel selection arterial spin labeling encoding method based on signal modulation, comprising the following steps: generating a plurality of labeling signal modulation maps on the labeling plane; obtaining the number of blood vessels to be encoded and the spatial position of the blood vessels to be encoded on the labeling plane; determining a plurality of groups of candidate encoding mode combinations according to the blood vessels to be encoded, selecting the labeling signal modulation map of each encoding mode of each group of candidate encoding mode combinations from the plurality of labeling signal modulation maps based on the spatial position; identifying the parameter combination and the blood vessel labeling signal value corresponding to the labeling signal modulation map of each encoding mode of each group of candidate encoding mode combinations, to form the blood vessel labeling signal value matrix corresponding to each group of candidate encoding mode combinations, and calculating the loss value of each group of candidate encoding mode combinations according to the parameter combination and the blood vessel labeling signal value matrix; selecting a target encoding mode combination from the plurality of groups of candidate encoding mode combinations according to the loss value, feeding back the parameter combination corresponding to each encoding mode in the target encoding mode combination to a magnetic resonance scanner, and performing blood vessel encoding arterial spin labeling encoding and scanning on the blood vessels to be encoded based on the parameter combination corresponding to each encoding mode in the target encoding mode combination.

[0007] Optionally, the parameter combination includes an angle, a wavelength, and a relative magnet center offset.

[0008] Optionally, generating a plurality of labeling signal modulation maps on the labeling plane comprises: abstracting the labeling plane into a first matrix, determining the representation dimension of the first matrix; given a parameter combination of the angle, the wavelength, and the relative magnet center offset of the signal modulation stripe, calculating the signal intensity of each spatial position on the labeling plane according to the signal modulation curve under specific pseudo-continuous arterial spin labeling parameters; for each group of parameter combinations, generating a labeling signal modulation map according to the representation dimension of the first matrix and the signal intensity of each spatial position.

[0009] Optionally, determining the multiple groups of candidate encoding mode combinations according to the to-be-encoded blood vessel comprises: determining a second matrix of a target order according to the number of the to-be-encoded blood vessel; removing one column from the second matrix of the target order, and randomly selecting multiple columns from the remaining columns, to form a group of candidate encoding modes according to the randomly selected multiple columns.

[0010] Optionally, selecting the marker signal modulation diagram of each encoding mode in each group of candidate encoding mode combinations based on the spatial position from the multiple marker signal modulation diagrams comprises: taking out the signal value at the position of the to-be-encoded blood vessel for any one marker signal modulation diagram; calculating the total value of the marker signal modulation diagram according to the signal value and the marker value in the encoding mode; and searching for the marker signal modulation diagram with the highest total value by traversing all the marker signal modulation diagrams.

[0011] Optionally, the loss value of the candidate encoding mode combination is calculated according to the following formula:

[0012]

[0013] Wherein, C is the condition number of the matrix; min(λ) is the shortest wavelength; motion is a preset motion displacement tolerance, used to control the motion sensitivity.

[0014] Optionally, the encoding of the to-be-encoded blood vessel by the blood vessel encoding arterial spin labeling comprises: setting the blood vessel encoding gradient and the radio frequency phase of the imaging sequence according to the angle, the wavelength and the offset from the center of the magnet corresponding to the encoding mode, to make the position of the to-be-encoded blood vessel produce a specific marker signal.

[0015] The second aspect of the present application provides a blood vessel selection arterial spin labeling encoding device based on signal modulation, comprising: a generation module for generating multiple marker signal modulation diagrams on a marker plane; an acquisition module for acquiring the number of to-be-encoded blood vessels and the spatial position of the to-be-encoded blood vessels on the marker plane; a determination module for determining multiple groups of candidate encoding mode combinations according to the to-be-encoded blood vessel, and selecting the marker signal modulation diagram of each encoding mode in each group of candidate encoding mode combinations based on the spatial position from the multiple marker signal modulation diagrams; an identification module for identifying the parameter combination and the blood vessel marker signal value corresponding to the marker signal modulation diagram of each encoding mode in each group of candidate encoding mode combinations, constructing the blood vessel marker signal value matrix corresponding to each group of candidate encoding mode combinations, and calculating the loss value of each group of candidate encoding mode combinations according to the parameter combination and the blood vessel marker signal value matrix; a selection module for selecting a target encoding mode combination from the multiple groups of candidate encoding mode combinations according to the loss value, and feeding back the parameter combination corresponding to each encoding mode in the target encoding mode combination to a magnetic resonance scanner; and the magnetic resonance scanner encodes and scans the to-be-encoded blood vessel by blood vessel encoding arterial spin labeling based on the parameter combination corresponding to each encoding mode in the target encoding mode combination.

[0016] The third aspect of the present application provides a medical device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the program to implement the signal modulation-based blood vessel selection arterial spin labeling encoding method of the first aspect.

[0017] The fourth aspect of the present application provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the signal modulation-based blood vessel selection arterial spin labeling encoding method of the first aspect.

[0018] Therefore, the present application includes the following beneficial effects:

[0019] The embodiments of the present application determine a plurality of groups of candidate encoding mode combinations by generating a plurality of label signal modulation maps on a label plane, and obtaining the number of blood vessels to be encoded and the spatial position of the blood vessels to be encoded on the label plane, select a label signal modulation map of each group of candidate encoding mode combinations from the plurality of label signal modulation maps based on the spatial position, identify the corresponding parameter combination and the blood vessel label signal value matrix, calculate the loss value of each group of candidate encoding mode combinations according to the two, select the target encoding mode combination from the plurality of groups of candidate encoding mode combinations according to the loss value, feed back to the magnetic resonance scanner, and the magnetic resonance scanner encodes the blood vessels to be encoded with the blood vessel encoding arterial spin labeling, which is compatible with a wider PCASL sequence parameter, fully utilizes the signal modulation of VEASL, and considers the size of the blood vessels, thereby improving the accuracy of the encoding. Therefore, the problems of requiring a signal modulation curve to be a cosine function, low universality, low accuracy of encoding, and low accuracy are solved.

[0020] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS

[0021] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the following description, including the accompanying drawings, in which:

[0022] Figure 1 A pseudo-continuous arterial spin labeling perfusion imaging sequence diagram provided by the related art;

[0023] Figure 2 A blood vessel encoding arterial spin labeling imaging sequence diagram and a typical label signal modulation map provided by the related art;

[0024] Figure 3 A basic principle diagram of a random encoding method provided by the related art;

[0025] Figure 4A schematic diagram of the basic principle of a blood vessel encoding mode design method provided by the related art through an optimized encoding strategy is shown in FIG. 1.

[0026] Figure 5 A flowchart of a blood vessel selection arterial spin labeling encoding method based on signal modulation provided according to an embodiment of the present application is shown in FIG. 2.

[0027] Figure 6 A labeling signal modulation curve under different PCASL parameters and flow rates provided according to an embodiment of the present application is shown in FIG. 3.

[0028] Figure 7 A typical labeling signal modulation diagram on a labeling plane provided according to an embodiment of the present application is shown in FIG. 4.

[0029] Figure 8 A flowchart of determining a labeling mode combination and a parameter combination provided according to an embodiment of the present application is shown in FIG. 5.

[0030] Figure 9 An example diagram of a blood vessel selection arterial spin labeling encoding device based on signal modulation provided according to an embodiment of the present application is shown in FIG. 6.

[0031] Figure 10 A structural schematic diagram of a medical device provided according to an embodiment of the present application is shown in FIG. 7. DETAILED DESCRIPTION

[0032] Embodiments of the present application are described in detail below with reference to the accompanying drawings, in which examples of the embodiments are shown, and in which like or similar designations denote like or similar elements or elements having the same or similar functions throughout. The embodiments described below by reference to the drawings are exemplary and are intended to explain the present application, and cannot be understood as limiting the present application.

[0033] In the related art, a RE (Random Encoding) method is to correlate the acquired VEASL signal with the simulation signal within a certain range of the labeling plane by 60 pairs of randomly generated encoding modes, so as to trace the potential blood supply artery source, such as Figure 3 as shown in A of FIG. 8, i.e., all the blood vessel positions on the labeling plane that can be effectively encoded, Figure 3 as shown in D of FIG. 8; the 60 encoding modes corresponding to these traced blood vessel positions are the encoding matrix referenced by the VEASL blood supply division. This method is completely based on data driving, and can resist the interference caused by partial resonance to a certain extent, such as Figure 3 as shown in B and C of FIG. 8, and does not need to accurately know the position of each blood vessel, so as to cope with a more complex blood vessel distribution situation, such as the area above the Willis circle. However, the current RE has the following shortcomings:

[0034] (1) For imaging sequences where imaging of a single coding pattern is slow, such as vascular imaging, the scanning time of 60 coding patterns is unacceptable; (2) In the step of marker plane signal simulation, the signal tuning is implicitly close to the cosine function, which restricts the parameter settings of PCASL; (3) Although it is not necessary to accurately obtain the coordinate information of each blood vessel, giving the marker plane a degree of freedom in placement, if the marker plane is completely parallel to a blood vessel segment or does not contain that blood vessel segment, it still cannot be correctly decoded; (4) The signal-to-noise ratio of VEASL is not the optimal strategy.

[0035] OES (Optimized Encoding Scheme) is a method for designing vascular coding schemes based on Fourier transform. Given vascular coordinates and an ideal vascular coding scheme (some vessels are in a labeled state with a signal value of -1, and some vessels are in a control state with a signal value of 1), OES constructs an ideal coding scheme diagram, such as... Figure 4 As shown in Figure a; perform a Fourier transform on this matrix as follows: Figure 4 As shown in b, the low-frequency components are weighted, and then the pixel with the largest weighted Fourier space amplitude is found, as shown in Figure b. Figure 4 As shown in Figure c, the optimal VEASL sequence design for this vessel coding pattern (including the phase of the RF pulse, the direction and magnitude of the gradient) is derived from the coordinates of this value in the complex plane. Figure 4 As shown in d, OES significantly reduces the number of vascular coding patterns by considering vascular coordinate information; and given vascular coordinates and coding patterns, it can automatically design a coding strategy with the optimal signal-to-noise ratio. Furthermore, because it weights the low-frequency Fourier space, OES can mitigate errors caused by head movement to some extent. In addition, OES can also incorporate the effects of partial resonance into the coding design, which is theoretically feasible for any number and distribution of vessels.

[0036] In addition, OES can also take into account the influence of partial resonance in the coding design, which is theoretically feasible for any number and distribution of blood vessels. However, OES also has obvious drawbacks: (1) This method relies on Fourier transform, that is, it assumes that the signal modulation should be a perfect cosine function, which only holds true under specific PCASL sequence parameter settings; (2) This method only provides the optimal signal-to-noise ratio strategy under a certain fixed coding mode, but it cannot obtain how to obtain an optimal coding combination under the premise of given blood vessel coordinates, and cannot design a coding matrix with the optimal signal-to-noise ratio efficiency; (3) This method assumes that the blood vessel is just a point on the marking plane and does not consider the influence of blood vessel geometry.

[0037] Based on the defects of the prior art, the application provides a blood vessel selection arterial spin labeling coding method and device based on signal modulation. Figure 5 A flowchart of the blood vessel selection arterial spin labeling coding method based on signal modulation provided by the application is shown in the figure.

[0038] As shown in the figure, the blood vessel selection arterial spin labeling coding method based on signal modulation comprises the following steps: Figure 5

[0039] In step S101, a plurality of labeling signal modulation maps on a labeling plane are generated.

[0040] The labeling plane refers to the spatial position of the arterial blood labeled in the arterial spin labeling process; the generation of the labeling signal modulation map will be described in detail below, and will not be described here.

[0041] In the embodiment of the application, the plurality of labeling signal modulation maps on the labeling plane are generated, comprising: abstracting the labeling plane into a first matrix, determining the representation dimension of the first matrix; given a parameter combination of the angle, wavelength, and offset of the signal modulation stripe relative to the center of the magnet, calculating the signal intensity of each spatial position on the labeling plane according to the signal modulation curve under specific pseudo-continuous arterial spin labeling parameters; for each parameter combination, generating a labeling signal modulation map according to the representation dimension of the first matrix and the signal intensity of each spatial position.

[0042] The first matrix is a two-dimensional matrix; the parameter combination will be described in detail below, and will not be described here; the shape of the signal modulation curve under specific pseudo-continuous arterial spin labeling parameters is determined by the pseudo-continuous arterial spin labeling parameters and blood flow velocity used by the blood vessel coding arterial spin labeling imaging sequence, the signal modulation curve is not affected by the parameter combination offset, but the position of the signal modulation curve can be determined according to the offset of the parameter combination, for example, the angle determines how much the curve is turned on the two-dimensional plane, the wavelength determines how much space distance each period actually corresponds to, and the offset determines the offset amount of the curve on the two-dimensional plane, no matter what pseudo-continuous arterial spin labeling parameters are set, the signal modulation curve can be considered, thereby realizing that the signal modulation curve is not forced to be a cosine function, and allowing more flexible pseudo-continuous arterial spin labeling parameter settings.

[0043] ​It can be understood that the embodiment of the application generates a plurality of marker signal modulation maps on the marker plane, first abstracts the marker plane into a two-dimensional matrix, that is, a first matrix, gives a set of parameter combinations, combines the signal modulation curve under the specific pseudo-continuous arterial spin labeling parameter, calculates the signal intensity of each spatial position on the marker plane, and for each parameter combination, generates a marker signal modulation map according to the representation dimension of the first matrix and the signal intensity of each spatial position, thereby generating a plurality of marker signal modulation maps under different parameter combinations by setting the parameter combinations, and constructing a marker signal modulation map dictionary.

[0044] In the embodiment of the application, the parameter combination includes an angle, a wavelength and a relative magnet center offset.

[0045] The relative magnet center offset represents the displacement of the marker signal modulation map relative to the center position of the magnetic resonance imaging device.

[0046] In step S102, the number of blood vessels to be encoded and the spatial positions of the blood vessels to be encoded on the marker plane are acquired.

[0047] It can be understood that the embodiment of the application first needs to acquire the number of blood vessels to be encoded and the spatial positions of the blood vessels on the marker plane, for determining the generation and optimization of the signal modulation map.

[0048] In step S103, a plurality of groups of candidate encoding mode combinations are determined according to the blood vessels to be encoded, and the marker signal modulation map of each encoding mode of each group of candidate encoding mode combinations is selected from the plurality of marker signal modulation maps based on the spatial positions.

[0049] It can be understood that the embodiment of the application can determine a plurality of groups of candidate encoding mode combinations through the blood vessels to be encoded, and the determination method will be described in detail below, which will not be described here. And the marker signal modulation map of each encoding mode of each group of candidate encoding mode combinations is selected from the plurality of marker signal modulation maps based on the spatial positions.

[0050] In the embodiment of the application, the plurality of groups of candidate encoding mode combinations are determined according to the blood vessels to be encoded, including: determining a second matrix of a target order according to the number of blood vessels to be encoded; removing a column full of ones in the second matrix of the target order, and randomly selecting a plurality of columns from the remaining columns, and according to the randomly selected plurality of columns, a group of candidate encoding modes is formed.

[0051] The second matrix is a Hadamard matrix; the second matrix of the target order is determined according to the number of blood vessels to be encoded, and the method is that if the number of blood vessels to be encoded is N, the target order of the second matrix is at least N+1.

[0052] It can be understood that, in the embodiment of the present application, the target order of the second matrix is determined to be at least N+1 by the number N of the blood vessels to be coded, and after the target order of the second matrix is determined, a column of all ones in the second matrix is removed, and a plurality of columns are randomly selected from the remaining columns to form a set of candidate coding modes. In this way, a plurality of sets of candidate coding modes can be obtained.

[0053] In the embodiment of the present application, selecting each coding mode marker signal modulation graph in each candidate coding mode combination based on the spatial position from the plurality of marker signal modulation graphs comprises: taking out the signal value at the position of the blood vessel to be coded for any one marker signal modulation graph; calculating the total value of the marker signal modulation graph according to the signal value and the marker value in the coding mode; and finding the marker signal modulation graph with the highest total value by traversing all the marker signal modulation graphs.

[0054] In the embodiment of the present application, the total value of the marker signal modulation graph is calculated according to the signal value and the marker value in the candidate coding mode combination, and the method is to multiply the signal value at the position of the blood vessel to be coded and the marker value in the coding mode, that is, +1 or -1 in the matrix, and then add all the blood vessels to obtain the total value of the marker signal modulation graph.

[0055] It can be understood that, in the embodiment of the present application, selecting each coding mode marker signal modulation graph in each candidate coding mode combination based on the spatial position from the plurality of marker signal modulation graphs comprises: taking out the signal value at the position of the blood vessel to be coded for any one marker signal modulation graph; calculating the total value of the marker signal modulation graph according to the signal value and the marker value in the coding mode; and finding the marker signal modulation graph with the highest total value by traversing all the marker signal modulation graphs.

[0056] In step S104, the parameter combination and the blood vessel marker signal value corresponding to the marker signal modulation graph of each coding mode in each candidate coding mode combination are identified to form a blood vessel marker signal value matrix corresponding to each candidate coding mode combination, and the loss value of each candidate coding mode combination is calculated according to the parameter combination and the blood vessel marker signal value matrix.

[0057] It can be understood that, in the embodiment of the present application, the parameter combination and the blood vessel marker signal value corresponding to the marker signal modulation graph of each coding mode in each candidate coding mode combination are used to form a blood vessel marker signal value matrix corresponding to each candidate coding mode combination, and the loss value of the candidate coding mode combination is calculated according to a specific formula. The formula will be described in detail below, and will not be described here. The lower the loss value, the better the performance of the coding mode under the influence of noise and head movement.

[0058] In the embodiment of the present application, the loss value of the alternative encoding mode combination is calculated according to the formula:

[0059]

[0060] Wherein, C is the condition number of the matrix; min(λ) is the shortest wavelength; motion is a preset motion displacement tolerance, used to control the motion sensitivity.

[0061] In step S105, the target encoding mode combination is selected from the multiple groups of alternative encoding mode combinations according to the loss value, and the parameter combination corresponding to each encoding mode in the target encoding mode combination is fed back to the magnetic resonance scanner. The magnetic resonance scanner encodes and scans the blood vessel to be encoded based on the parameter combination corresponding to each encoding mode in the target encoding mode combination.

[0062] It can be understood that, in the embodiment of the present application, the performance of each group of alternative encoding modes is evaluated by calculating the loss value of each group of alternative encoding modes, and the group with the lowest loss value is the target encoding mode combination. Then, the parameter combination corresponding to each encoding mode in the target encoding mode combination, including the angle, wavelength and relative magnet center offset, is fed back to the magnetic resonance scanner. The magnetic resonance scanner encodes the blood vessel to be encoded according to the optimized parameter combination, so as to realize high-precision blood vessel perfusion imaging.

[0063] In the embodiment of the present application, the encoding of the blood vessel to be encoded by the blood vessel encoding arterial spin labeling includes: setting the blood vessel encoding gradient and the radio frequency phase of the imaging sequence according to the angle, wavelength and offset to the magnet center corresponding to the encoding mode, so that the position of the blood vessel to be encoded generates a specific marker signal.

[0064] It can be understood that, in the embodiment of the present application, the parameter combination corresponding to each encoding mode in the target encoding mode combination, including the angle, wavelength and relative magnet center offset, is fed back to the magnetic resonance scanner. The magnetic resonance scanner sets the blood vessel encoding gradient and the radio frequency phase of the imaging sequence according to these parameters, so that the position of the blood vessel to be encoded generates a marker signal.

[0065] According to the blood vessel selection arterial spin labeling encoding method based on signal modulation provided in the embodiments of the present application, a plurality of labeling signal modulation maps on a labeling plane are generated, the number of blood vessels to be encoded and the spatial positions of the blood vessels to be encoded on the labeling plane are acquired, a plurality of groups of candidate encoding mode combinations are determined, the labeling signal modulation maps of each group of candidate encoding mode combinations are selected from the plurality of labeling signal modulation maps based on the spatial positions, the corresponding parameter combinations and blood vessel labeling signal value matrices are identified, the loss values of each group of candidate encoding mode combinations are calculated according to the two, the target encoding mode combination is selected from the plurality of groups of candidate encoding mode combinations according to the loss values, and the target encoding mode combination is fed back to a magnetic resonance scanner. The magnetic resonance scanner encodes the blood vessels to be encoded by using the blood vessel encoding arterial spin labeling. The method is compatible with more extensive PCASL sequence parameters, fully utilizes the signal modulation of VEASL, and considers the size of the blood vessels, and the accuracy of encoding is improved.

[0066] The blood vessel selection arterial spin labeling encoding method based on signal modulation is further described below through a specific embodiment. A new encoding method (MOdulation-Guided Encoding, MOGEN) provided in the embodiment is similar to the OES method, but discards the assumption that the signal modulation is a cosine function, and allows the use of a more general signal modulation function close to the actual situation. As shown in Figure 6 , the signal modulation curves in a period obtained by theoretical calculation are respectively under two different PCASL sequence parameters and different flow rates. As an example, the signal modulation curve under a flow rate of 30 cm / s is selected for MOGEN design. The following is a specific implementation mode of MOGEN:

[0067] (1) MOGEN generates the labeling signal modulation map on the labeling plane by the following method: first, the labeling plane is abstracted into a 1024*1024 matrix, and x and y represent two dimensions of the matrix. Given a combination of the angle (a), the wavelength (l), and the offset (e) relative to the center of the magnet (a, l, e), the signal intensity of each position on the labeling plane is calculated in combination with the above signal modulation curve, and a labeling signal modulation map is obtained, as shown in Figure 7 , the curve in the figure is a one-dimensional labeling signal modulation curve under a specific PCASL sequence parameter setting. In combination with the actual demand, the ranges and steps of a, l and e are set, the labeling signal modulation maps under all combinations are calculated, and a labeling signal modulation map dictionary is constructed.

[0068] (2) After the number of vessels to be encoded and their spatial positions on the labeling plane are given, we need to pick out several combinations from the above (a, l, e) combinations to apply to the VEASL sequence scanning based on a certain method. The specific implementation is as follows: (1) According to the number N of vessels to be encoded, a Hadamard matrix of the minimum order is determined. As an example, Figure 8 (a) is an 8-order Hadamard matrix, which can be used for VEASL imaging of no more than 7 vessels; (2) remove a column of all 1s in the Hadamard matrix, and randomly select N columns from the remaining columns to form a set of alternative encoding modes, as shown in Figure 8 (b); (3) For any one of the encoding modes in the set of alternative encoding modes, find a most matched graph in the above labeling signal modulation graph dictionary, so as to obtain a most matched (a, l, e) combination. The specific implementation is as follows: for any one labeling signal modulation graph, take out the signal values at the positions of the vessels to be encoded, multiply them with the ideal labeling values (i.e. +1 or -1) set in the encoding mode one by one, and then add all the vessels to obtain a total value. Traverse all the labeling signal modulation graphs, and the signal modulation graph with the highest total value is the most matched labeling signal modulation graph, and the (a, l, e) corresponding thereto is the most matched combination; (4) Based on the method of (3), a set of (a, l, e) combinations corresponding to the set of alternative encoding modes in (2) and the corresponding vessel labeling signal value matrix are obtained, as shown in Figure 8 (c); (5) As shown in Figure 8 , in order to select the optimal set from the multiple sets of alternative encoding modes randomly generated in (2), the loss value of each set of alternative encoding modes is calculated using the following formula:

[0069]

[0070] Where C is the condition number of the above vessel labeling signal value matrix, min(l) is the minimum value of all wavelengths l in the labeling mode, and motion is a preset motion displacement tolerance for controlling the sensitivity to motion. The lower the loss value, the stronger the resistance of the MOGEN-based VEASL imaging to noise and head motion. The set of encoding modes with the lowest loss value is the final selected encoding mode combination, and the corresponding (a, l, e) combination will be fed back to the magnetic resonance scanner for determining the vessel encoding gradient (size and direction) and radio frequency phase of VEASL.

[0071] MOGEN also employs the following methods to improve imaging and computational performance: (1) To further avoid the impact of head movement on coding accuracy, an exponentially decaying smoothing kernel can be used to smooth the signal modulation map in the ε dimension; (2) When generating the above signal modulation map, MOGEN can use either a single point or a ROI of a certain shape to represent a blood vessel. For the latter, it is only necessary to use multiple points on the marker plane to represent a blood vessel, and then optimize the search for the best matching signal modulation map ( Figure 8 (c) Consider the sum of the products of the ideal coding values ​​and the simulated signal values ​​of all blood vessel points. This design can improve the difference in the marking patterns between blood vessels when the distance between the blood vessels to be encoded is small; (3) In actual calculation, in order to speed up the calculation, the above signal intensity is only calculated for the location (x, y) of the blood vessel to form a signal modulation dictionary; (4) In the above step (ii), in order to ensure that the optimal set of candidate coding patterns is obtained, MOGEN randomly selects the column combinations of the Hadamard matrix a sufficient number of times; another method is to traverse all possible column combinations; (5) In order to improve the calculation effect, MOGEN will temporarily store the results of the processed coding patterns during the calculation process to avoid repeated calculations in the subsequent search process.

[0072] Next, referring to the accompanying drawings, a signal modulation-based arterial spin label encoding device for blood vessel selection according to an embodiment of this application is described.

[0073] Figure 9 This is a block diagram of a signal modulation-based arterial spin label encoding device for blood vessel selection according to an embodiment of this application.

[0074] like Figure 9 As shown, the signal modulation-based vascular selective arterial spin label encoding device 10 includes: a generation module 201, an acquisition module 202, a determination module 203, an identification module 204, and a selection module 205.

[0075] The generation module 201 is configured to generate a plurality of marker signal modulation maps on a marker plane; the acquisition module 202 is configured to acquire the number of blood vessels to be encoded and the spatial positions of the blood vessels to be encoded on the marker plane; the determination module 203 is configured to determine a plurality of groups of candidate encoding mode combinations according to the blood vessels to be encoded, select, based on the spatial positions, the marker signal modulation map of each encoding mode in each group of candidate encoding mode combinations from the plurality of marker signal modulation maps; the identification module 204 is configured to identify the parameter combination and the blood vessel marker signal value corresponding to the marker signal modulation map of each encoding mode in each group of candidate encoding mode combinations, and form the blood vessel marker signal value matrix corresponding to each group of candidate encoding mode combinations, calculate the loss value of each group of candidate encoding mode combinations according to the parameter combination and the blood vessel marker signal value matrix; the selection module 205 is configured to select a target encoding mode combination from the plurality of groups of candidate encoding mode combinations according to the loss value, and feed back the parameter combination corresponding to each encoding mode in the target encoding mode combination to a magnetic resonance scanner; and the magnetic resonance scanner performs blood vessel encoding and scanning of the blood vessel encoding arterial spin labeling based on the parameter combination corresponding to each encoding mode in the target encoding mode combination.

[0076] In the embodiment of the present application, the parameter combination includes an angle, a wavelength, and a relative magnet center offset.

[0077] In the embodiment of the present application, the plurality of marker signal modulation maps on the marker plane are generated, and the generation module 201 is further configured to: abstract the marker plane into a first matrix, determine the representation dimension of the first matrix; given a parameter combination of the angle, the wavelength, and the offset of the magnet center, calculate the signal intensity of each spatial position on the marker plane according to the signal modulation curve under a specific pseudo-continuous arterial spin labeling parameter; for each group of parameter combinations, generate a marker signal modulation map according to the representation dimension of the first matrix and the signal intensity of each spatial position.

[0078] In the embodiment of the present application, the plurality of groups of candidate encoding mode combinations are determined according to the blood vessels to be encoded, and the determination module 203 is further configured to: determine a second matrix of a target order according to the number of blood vessels to be encoded; remove a column in the second matrix of the target order, randomly select a plurality of columns from the remaining columns, and form a group of candidate encoding modes according to the randomly selected plurality of columns.

[0079] In the embodiment of the present application, the marker signal modulation map of each encoding mode in each group of candidate encoding mode combinations is selected from the plurality of marker signal modulation maps based on the spatial positions, and the determination module 203 is further configured to: for any one marker signal modulation map, take out the signal value at the position of the blood vessel to be encoded; calculate the total value of the marker signal modulation map according to the signal value and the marker value in the candidate encoding mode combination; and traverse all the marker signal modulation maps to find the marker signal modulation map with the highest total value.

[0080] In the embodiment of the present application, the loss value calculation formula of the alternative encoding mode combination is:

[0081]

[0082] Wherein, C is the condition number of the matrix; min (lambda) is the shortest wavelength; motion is a preset motion displacement tolerance, used to control the motion sensitivity.

[0083] In the embodiment of the present application, the selection module 205 is further used for: performing blood vessel encoding arterial spin labeling coding on the blood vessel to be encoded, comprising: setting the blood vessel encoding gradient and radio frequency phase of the imaging sequence according to the angle, wavelength and offset to the magnet center corresponding to the encoding mode, so as to make the blood vessel to be encoded position produce a specific marker signal.

[0084] It should be noted that the foregoing explanation and description of the embodiment of the blood vessel selection arterial spin labeling coding method based on signal modulation also applies to the embodiment of the blood vessel selection arterial spin labeling coding device based on signal modulation, which will not be described here.

[0085] According to the blood vessel selection arterial spin labeling coding device based on signal modulation provided in the embodiment of the present application, a plurality of marker signal modulation maps on the marker plane are generated, the number of blood vessels to be encoded and the spatial position of the blood vessels to be encoded on the marker plane are obtained, a plurality of groups of alternative encoding mode combinations are determined, the marker signal modulation map of each group of alternative encoding mode combinations is selected from the plurality of marker signal modulation maps based on the spatial position, the corresponding parameter combination and the blood vessel marker signal value matrix are identified, the loss value of each group of alternative encoding mode combinations is calculated according to the two, the target encoding mode combination is selected from the plurality of groups of alternative encoding mode combinations according to the loss value, and the magnetic resonance scanner is fed back. The magnetic resonance scanner performs blood vessel encoding arterial spin labeling coding on the blood vessel to be encoded, is compatible with a wider PCASL sequence parameter, fully utilizes the signal modulation of VEASL, and considers the size of the blood vessel, thereby improving the accuracy of the coding.

[0086] Figure 10 The medical device provided in the embodiment of the present application is shown in the structural schematic diagram. The medical device can include:

[0087] The memory 301, the processor 302, and the computer program stored in the memory 301 and executable on the processor 302.

[0088] The processor 302 executes the program to implement the blood vessel selection arterial spin labeling coding method based on signal modulation provided in the above embodiments.

[0089] Further, the medical device further includes:

[0090] The communication interface 303 is configured to communicate between the memory 301 and the processor 302.

[0091] The memory 301 is configured to store a computer program executable in the processor 302.

[0092] The memory 301 can include a high-speed RAM (Random Access Memory) memory, and can further include a non-volatile memory, for example, at least one disk memory.

[0093] If the memory 301, the processor 302 and the communication interface 303 are independently implemented, the communication interface 303, the memory 301 and the processor 302 can be connected through a bus and complete communication between each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, Figure 10 In the figure, only one thick line is used to represent, but it does not mean that there is only one bus or one type of bus.

[0094] Optionally, in a specific implementation, if the memory 301, the processor 302 and the communication interface 303 are integrated on a chip, the memory 301, the processor 302 and the communication interface 303 can complete communication between each other through an internal interface.

[0095] The processor 302 can be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement one or more embodiments of the present application.

[0096] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the program is executed by the processor to implement the above-mentioned blood vessel selection arterial spin labeling encoding method based on signal modulation.

[0097] In the description of the application, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the application. In the description of the application, the illustrative description of the above terms is not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or N embodiments or examples. In addition, different embodiments or examples described in the description of the application and the features of different embodiments or examples can be combined and combined by those skilled in the art without contradiction.

[0098] In addition, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the application, the meaning of "N" is at least two, for example, two, three, etc., unless otherwise specifically limited.

[0099] Any process or method descriptions in flow charts or otherwise described herein can be understood as representing code modules, segments, or portions of code that include one or more executable instructions for implementing the specified logical functions or steps, and the preferred embodiments of the application include additional or fewer steps or processes in addition to or other than those shown and discussed. The various steps or processes described herein can be carried out by a device (e.g., a computer) that desires to implement the functions of the present application, either locally or remotely.

[0100] It should be understood that parts of the application can be implemented in hardware, software, firmware or a combination thereof. In the above-described embodiments, the steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. As in another embodiment, if implemented in hardware, any of the following technologies known in the art or their combinations can be used: discrete logic circuit with logic gate circuit for implementing logical functions on data signals, application specific integrated circuit with suitable combination of logic gate circuit, programmable gate array, field programmable gate array, etc.

[0101] Those skilled in the art of the present technology can understand that all or part of the steps carried out by the above-described embodiments can be instructed by a program to complete the relevant hardware, and the above-mentioned program can be stored in a computer readable storage medium. The program, when executed, includes one or a combination of steps of the method embodiments.

[0102] Although the embodiments of the present application have been shown and described above, it is understood that the above-described embodiments are exemplary and are not to be construed as limiting the present application, and that changes, modifications, substitutions and variations can be made by those skilled in the art without departing from the scope of the present application.

Claims

1. A signal modulation based, vessel-selective, arterial spin labeling encoding method, characterized by, The method comprises the following steps: generating a plurality of marker signal modulation maps on a marker plane; acquiring the number of blood vessels to be encoded and the spatial positions of the blood vessels to be encoded on the marker plane; determining a plurality of groups of candidate encoding mode combinations according to the blood vessels to be encoded, and selecting, based on the spatial positions, a marker signal modulation map of each encoding mode in each group of candidate encoding mode combinations from the plurality of marker signal modulation maps; identifying a parameter combination and a blood vessel marker signal value corresponding to the marker signal modulation map of each encoding mode in each group of candidate encoding mode combinations, to form a blood vessel marker signal value matrix corresponding to each group of candidate encoding mode combinations, and calculating a loss value of each group of candidate encoding mode combinations according to the parameter combination and the blood vessel marker signal value matrix; selecting a target encoding mode combination from the plurality of groups of candidate encoding mode combinations according to the loss value, and feeding back a parameter combination corresponding to each encoding mode in the target encoding mode combination to a magnetic resonance scanner, so that the magnetic resonance scanner performs blood vessel encoding and scanning on the blood vessels to be encoded based on the parameter combination corresponding to each encoding mode in the target encoding mode combination.

2. The signal modulation based, blood vessel selective arterial spin labeling encoding method of claim 1, wherein, The parameter combination comprises an angle, a wavelength, and a relative magnet center offset.

3. The signal modulation based, blood vessel selective, arterial spin labeling encoding method of claim 1, wherein, The generating of the plurality of marker signal modulation maps on the marker plane comprises: abstracting the marker plane into a first matrix, and determining a representation dimension of the first matrix; given a parameter combination of an angle, a wavelength, and a relative magnet center offset of a signal modulation stripe, calculating a signal intensity of each spatial position on the marker plane according to a signal modulation curve under a specific pseudo-continuous arterial spin labeling parameter; for each parameter combination, generating a marker signal modulation map according to the representation dimension of the first matrix and the signal intensity of each spatial position.

4. The signal modulation based, blood vessel selective arterial spin labeling encoding method of claim 1, wherein, The determining of the plurality of groups of candidate encoding mode combinations according to the blood vessels to be encoded comprises: determining a second matrix of a target order according to the number of blood vessels to be encoded; removing one column from the second matrix of the target order, randomly selecting a plurality of columns from the remaining columns, and determining a group of candidate encoding modes according to the randomly selected plurality of columns.

5. The signal modulation based blood vessel selective arterial spin labeling encoding method of claim 1, wherein, The selecting of the marker signal modulation map of each encoding mode in each group of candidate encoding mode combinations based on the spatial positions from the plurality of marker signal modulation maps comprises: for any one marker signal modulation map, taking out signal values at positions of blood vessels to be encoded; calculating a total value of the marker signal modulation map according to the signal values and marker values in the encoding mode; traversing all the marker signal modulation maps to find a marker signal modulation map with the highest total value.

6. The signal modulation based, blood vessel selective, arterial spin labeling encoding method of claim 1, wherein, A calculation formula of the loss value of the group of candidate encoding modes is as follows: wherein C is a condition number of a matrix; min(λ) is a shortest wavelength; motion is a preset motion displacement tolerance, and is used for controlling motion sensitivity.

7. The signal modulation based, blood vessel selective arterial spin labeling encoding method of claim 1, wherein, The encoding of the blood vessels to be encoded comprises: setting a blood vessel encoding gradient and a radio frequency phase of an imaging sequence according to the angle, the wavelength, and the offset to the magnet center corresponding to the encoding mode, so that a specific marker signal is generated at positions of the blood vessels to be encoded.

8. A signal modulation based blood vessel selective arterial spin labeling encoding device, characterized by, The method comprises the following steps: a generating module configured to generate a plurality of marker signal modulation maps on a marker plane; a selecting module configured to select, based on spatial positions, a marker signal modulation map of each encoding mode in each group of candidate encoding mode combinations from the plurality of marker signal modulation maps; An acquisition module is configured to acquire a number of blood vessels to be encoded and spatial positions of the blood vessels to be encoded on the labeling plane; A determination module is configured to determine a plurality of groups of candidate encoding mode combinations according to the blood vessels to be encoded, and select, from the plurality of labeling signal modulation maps, a labeling signal modulation map of each encoding mode in each group of candidate encoding mode combinations based on the spatial positions; An identification module is configured to identify a parameter combination and a blood vessel labeling signal value corresponding to the labeling signal modulation map of each encoding mode in each group of candidate encoding mode combinations, and construct a blood vessel labeling signal value matrix corresponding to each group of candidate encoding mode combinations, and calculate a loss value of each group of candidate encoding mode combinations according to the parameter combination and the blood vessel labeling signal value matrix; A selection module is configured to select a target encoding mode combination from the plurality of groups of candidate encoding mode combinations according to the loss value, and feed a parameter combination corresponding to each encoding mode in the target encoding mode combination to a magnetic resonance scanner, and the magnetic resonance scanner performs blood vessel encoding and scanning of the blood vessel labeling encoding of the blood vessels to be encoded based on the parameter combination corresponding to each encoding mode in the target encoding mode combination.

9. A medical device, characterized by Comprise: A memory, a processor and a computer program stored on the memory and executable on the processor, the processor executes the program to implement the signal modulation based blood vessel selection arterial spin labeling encoding method of any one of claims 1-7.

10. A computer readable storage medium having stored thereon a computer program or instructions, characterized in that, The computer program or instructions are executed to implement the signal modulation based blood vessel selection arterial spin labeling encoding method of any one of claims 1-7.

Images