Medical image processing method, device and storage medium

Abstract

The present application relates to a medical image processing method, apparatus, and storage medium. The method comprises: acquiring multiple initial medical images of a tissue of interest via a pulse sequence; obtaining, based on the multiple initial medical images, a first initial parameter map including an initial blood flow velocity at each voxel point, and a second initial parameter map including an initial arterial transit time at each voxel point; optimizing the first initial parameter map and the second initial parameter map to obtain a first target parameter map corresponding to the first initial parameter map and a second target parameter map corresponding to the second initial parameter map, wherein the first target parameter map includes a target blood flow velocity at each voxel point, and the second target parameter map includes a target arterial transit time at each voxel point. Specifically, multiple target parameter maps are obtained by acquiring multiple initial parameter maps and iteratively optimizing each initial parameter map; determining multiple parameter maps corresponding to blood concentration data within different data ranges, and improving the accuracy of each parameter map.

Description

Technical Field

[0001] The present application relates to the field of big data processing technology, and in particular to a medical image processing method, device and storage medium. Background Art

[0002] Currently, the process of obtaining a brain MRI begins by labeling the hydrogen nuclei in the blood at the neck using a saturation or inversion pulse. After a preset period of time, a pulse sequence is used to generate a magnetic image of the brain. Finally, the data generated by the flow of labeled hydrogen nuclei into the brain are used to generate parameter maps corresponding to various data sets. These parameter maps are then used to assist doctors in diagnosing and analyzing stroke-related illnesses.

[0003] The existing technology calculates concentration data based on the grayscale value of each voxel point in the magnetic imaging, and then obtains the various data generated after the labeled hydrogen nuclei flow into the human brain through a table lookup method based on the correspondence between the preset concentration data and multiple data, and further obtains the parameter map corresponding to the multiple data.

[0004] However, the above method of obtaining the parameter map may cause the calculated concentration data to exceed the range recorded in the corresponding relationship and fail to determine the multiple data corresponding to the concentration data, thereby affecting the accuracy of the parameter map corresponding to the multiple data. Summary of the Invention

[0005] Based on this, it is necessary to provide a medical image processing method, device and storage medium that can obtain accurate blood parameter maps for blood concentrations within different ranges and improve the accuracy of the parameter maps in order to address the above technical problems.

[0006] In a first aspect, an embodiment of the present application provides a medical image processing method, the method comprising:

[0007] acquiring a plurality of initial medical images of the tissue of interest by a pulse sequence;

[0008] obtaining a first initial parameter map and a second initial parameter map based on the plurality of initial medical images, wherein the first initial parameter map includes an initial blood flow velocity at each voxel point, and the second initial parameter map includes an initial arterial transit time at each voxel point;

[0009] The first initial parameter map and the second initial parameter map are optimized to obtain a first target parameter map corresponding to the first initial parameter map and a second target parameter map corresponding to the second initial parameter map, wherein the first target parameter map includes a target blood flow velocity at each voxel point, and the second target parameter map includes a target arterial transit time at each voxel point.

[0010] In one embodiment, the method further comprises:

[0011] A third target parameter map is obtained according to the first target parameter map and the second target parameter map, and the third target parameter map includes a target blood volume of each voxel point.

[0012] In one embodiment, obtaining a first initial parameter map based on a plurality of initial medical images includes:

[0013] Inputting each initial medical image into the first model to obtain a first parameter map corresponding to each initial medical image;

[0014] The first parameter maps are fused to obtain a first initial parameter map.

[0015] In one embodiment, each initial medical image is input into a first model to obtain a first parameter map corresponding to each initial medical image, including:

[0016] For each initial medical image, a plurality of target data corresponding to the initial medical image is acquired, the plurality of target data including a distribution coefficient, a labeling efficiency, a first grayscale value, a second grayscale value, a third grayscale value, a first time, a second time, and a time difference; the first grayscale value is a grayscale value of the initial medical image excluding the target marker, the second grayscale value is an image grayscale value of the initial medical image including the target marker, the third grayscale value is a grayscale value of a voxel point in a proton density-weighted image including the target marker, the first time is a relaxation time of the target marker in the blood, the second time is a time between the end of a labeling pulse and the start of an acquisition pulse, and the time difference is a time difference between the start and end of labeling of the target marker in the blood;

[0017] Calculating the original blood flow velocity of each voxel point in the initial medical image according to a relationship formula containing multiple target data;

[0018] Based on the original blood flow velocity of each voxel point in the initial medical image and the corresponding initial medical image, a first parameter map corresponding to the initial medical image is obtained.

[0019] In one embodiment, the first parameter maps are fused to obtain a first initial parameter map, including:

[0020] Obtaining the original blood flow velocity of the corresponding voxel point in each first parameter map;

[0021] Calculating the average of the original blood flow velocity of the voxel points at the corresponding position of each first parameter map to obtain the average blood flow velocity of each voxel point;

[0022] A first initial parameter map is obtained based on the average blood flow velocity of each voxel point and any one of the multiple initial medical images.

[0023] In one embodiment, obtaining a first initial parameter map based on the average blood flow velocity of each voxel point and any one of the initial medical images includes:

[0024] The average blood flow velocity of each voxel point is marked on the corresponding voxel point of any initial medical image in the multiple initial medical images to obtain a first initial parameter map.

[0025] In one embodiment, obtaining a second initial parameter map based on each initial medical image includes:

[0026] A second initial parameter map is obtained based on the first initial medical image, the second initial medical image and the second model. The first initial medical image and the second initial medical image are two initial medical images among the multiple initial medical images that are continuously acquired using a pulse sequence.

[0027] In one embodiment, obtaining a second initial parameter map based on the first initial medical image, the second initial medical image, and the second model includes:

[0028] The first initial medical image and the second initial medical image are input into the second model to obtain a second initial parameter map.

[0029] In one embodiment, inputting the first initial medical image and the second initial medical image into the second model to obtain the second initial parameter map includes:

[0030] Acquiring a first acquisition time of a pulse corresponding to the first initial medical image and a second acquisition time of a pulse corresponding to the second initial medical image;

[0031] determining an initial arterial transit time based on the first acquisition time and the second acquisition time;

[0032] Obtaining a second initial parameter map according to the initial artery transit time and the first initial medical image; or,

[0033] A second initial parameter map is obtained according to the initial artery transit time and the second initial medical image.

[0034] In one embodiment, optimizing the first initial parameter map and the second initial parameter map to obtain a first target parameter map corresponding to the first initial parameter map and a second target parameter map corresponding to the second initial parameter map includes:

[0035] The first initial parameter map and the second initial parameter map are input into the optimizer for optimization. When the optimizer satisfies the optimization termination condition, a first target parameter map corresponding to the first initial parameter map and a second target parameter map corresponding to the second initial parameter map are output.

[0036] In one embodiment, the first initial parameter map and the second initial parameter map are input into the optimizer for optimization, and when the optimizer satisfies the optimization termination condition, the first target parameter map corresponding to the first initial parameter map and the second target parameter map corresponding to the second initial parameter map are output, including:

[0037] determining a magnetic flux change ratio based on an initial blood flow velocity at each voxel point in the first initial parameter map and an initial arterial transit time at each voxel point in the second initial parameter map;

[0038] Optimize the magnetic flux change ratio, and output the target magnetic flux change ratio when the optimization termination condition is met;

[0039] determining a target blood flow velocity at each voxel point and a target arterial transit time at each voxel point based on a target magnetic flux change ratio;

[0040] Based on the target blood flow velocity of each voxel point, the target arterial transit time of each voxel point, the first initial parameter map and the second initial parameter map, a first target parameter map and a second target parameter map are output.

[0041] In a second aspect, an embodiment of the present application further provides a medical image processing device, the device comprising:

[0042] an acquisition module, configured to acquire a plurality of initial medical images of a tissue of interest through a pulse sequence;

[0043] a first acquisition module, configured to obtain a first initial parameter map and a second initial parameter map based on the plurality of initial medical images, the first initial parameter map including an initial blood velocity at each voxel point, and the second initial parameter map including an initial arterial transit time at each voxel point;

[0044] The optimization module is used to optimize the first initial parameter map and the second initial parameter map to obtain a first target parameter map corresponding to the first initial parameter map and a second target parameter map corresponding to the second initial parameter map, wherein the first target parameter map includes a target blood flow velocity at each voxel point, and the second target parameter map includes a target arterial transit time at each voxel point.

[0045] In a third aspect, an embodiment of the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the methods in the first aspect when executing the computer program.

[0046] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of any one of the methods in the first aspect when executed by a processor.

[0047] The medical image processing method, apparatus, and storage medium described above acquire multiple initial medical images of a tissue of interest using a pulse sequence, and based on the multiple initial medical images, obtain a first initial parameter map including an initial blood velocity at each voxel point, and a second initial parameter map including an initial arterial transit time at each voxel point. Subsequently, the first initial parameter map and the second initial parameter map are optimized to obtain a first target parameter map corresponding to the first initial parameter map and a second target parameter map corresponding to the second initial parameter map. The first target parameter map includes a target blood velocity at each voxel point, and the second target parameter map includes a target arterial transit time at each voxel point. In other words, in the embodiment of the present application, multiple target parameter maps are obtained by acquiring multiple initial parameter maps and iteratively optimizing each of the initial parameter maps. This iterative optimization process does not involve table lookup to obtain the corresponding multiple data, and therefore does not involve limiting the data range in the table lookup operation. This avoids the situation where concentration data exceeds the range recorded in the table lookup correspondence relationship, making it impossible to determine the multiple data corresponding to the concentration data. This allows for the determination of multiple parameter maps corresponding to blood concentration data within different data ranges, and improves the accuracy of each parameter map. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is a flowchart of a medical image processing method in one embodiment;

[0049] Figure 2 is a flowchart of a medical image processing method in another embodiment;

[0050] Figure 3 Schematic diagram of the process of obtaining the first parameter map in one embodiment;

[0051] Figure 4 Schematic diagram of the process of obtaining the first initial parameter map in one embodiment;

[0052] Figure 5 1 is a schematic diagram of a process for obtaining a second initial parameter map in one embodiment;

[0053] Figure 6 1. It is a schematic diagram of a process for optimizing the first initial parameter map and the second initial parameter map in one embodiment;

[0054] Figure 7 is a structural block diagram of a medical image processing device in one embodiment;

[0055] Figure 8 is a structural block diagram of a medical image processing device in another embodiment;

[0056] Figure 9 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0058] The medical image processing method provided in the embodiments of the present application can be applied to a medical imaging device, or to a computer device that is communicatively connected to the medical imaging device, such as a terminal or a server. Optionally, the medical imaging device can be a perfusion imaging device, such as a magnetic resonance imaging (MR) device.

[0059] In one embodiment, Figure 1 As shown, a medical image processing method is provided, which is described by taking the application of the method to the above-mentioned medical imaging device or a computer device connected to the medical imaging device for example, and includes the following steps:

[0060] Step 101 : Acquire multiple initial medical images of a tissue of interest through a pulse sequence.

[0061] Among them, different initial medical images correspond to different PLD parameters. PLD (postlabeling delay) refers to the waiting time between the labeling sequence and the imaging sequence. In other words, it is the waiting time from the end of the labeling pulse to the start of ASL perfusion image acquisition. During this period, the labeled blood enters the imaging area of ​​the tissue of interest. Different PLD parameters result in different blood labeling conditions in the acquired initial medical images.

[0062] Optionally, multiple different PLD parameters can be obtained, and for each PLD parameter, multiple initial medical images of the tissue of interest are acquired through a pulse sequence; optionally, the multiple PLD parameters can be multiple default PLD parameters, or can be multiple PLD parameters input by the user before the medical image scan, and this application does not make any specific limitations on this.

[0063] Step 102 : obtaining a first initial parameter map and a second initial parameter map based on a plurality of initial medical images.

[0064] The first initial parameter map includes the initial blood flow velocity of each voxel point, and the second initial parameter map includes the initial arterial transit time of each voxel point; for example, the first initial parameter map may be an initial CBF parameter map, and the second initial parameter map may be an initial ATT parameter map.

[0065] Optionally, after obtaining multiple initial medical images corresponding to different PLD parameters, a first processing operation can be performed on the multiple initial medical images to obtain the first initial parameter map. The first processing operation is an operation related to determining the blood flow velocity. For example, the first processing operation can be a first preset algorithm for determining the initial blood flow velocity of each voxel point. By inputting the multiple initial medical images into the first preset algorithm, the first initial parameter map is output.

[0066] Optionally, a second processing operation may be performed on the multiple initial medical images to obtain the second initial parameter map. The second processing operation is an operation related to determining the arterial transit time. For example, the second processing operation may be a second preset algorithm for determining the initial arterial transit time of each voxel point. By inputting the multiple initial medical images into the second preset algorithm, the second initial parameter map is output.

[0067] It should be noted that, in practical applications, the second initial parameter map may also be pre-set, and the pre-set second initial parameter map may be based on historical empirical values. Of course, the above-mentioned second processing operation may also be used to process multiple initial medical images to obtain an estimated value.

[0068] Step 103 : Optimize the first initial parameter map and the second initial parameter map to obtain a first target parameter map corresponding to the first initial parameter map and a second target parameter map corresponding to the second initial parameter map.

[0069] The first target parameter map includes a target blood flow velocity at each voxel point, and the second target parameter map includes a target arterial transit time at each voxel point.

[0070] Optionally, the first initial parameter map can be input into the first optimization model, and multiple iterative optimization operations can be performed until a termination condition is met. The parameter map obtained from the last optimization is used as the first target parameter map. The termination condition can be the number of optimizations or a judgment condition related to the first target parameter map. Based on the above example, the first target parameter map can be a CBF parameter map.

[0071] Similarly, the second initial parameter map can be input into the second optimization model, and multiple iterative optimization operations can be performed until a termination condition is met. The parameter map obtained from the last optimization is used as the second target parameter map. The termination condition can be the number of optimizations or a judgment condition related to the second target parameter map. Based on the above example, the second target parameter map can be the ATT parameter map.

[0072] In the above-described medical image processing method, multiple initial medical images of a tissue of interest are acquired using a pulse sequence. Based on the multiple initial medical images, a first initial parameter map including an initial blood velocity at each voxel and a second initial parameter map including an initial arterial transit time at each voxel are obtained. Subsequently, the first initial parameter map and the second initial parameter map are optimized to obtain a first target parameter map corresponding to the first initial parameter map and a second target parameter map corresponding to the second initial parameter map. The first target parameter map includes a target blood velocity at each voxel, and the second target parameter map includes a target arterial transit time at each voxel. In other words, in the embodiment of the present application, multiple target parameter maps are obtained by acquiring multiple initial parameter maps and iteratively optimizing each of the initial parameter maps. This iterative optimization process does not involve table lookup to obtain the corresponding multiple data, and therefore does not involve limiting the data range in the table lookup. This avoids the situation where concentration data exceeds the range recorded in the table lookup correspondence relationship, making it impossible to determine the multiple data corresponding to the concentration data. This allows for the determination of multiple parameter maps corresponding to blood concentration data within different data ranges, and improves the accuracy of each parameter map.

[0073] In an optional embodiment of the present application, after obtaining the above-mentioned first target parameter map including the target blood flow velocity of each voxel point and the second target parameter map including the target arterial transit time of each voxel point, a third target parameter map can also be obtained based on the first target parameter map and the second target parameter map, wherein the third target parameter map includes the target blood volume of each voxel point.

[0074] Based on the above example, when the first target parameter map is a CBF parameter map and the second target parameter map is an ATT parameter map, the third target parameter map may be a CBV parameter map.

[0075] Optionally, a hemodynamic model can be established based on relevant principles of hemodynamics, and then a third target parameter map can be determined based on the first target parameter map, the second target parameter map, and the hemodynamic model. Since the principle and implementation process of determining the CBV parameter map based on the CBF parameter map and the ATT parameter map are existing technologies, they will not be explained in detail here.

[0076] Figure 2 FIG. 1 is a flow chart of a medical image processing method in another embodiment. This embodiment involves an optional implementation process of obtaining a first initial parameter map based on multiple initial medical images. Based on the above embodiment, Figure 2 As shown, the above step 102 includes:

[0077] In step 201 , each initial medical image is input into a first model to obtain a first parameter map corresponding to each initial medical image.

[0078] Optionally, the first model can be a mathematical model or an image processing model, such as a neural network model, a deep learning model, a machine learning model, etc. The form of the first model is not specifically limited in the embodiments of the present application. Through the preset first model, a first parameter map corresponding to the initial medical image can be obtained based on the initial medical image.

[0079] The following describes in detail the first model as a data model. Figure 3 As shown, the following steps are included:

[0080] In step 301 , for each initial medical image, a plurality of target data corresponding to the initial medical image may be obtained.

[0081] The multiple target data may include a distribution coefficient, labeling efficiency, a first grayscale value, a second grayscale value, a third grayscale value, a first time, a second time, and a time difference. The first grayscale value is the grayscale value of the initial medical image excluding the target marker, the second grayscale value is the image grayscale value of the initial medical image including the target marker, and the third grayscale value is the grayscale value of a voxel in a proton density-weighted image including the target marker. The first time is the relaxation time of the target marker in the blood, the second time is the time between the end of the labeling pulse and the start of the acquisition pulse, i.e., the time corresponding to the PLD parameter, and the time difference is the time difference between the start and end of labeling of the target marker in the blood. It should be noted that the target markers described above are markers in the tissue of interest after pulse labeling.

[0082] It should also be noted that the initial medical image refers to the medical image collected corresponding to a PLD parameter, which may include one or more different types of medical images. The above-mentioned target data can be determined by at least one type of initial medical image collected corresponding to the PLD parameter.

[0083] Step 302 : Calculate the original blood flow velocity of each voxel point in the initial medical image according to a relationship formula containing multiple target data.

[0084] Optionally, the following formula (1) may be used to calculate the original blood flow velocity of each voxel point in the initial medical image.

[0085]

[0086] Where CBF is the original blood flow velocity of the voxel point, λ is the brain-blood partition coefficient, which is usually simplified to 0.9 mL / g, and SI control is the time-averaged signal intensity in the control image, i.e., the first grayscale value in the initial medical image excluding the target marker, SI labelis the time-averaged signal intensity in the labeled image, i.e., the second grayscale value in the initial medical image including the target marker, T 1,blood is the first time, i.e., the longitudinal relaxation time of the target marker in the blood, in seconds; PLD is the PLD time corresponding to the initial medical image, i.e., the second time between the end of the labeling pulse and the acquisition pulse of the initial medical image; α is the labeling efficiency; SI PD is the signal intensity in the proton density-weighted image, that is, the third grayscale value of the voxel point in the proton density-weighted image of the target marker. τ is the labeling duration, that is, the time difference between the start and end labeling of the target marker in the blood. The constant 6000 is used to convert the CBF unit to mL / (100 g) / min commonly used in clinical practice.

[0087] It should be noted that the at least one type of initial medical image acquired under a PLD parameter mentioned in step 301 may include a control image, a labeled image, and a proton density weighted image acquired under the PLD parameter.

[0088] Step 303 : obtaining a first parameter map corresponding to the initial medical image based on the original blood flow velocity of each voxel point in the initial medical image and the corresponding initial medical image.

[0089] Optionally, the original blood flow velocity of each voxel point in the initial medical image may be marked on the corresponding voxel point in the initial medical image to obtain a first parameter map corresponding to the initial medical image.

[0090] Step 202: Fusing the first parameter maps to obtain a first initial parameter map.

[0091] Optionally, for the first parameter maps corresponding to each initial medical image, the average blood flow velocity of the corresponding voxel point in each first parameter map can be used as the blood flow velocity of the voxel point in the first initial parameter map; the median blood flow velocity of the corresponding voxel point in each first parameter map can be used as the blood flow velocity of the voxel point in the first initial parameter map; the maximum / minimum blood flow velocity of the corresponding voxel point in each first parameter map can also be used as the blood flow velocity of the voxel point in the first initial parameter map; and the first parameter maps can be fused to obtain the first initial parameter map. The fusion method is not specifically limited in the embodiments of the present application.

[0092] In this embodiment, each initial medical image is input into the first model respectively to obtain a first parameter map corresponding to each initial medical image. Then, each first parameter map is fused to obtain a first initial parameter map; so that the first initial parameter map can more accurately express the blood flow velocity characteristics of the tissue of interest, and thus the first target parameter map obtained by optimizing the first initial parameter map is more accurate, that is, the accuracy of the blood flow velocity parameter map is improved.

[0093] Figure 4 FIG. 1 is a flow chart of a medical image processing method in another embodiment. This embodiment involves fusing the first parameter maps to obtain an optional implementation process of the first initial parameter map. Based on the above embodiment, Figure 4 As shown, the above step 202 includes:

[0094] Step 401: Obtain the original blood flow velocity of the corresponding voxel point in each first parameter map.

[0095] Step 402 : Calculate the average value of the original blood flow velocity of the voxel points at the corresponding position of each first parameter map to obtain the average blood flow velocity of each voxel point.

[0096] Optionally, the average value of the original blood flow velocities of each voxel point at the corresponding position of each first parameter map can be directly calculated to obtain the average blood flow velocity; or the original blood flow velocities of each voxel point at the corresponding position of each first parameter map can be first screened, and the average blood flow velocity can be calculated based on the original blood flow velocities of the voxel points in the multiple screened first parameter maps; optionally, the maximum blood flow velocity and the minimum blood flow velocity in the original blood flow velocities of each voxel point at the corresponding position of each first parameter map can be eliminated, that is, after removing the maximum value and the minimum value, the average value of the remaining original blood flow velocities is calculated to eliminate interference.

[0097] Step 403 : Obtain a first initial parameter map based on the average blood flow velocity of each voxel point and any one of the multiple initial medical images.

[0098] Optionally, the average blood flow velocity of each voxel point may be marked on a corresponding voxel point of any initial medical image in a plurality of initial medical images to obtain a first initial parameter map.

[0099] In this embodiment, the average blood flow velocity of each voxel point is obtained by obtaining the original blood flow velocity of the corresponding voxel point in each first parameter map and calculating the average value of the original blood flow velocity of the voxel point at the corresponding position of each first parameter map; then, based on the average blood flow velocity of each voxel point and any initial medical image of the multiple initial medical images, the first initial parameter map is obtained, which can improve the matching degree between the first initial parameter map and the final first target parameter map, improve the optimization efficiency, and thus improve the accuracy of the first target parameter map.

[0100] In an optional embodiment of the present application, the implementation of "obtaining a second initial parameter map based on each initial medical image" in the above-mentioned step 102 may include: obtaining a second initial parameter map based on the first initial medical image, the second initial medical image and the second model, wherein the first initial medical image and the second initial medical image are two initial medical images continuously acquired using a pulse sequence from multiple initial medical images; that is, two adjacent initial medical images can be arbitrarily selected from the multiple initial medical images as the first initial medical image and the second initial medical image.

[0101] Optionally, the first initial medical image and the second initial medical image can be input into a second model to obtain a second initial parameter map. The second model can be a mathematical model or an image processing model, such as a neural network model, a deep learning model, or a machine learning model. The embodiment of this application does not specifically limit the form of the second model. Using the preset second model, a second initial parameter map representing arterial transit time can be obtained based on the first and second initial medical images.

[0102] In an optional implementation of this embodiment, as Figure 5 As shown, the second initial parameter map can also be obtained by the following steps, including:

[0103] Step 501 : Acquire a first acquisition time of a pulse corresponding to a first initial medical image and a second acquisition time of a pulse corresponding to a second initial medical image.

[0104] Step 502: Determine an initial arterial transit time based on the first acquisition time and the second acquisition time.

[0105] That is, the time difference between the first acquisition time of the adjacent first initial medical image and the second acquisition time of the second initial medical image is the initial artery transit time.

[0106] Step 503 : Obtain a second initial parameter map based on the initial artery transit time and the first initial medical image; or obtain a second initial parameter map based on the initial artery transit time and the second initial medical image.

[0107] Optionally, the initial arterial transit time may be marked on each voxel point in the first initial medical image to obtain a second initial parameter map; or the initial arterial transit time may be marked on each voxel point in the second initial medical image to obtain a second initial parameter map.

[0108] Based on this embodiment, another implementation method may be included, namely, obtaining a first initial medical image and a second initial medical image of multiple adjacent pulses as multiple initial medical image pairs; determining the arterial transit time corresponding to each initial medical image pair for each initial medical image pair; then, the average arterial transit time of the arterial transit times corresponding to each initial medical image pair may be used as the initial arterial transit time; finally, a second initial parameter map is obtained based on the initial arterial transit time and any initial medical image; wherein the any initial medical image may be any one of the multiple initial medical image pairs.

[0109] In this embodiment, a first acquisition time of a pulse corresponding to a first initial medical image and a second acquisition time of a pulse corresponding to a second initial medical image are obtained; an initial arterial transit time is determined based on the first and second acquisition times; a second initial parameter map is then obtained based on the initial arterial transit time and the first initial medical image; or a second initial parameter map is obtained based on the initial arterial transit time and the second initial medical image. The second initial parameter map obtained by estimating the initial medical image can more accurately represent the pulse transit time of the tissue of interest and has higher accuracy than that set using empirical values. Furthermore, the second target parameter map obtained by optimizing the second initial parameter map with higher accuracy has higher accuracy, improves optimization efficiency, and enables rapid convergence of the optimization process.

[0110] In an optional embodiment of the present application, step 103 may be implemented by inputting the first initial parameter map and the second initial parameter map into an optimizer for optimization. When the optimizer satisfies a termination optimization condition, the optimizer outputs a first target parameter map corresponding to the first initial parameter map and a second target parameter map corresponding to the second initial parameter map. Optionally, the optimizer's termination optimization condition may be the number of optimization attempts or a termination optimization condition related to the accuracy of the first target parameter map and / or the second target parameter map.

[0111] In an optional implementation of this embodiment, as Figure 6 As shown, the first initial parameter map and the second initial parameter map can be optimized by the following steps to obtain the first target parameter map and the second target parameter map, including:

[0112] Step 601 : determining a magnetic flux change ratio according to the initial blood flow velocity of each voxel point in the first initial parameter map and the initial arterial transit time of each voxel point in the second initial parameter map.

[0113] The magnetic flux change ratio is the ratio of the difference between the first grayscale value and the second grayscale value to the third grayscale value.

[0114] Optionally, the initial blood flow velocity at each voxel point in the first initial parameter map and the initial arterial transit time at each voxel point in the second initial parameter map can be input into a preset formula to calculate the magnetic flux change ratio. The preset formula can be a formula derived by inversely calculating the blood flow velocity and arterial transit time based on the magnetic flux change ratio.

[0115] Step 602 , optimizing the magnetic flux change ratio, and outputting the target magnetic flux change ratio when the optimization termination condition is met.

[0116] Optionally, the magnetic flux change ratio can be optimized using the least linear squares method to obtain the optimized magnetic flux change ratio. Then, the first initial parameter map and the second initial parameter map can be adjusted according to the optimized magnetic flux change ratio, that is, the first intermediate parameter map and the second intermediate parameter map after the first optimization process are obtained. If the optimization termination condition is not met at this time, the magnetic flux change ratio can be re-determined according to the first intermediate parameter map and the second intermediate parameter map after the first optimization process. Then, the first intermediate parameter map and the second intermediate parameter map after the first optimization process are optimized and adjusted for a second time according to the re-determined magnetic flux change ratio, to obtain the first intermediate parameter map and the second intermediate parameter map after the second optimization process. The optimization operation is performed in this cycle until the optimization termination condition is met, and the target magnetic flux change ratio obtained by the last optimization process is output.

[0117] Step 603 : Determine the target blood flow velocity and the target artery transit time of each voxel point based on the target magnetic flux change ratio.

[0118] Optionally, the first and second intermediate parameter maps in the optimization process are adjusted based on the target magnetic flux change ratio to calculate the optimized target blood flow velocity and target arterial transit time for each voxel. Optionally, the target magnetic flux change ratio can be substituted into the formula for calculating blood flow velocity and arterial transit time based on the magnetic flux change ratio to calculate the optimized target blood flow velocity and target arterial transit time for each voxel.

[0119] Step 604 : outputting a first target parameter map and a second target parameter map based on the target blood flow velocity of each voxel point, the target artery transit time of each voxel point, the first initial parameter map, and the second initial parameter map.

[0120] Optionally, the initial blood flow velocity of each voxel point in the first initial parameter map can be replaced with the target blood flow velocity of each voxel point to obtain the first target parameter map; and the initial arterial transit time of each voxel point in the second initial parameter map can be replaced with the target arterial transit time of each voxel point to obtain the second target parameter map.

[0121] In this embodiment, the magnetic flux change ratio is determined based on the initial blood flow velocity of each voxel point in the first parameter map and the initial arterial transit time of each voxel point in the second parameter map; the magnetic flux change ratio is optimized, and when the optimization termination condition is met, the target magnetic flux change ratio is output; then, the target blood flow velocity of each voxel point and the target arterial transit time of each voxel point are determined based on the target magnetic flux change ratio, and the first target parameter map and the second target parameter map are output based on the target blood flow velocity of each voxel point, the target arterial transit time of each voxel point, the first parameter map and the second parameter map; through this optimization process, the optimization efficiency and the accuracy of the optimization results can be improved, thereby improving the accuracy of the first target parameter map and the second target parameter map.

[0122] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0123] Based on the same inventive concept, embodiments of the present application also provide a medical image processing device for implementing the aforementioned medical image processing method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more medical image processing device embodiments provided below can be found in the above-described limitations of the medical image processing method and will not be further elaborated here.

[0124] In one embodiment, Figure 7As shown, a medical image processing device is provided, including: an acquisition module 701, a first acquisition module 702 and an optimization module 703, wherein:

[0125] An acquisition module 701 is configured to acquire a plurality of initial medical images of a tissue of interest through a pulse sequence;

[0126] A first acquisition module 702 is configured to obtain a first initial parameter map and a second initial parameter map based on the multiple initial medical images, wherein the first initial parameter map includes an initial blood velocity at each voxel point, and the second initial parameter map includes an initial arterial transit time at each voxel point;

[0127] The optimization module 703 is used to optimize the first initial parameter map and the second initial parameter map to obtain a first target parameter map corresponding to the first initial parameter map and a second target parameter map corresponding to the second initial parameter map, wherein the first target parameter map includes a target blood flow velocity for each voxel point, and the second target parameter map includes a target arterial transit time for each voxel point.

[0128] In one embodiment, Figure 8 As shown, the device further includes a second acquisition module 704; the second acquisition module 704 is used to obtain a third target parameter map according to the first target parameter map and the second target parameter map, and the third target parameter map includes the target blood volume of each voxel point.

[0129] In one embodiment, the first acquisition module 702 includes a first acquisition unit and a second acquisition unit; wherein the first acquisition unit is used to input each initial medical image into the first model respectively to obtain a first parameter map corresponding to each initial medical image; the second acquisition unit is used to fuse each first parameter map to obtain a first initial parameter map.

[0130] In one embodiment, the first acquisition unit is specifically configured to acquire, for each initial medical image, a plurality of target data corresponding to the initial medical image, and calculate the original blood flow velocity of each voxel in the initial medical image according to a relationship formula containing the plurality of target data; and obtain a first parameter map corresponding to the initial medical image based on the original blood flow velocity of each voxel in the initial medical image and the corresponding initial medical image; wherein the plurality of target data include a distribution coefficient, a labeling efficiency, a first grayscale value, a second grayscale value, a third grayscale value, a first time, a second time, and a time difference; the first grayscale value is the grayscale value of the initial medical image excluding the target marker, the second grayscale value is the image grayscale value of the initial medical image including the target marker, the third grayscale value is the grayscale value of the voxel in the proton density weighted image including the target marker, the first time is the relaxation time of the target marker in the blood, the second time is the time between the end of the labeling pulse and the start of the acquisition pulse, and the time difference is the time difference between the start labeling and the end labeling of the target marker in the blood.

[0131] In one embodiment, the second acquisition unit is specifically used to obtain the original blood flow velocity of the corresponding voxel points in each first parameter map; calculate the average value of the original blood flow velocity of the voxel points at the corresponding position of each first parameter map to obtain the average blood flow velocity of each voxel point; and obtain the first initial parameter map based on the average blood flow velocity of each voxel point and any one of the multiple initial medical images.

[0132] In one embodiment, the second acquisition unit is specifically configured to mark the average blood flow velocity of each voxel point on a corresponding voxel point of any initial medical image among the multiple initial medical images to obtain a first initial parameter map.

[0133] In one embodiment, the above-mentioned first acquisition module 702 also includes a third acquisition unit; the third acquisition unit is used to obtain a second initial parameter map based on the first initial medical image, the second initial medical image and the second model, and the first initial medical image and the second initial medical image are two initial medical images continuously acquired using a pulse sequence among multiple initial medical images.

[0134] In one embodiment, the third acquisition unit is specifically configured to input the first initial medical image and the second initial medical image into the second model to obtain a second initial parameter map.

[0135] In one embodiment, the third acquisition unit is specifically configured to acquire a first acquisition time of a pulse corresponding to the first initial medical image and a second acquisition time of a pulse corresponding to the second initial medical image; determine an initial arterial transit time based on the first acquisition time and the second acquisition time; obtain a second initial parameter map based on the initial arterial transit time and the first initial medical image; or obtain a second initial parameter map based on the initial arterial transit time and the second initial medical image.

[0136] In one embodiment, the above-mentioned optimization module 703 is specifically used to input the first initial parameter graph and the second initial parameter graph into the optimizer for optimization, and when the optimizer meets the optimization termination condition, output the first target parameter graph corresponding to the first initial parameter graph and the second target parameter graph corresponding to the second initial parameter graph.

[0137] In one embodiment, the above-mentioned optimization module 703 is specifically used to determine the magnetic flux change ratio based on the initial blood flow velocity of each voxel point in the first initial parameter map and the initial arterial transit time of each voxel point in the second initial parameter map; optimize the magnetic flux change ratio and output the target magnetic flux change ratio when the optimization termination condition is met; determine the target blood flow velocity of each voxel point and the target arterial transit time of each voxel point based on the target magnetic flux change ratio; and output the first target parameter map and the second target parameter map based on the target blood flow velocity of each voxel point, the target arterial transit time of each voxel point, the first initial parameter map and the second initial parameter map.

[0138] Each module in the aforementioned medical image processing apparatus may be implemented in whole or in part through software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0139] In one embodiment, a computer device is provided. The computer device may be a terminal or a server, or a processing device integrated into a medical imaging device. The internal structure diagram thereof may be as follows: Figure 9As shown. The computer device includes a processor, a memory and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store multiple initial medical images of the collected tissue of interest, multiple parameter maps after processing, a first model, a second model, and at least one data in the optimizer. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a medical image processing method is implemented.

[0140] Those skilled in the art will understand that Figure 9 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0141] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps of the medical image processing method in any of the above embodiments when executing the computer program.

[0142] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the medical image processing method in any of the above embodiments are implemented.

[0143] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0144] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0145] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0146] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A medical image processing method, characterized in that: The method comprises: acquiring a plurality of initial medical images of the tissue of interest by a pulse sequence; obtaining a first initial parameter map and a second initial parameter map based on the multiple initial medical images, wherein the first initial parameter map includes an initial blood flow velocity at each voxel point, and the second initial parameter map includes an initial arterial transit time at each voxel point; determining a magnetic flux change ratio according to an initial blood flow velocity at each voxel point in the first initial parameter map and an initial arterial transit time at each voxel point in the second initial parameter map; Optimizing the magnetic flux change ratio, and outputting a target magnetic flux change ratio if a termination optimization condition is met; determining a target blood flow velocity at each voxel point and a target artery transit time at each voxel point based on the target magnetic flux change ratio; Based on the target blood flow velocity of each voxel point, the target arterial transit time of each voxel point, the first initial parameter map and the second initial parameter map, a first target parameter map and a second target parameter map are output, wherein the first target parameter map includes the target blood flow velocity of each voxel point, and the second target parameter map includes the target arterial transit time of each voxel point.

2. The method according to claim 1, characterized in that The method further comprises: A third target parameter map is obtained according to the first target parameter map and the second target parameter map, wherein the third target parameter map includes the target blood volume of each voxel point.

3. The method according to claim 1, characterized in that The obtaining of a first initial parameter map based on the multiple initial medical images comprises: Inputting each of the initial medical images into a first model to obtain a first parameter map corresponding to each of the multiple initial medical images; The first parameter maps are fused to obtain the first initial parameter map.

4. The method according to claim 3, characterized in that Inputting each of the initial medical images into the first model to obtain a first parameter map corresponding to each of the initial medical images includes: For each of the initial medical images, a plurality of target data corresponding to the initial medical image is acquired, the plurality of target data including a distribution coefficient, a labeling efficiency, a first grayscale value, a second grayscale value, a third grayscale value, a first time, a second time, and a time difference; the first grayscale value is the grayscale value of the initial medical image excluding the target marker, the second grayscale value is the image grayscale value of the initial medical image including the target marker, the third grayscale value is the grayscale value of a voxel point in a proton density-weighted image including the target marker, the first time is the relaxation time of the target marker in the blood, the second time is the time between the end of the labeling pulse and the start of the acquisition pulse, and the time difference is the time difference between the start and end labeling of the target marker in the blood; calculating the original blood flow velocity of each voxel point in the initial medical image according to a relationship formula including the plurality of target data; Based on the original blood flow velocity of each voxel point in the initial medical image and the corresponding initial medical image, a first parameter map corresponding to the initial medical image is obtained.

5. The method according to claim 4, characterized in that The fusing of the first parameter maps to obtain the first initial parameter map includes: Obtaining the original blood flow velocity of the corresponding voxel point in each of the first parameter maps; Calculating the average of the original blood flow velocities of the voxel points at the corresponding positions of the first parameter maps to obtain the average blood flow velocity of each voxel point; The first initial parameter map is obtained based on the average blood flow velocity of each voxel point and any initial medical image of the multiple initial medical images.

6. The method according to claim 5, characterized in that The obtaining of the first initial parameter map based on the average blood flow velocity of each voxel point and any one of the initial medical images includes: The average blood flow velocity of each voxel point is marked on a corresponding voxel point of any initial medical image in the multiple initial medical images to obtain the first initial parameter map.

7. The method according to claim 1, characterized in that The obtaining of a second initial parameter map based on the multiple initial medical images comprises: The second initial parameter map is obtained based on the first initial medical image, the second initial medical image and the second model, where the first initial medical image and the second initial medical image are two initial medical images among the multiple initial medical images that are continuously acquired using the pulse sequence.

8. The method according to claim 7, characterized in that The obtaining the second initial parameter map based on the first initial medical image, the second initial medical image, and the second model includes: The first initial medical image and the second initial medical image are input into the second model to obtain the second initial parameter map.

9. The method according to claim 8, characterized in that Inputting the first initial medical image and the second initial medical image into the second model to obtain the second initial parameter map includes: Acquire a first acquisition time of a pulse corresponding to the first initial medical image and a second acquisition time of a pulse corresponding to the second initial medical image; determining an initial arterial transit time based on the first acquisition time and the second acquisition time; obtaining the second initial parameter map according to the initial artery transit time and the first initial medical image; or, The second initial parameter map is obtained according to the initial artery transit time and the second initial medical image.

10. A medical image processing device, characterized in that: The device comprises: an acquisition module, configured to acquire a plurality of initial medical images of a tissue of interest through a pulse sequence; a first acquisition module, configured to obtain a first initial parameter map and a second initial parameter map based on the multiple initial medical images, wherein the first initial parameter map includes an initial blood velocity at each voxel point, and the second initial parameter map includes an initial arterial transit time at each voxel point; An optimization module is configured to determine a magnetic flux change ratio based on an initial blood flow velocity at each voxel point in the first initial parameter map and an initial arterial transit time at each voxel point in the second initial parameter map; optimize the magnetic flux change ratio, and output a target magnetic flux change ratio if the optimization termination condition is satisfied; determine a target blood flow velocity at each voxel point and a target arterial transit time at each voxel point based on the target magnetic flux change ratio; and output a first target parameter map and a second target parameter map based on the target blood flow velocity at each voxel point, the target arterial transit time at each voxel point, the first initial parameter map, and the second initial parameter map, wherein the first target parameter map includes the target blood flow velocity at each voxel point and the second target parameter map includes the target arterial transit time at each voxel point.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

Images