Magnetic resonance image processing methods, devices and computer equipment

Abstract

This application relates to a magnetic resonance imaging method, apparatus, computer device, and storage medium. The method includes: acquiring a diffusion-weighted image group with multiple b-values; determining a region of interest (ROI) within the diffusion-weighted image group; performing diffusion relaxation coupling spectral imaging analysis on at least one voxel within the ROI; acquiring the coupling spectrum corresponding to the voxel; and determining the distribution of target components contained within the ROI based on the coupling spectrum. By acquiring the distribution and proportion of target components within each voxel, this method can estimate microstructural compartments with higher sensitivity and specificity, comprehensively and accurately reflecting the microstructural characteristics of tissues, and achieving multidimensional sub-voxel microstructural quantification. Compared to existing voxel-level analysis, this method provides more accurate quantitative analysis results.

Description

Technical Field

[0001] This application relates to the field of magnetic resonance imaging technology, and in particular to a magnetic resonance image processing method, apparatus and computer equipment. Background Technology

[0002] Magnetic resonance imaging (MRI) is an imaging technique that reconstructs images from signals generated by the resonance of atomic nuclei within a strong magnetic field. It features multi-parameter, multi-sequence, and multi-directional capabilities, offering high resolution for soft tissues, being non-invasive, and effectively displaying the structure and functional state of organs and tissues. Its imaging principle includes: exciting atomic nuclei with non-zero spin placed in a magnetic field using radio frequency pulses; relaxing the nuclei after the pulse stops; acquiring signals during relaxation using induction coils; and performing a series of processing steps, such as analog-to-digital conversion, amplification, filtering, and Fourier transform, to obtain the magnetic resonance image.

[0003] However, many important biological changes in living tissues (due to development, aging, injury, disease, scientific intervention, etc.) initially occur at the microscopic scale. The inherent macroscopic resolution of magnetic resonance imaging (MRI) limits the direct exploration of microscopic tissue characteristics. Due to the limited sensitivity of MRI, generating high-resolution MRI images within a reasonable timeframe is extremely challenging. Therefore, existing MRI-based methods utilize the fact that certain contrast mechanisms are sensitive to microstructures and combine them with appropriate mathematical models to perform quantitative imaging and indirectly infer tissue information at the microscopic scale. Quantitative magnetic resonance imaging (MRI) has two main branches: diffusion-weighted imaging (DWI) and MR relaxation measurement. DWI is a crucial branch of DWI, a powerful method for probing the microscopic structure of living biological tissues, and is widely used in disease detection and diagnosis, tumor benign / malignant differentiation, TMN (Tumor Node Metastasis Classification) staging, treatment efficacy evaluation, and prognosis assessment. Relaxation measurement techniques utilize the inherent sensitivity of magnetic resonance to the biochemical environment of tissues to quantify transverse relaxation time (T2 and T2*) and longitudinal relaxation time (T1), thereby further reflecting the internal physical and chemical microenvironment of tissues. The T1 relaxation process describes the recovery of longitudinal magnetization, while the T2 relaxation process describes the loss of transverse magnetization vector due to dephase effects caused by molecular-level interactions. Multiple studies have shown that quantification of T2 (T2 mapping) is fundamental for quantitatively measuring tissue properties in both healthy and diseased tissues, and this has been confirmed in various tissues, including joints, heart, and prostate.

[0004] While significant progress has been made in assessing tissue microstructure using quantitative information on diffusion and relaxation, existing methods still have some limitations: single T2 mapping or apparent diffusion coefficient (ADC) mapping only provides one-dimensional biological information and cannot comprehensively and accurately reflect the microstructural characteristics of tissues; T2 mapping and ADC mapping yield the apparent T2 and ADC values ​​of a voxel, but each voxel is composed of multiple microstructural compartments with different properties and interactions, and existing quantitative analysis methods cannot calculate the spatial distribution and proportion of different microstructural compartments. Therefore, it is necessary to improve existing quantitative analysis methods. Summary of the Invention

[0005] Therefore, it is necessary to provide a magnetic resonance image processing method, apparatus, and computer equipment to address the aforementioned technical problems and improve the accuracy of quantitative analysis.

[0006] A magnetic resonance image processing method, the method comprising:

[0007] Obtain diffusion-weighted image groups with multiple b values, where each b-value diffusion-weighted image group corresponds to multiple TI values ​​and / or multiple TE values;

[0008] Determine the region of interest in the diffusion-weighted image group;

[0009] At least one voxel in the region of interest is subjected to diffusion relaxation coupling spectral imaging analysis to obtain the coupling spectrum corresponding to the voxel.

[0010] Based on the coupling spectrum, the distribution of target components contained in the region of interest is determined.

[0011] In one embodiment, determining the region of interest in the diffusion-weighted image group includes:

[0012] A reference diffusion-weighted image is selected from the group of diffusion-weighted images with multiple b values;

[0013] Delineate the region of interest in the reference diffusion-weighted image; and,

[0014] The delineated region of interest is mapped to other diffusion-weighted images in the multi-b value diffusion-weighted image group.

[0015] In one embodiment, determining the distribution of target components contained in the region of interest based on the coupling spectrum includes:

[0016] Based on the coupling spectrum, determine the proportion of the target component contained in each voxel;

[0017] The proportion of the target component contained in each voxel is mapped to at least one of the multi-b value diffusion-weighted image groups, and the distribution of the target component contained in the region of interest is determined in the diffusion-weighted image groups.

[0018] In one embodiment, determining the proportion of the target component contained in each voxel based on the coupling spectrum includes:

[0019] Based on the prior quantitative parameter values, the coupled spectrum is divided into multiple regions, each region corresponding to a target component;

[0020] The two-dimensional distribution function corresponding to the coupling spectrum in each region is summed to determine the proportion of the target component contained in each voxel.

[0021] In one embodiment, the prior quantitative parameter value is determined as follows:

[0022] A quantitative parameter mapping map of the region of interest is reconstructed based on the diffusion-weighted image group with multiple b values.

[0023] The prior quantitative parameter values ​​are determined based on the quantitative mapping map of the region of interest.

[0024] In one embodiment, the prior quantitative parameter values ​​are the longitudinal relaxation time, the transverse relaxation time, and / or the apparent diffusion coefficient.

[0025] In one embodiment, the method further comprises:

[0026] The region of interest is rendered based on the distribution of target components contained therein.

[0027] A magnetic resonance imaging processing apparatus, the apparatus comprising:

[0028] The acquisition module is used to acquire diffusion-weighted image groups with multiple b values, where each b-value diffusion-weighted image group corresponds to multiple TI values ​​and / or multiple TE values;

[0029] A region of interest determination module is used to determine the region of interest in the diffusion-weighted image group;

[0030] The imaging analysis module is used to perform diffusion relaxation coupling spectrum imaging analysis on at least one voxel in the region of interest to obtain the coupling spectrum corresponding to the voxel.

[0031] The component distribution determination module is used to determine the target component distribution contained in the region of interest based on the coupled spectrum.

[0032] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program performing the following steps:

[0033] Obtain diffusion-weighted image groups with multiple b values, where each b-value diffusion-weighted image group corresponds to multiple TI values ​​and / or multiple TE values;

[0034] Determine the region of interest in the diffusion-weighted image group;

[0035] At least one voxel in the region of interest is subjected to diffusion relaxation coupling spectral imaging analysis to obtain the coupling spectrum corresponding to the voxel.

[0036] Based on the coupling spectrum, the distribution of target components contained in the region of interest is determined.

[0037] In one embodiment, the processor, when executing the computer program, further implements:

[0038] Diagnostic data are determined based on the distribution of target components contained within the region of interest; and

[0039] Output the determined diagnostic data.

[0040] The aforementioned magnetic resonance image processing method, apparatus, and computer equipment allow the computer equipment to acquire a multi-b-value diffusion-weighted image set, determine the region of interest (ROI) within the diffusion-weighted image set, perform diffusion relaxation coupling spectrum imaging analysis on at least one voxel within the ROI, acquire the coupling spectrum corresponding to the voxel, and determine the distribution of target components contained within the ROI based on the coupling spectrum. This method, by acquiring the distribution and proportion of target components within each voxel, can estimate microstructural compartments with higher sensitivity and specificity, comprehensively and accurately reflecting the microstructural characteristics of the tissue, and achieving multidimensional sub-voxel microstructural quantification. Compared to existing voxel-level analysis, this method provides more accurate quantitative analysis results. Attached Figure Description

[0041] Figure 1 This is an application environment diagram of a magnetic resonance image processing method in one embodiment;

[0042] Figure 2 This is a flowchart illustrating a magnetic resonance image processing method in one embodiment;

[0043] Figure 3 This is a two-dimensional spectrum carrying D and T2 in one embodiment;

[0044] Figure 4 This is a flowchart illustrating a method for determining a region of interest in another embodiment;

[0045] Figure 5 This is a flowchart illustrating a method for determining the distribution of target components within a region of interest in another embodiment.

[0046] Figure 5A This is a schematic diagram of a single voxel coupling spectrum divided into multiple regions in another embodiment;

[0047] Figure 6 This is a flowchart illustrating a method for determining the proportion of a target component contained in each voxel in another embodiment.

[0048] Figure 7 A spatial distribution and proportion of a component contained in a pituitary tumor in another embodiment;

[0049] Figure 8 A diagram showing the percentage composition of knee cartilage with varying degrees of arthritis in another embodiment;

[0050] Figure 9 This is a structural block diagram of a magnetic resonance image processing device in one embodiment;

[0051] Figure 10 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0053] The magnetic resonance image processing method provided in this application can be applied to Figure 1 In the application scenarios shown. For example... Figure 1As shown, the magnetic resonance image processing system includes a magnetic resonance scanning device and a computer device. In this embodiment, the magnetic resonance image processing method can be executed through the computer device. The magnetic resonance scanning device and the computer device can communicate via wired or wireless connection. The scanner of the magnetic resonance scanning device can transmit data to a host computer via optical fiber, then archive the data from the host computer to a hard drive, and then transfer it from the hard drive to the computer device for post-processing; alternatively, the data can be transmitted from the scanner of the magnetic resonance scanning device to a host computer and a post-processing workstation for direct processing at the workstation. Optionally, the wireless connection can be via Wi-Fi, mobile network, or Bluetooth, etc. The computer device can be an electronic device capable of performing image processing, such as a smartphone, tablet, laptop, desktop computer, or personal digital assistant. In this embodiment, the executing entity of the magnetic resonance image processing method can be a computer device. The specific process of the magnetic resonance image processing method will be described in the following embodiments.

[0054] In one embodiment, such as Figure 2 As shown, a magnetic resonance image processing method is provided, which can be applied to... Figure 1 Taking a computer device as an example, the explanation includes the following steps:

[0055] S100. Obtain a diffusion-weighted image group with multiple b values, where each b value corresponds to multiple inversion time (TI) values ​​and / or multiple echo time (TE) values.

[0056] Specifically, an MRI scanner can generate an MRI scan sequence based on an MRI scanning protocol. The scanner of the MRI scanner then scans the area to be scanned using this sequence to obtain MRI images. These images are then sent to a computer. The series of MRI images corresponding to the parameters b and TE of the MRI scan sequence constitutes a diffusion-weighted image set. The MRI scan sequence can be a spin echo sequence or a planar echo sequence, etc. The area to be scanned can be a part of the human body or an organism, such as the liver, prostate, uterus, brain, or abdominal organs, etc. In this embodiment, a diffusion-weighted image set with multiple b-values ​​is obtained by setting different b-value parameters.

[0057] In this embodiment, the aforementioned magnetic resonance scanning sequence can be a single-excitation spin-echo-planar imaging (SEPI) sequence, which is a combination of planar echo imaging and a spin-echo sequence. After a 90° radio frequency pulse excitation, a 180° pulse is applied, followed by continuous gradient oscillations to generate echo signals. The aforementioned magnetic resonance image can be a diffusion-weighted image. The aforementioned magnetic resonance scanning protocol can be a multi-b-value multi-TI value protocol, a multi-b-value multi-TE value protocol, a multi-b-value multi-TI value multi-TE value protocol, etc. The magnetic resonance device can obtain a diffusion-weighted image by scanning the scanned area using a single-excitation spin-echo-planar imaging sequence based on a multi-b-value multi-TI value protocol, a multi-b-value multi-TE value protocol, or a multi-b-value multi-TI value multi-TE value protocol. The imaging matrix or imaging angle of each diffusion-weighted image in the diffusion-weighted image group can be set to the same value.

[0058] For example, if the magnetic resonance imaging (MRI) scanning protocol is a multi-b-value, multi-TI-value protocol, then the MRI scanning protocol can acquire the first set of diffusion-weighted imaging (DWI) images based on a single TI-value, multi-b-value protocol. Further, different TI values ​​are set to generate multiple sets of TI-value multi-b-value protocols, and these multi-set TI-value multi-b-value protocols are executed. The multi-b-value, multi-TE-value protocol is similar to the multi-b-value, multi-TI-value protocol, and will not be described further. If the MRI scanning protocol is a multi-b-value, multi-TI-value, multi-TE-value protocol, then the MRI scanning protocol can acquire the first set of diffusion-weighted images based on a single TI-value, single TE-value multi-b-value protocol. Further, different TI values ​​and TE values ​​are set sequentially to generate multiple sets of TI-value and TE-value multi-b-value protocols, and these multi-set TI-value and TE-value multi-b-value protocols are executed.

[0059] Furthermore, the diffusion-weighted images corresponding to the multi-b-value multi-TI-value protocol, the multi-b-value multi-TE-value protocol, and the multi-b-value multi-TI-value multi-TE-value protocol are combined to obtain diffusion-weighted image groups corresponding to each protocol. For example... Figure 3 The diagram shows a diffusion-weighted image group under a multi-b-value and multi-TE-value protocol, i.e., a multi-b-value diffusion-weighted image group. Each circle in the diagram corresponds to a diffusion-weighted image under a single-b-value and single-TE-value protocol.

[0060] In magnetic resonance imaging (MRI), the aforementioned b-value can be understood as the diffusion-sensitive gradient field parameter applied in diffusion-weighted imaging, also known as the diffusion sensitivity coefficient. The aforementioned TI can appear in pulse sequences with a 180-degree inversion prepulse, including inversion recovery sequences, fast inversion recovery sequences, and inversion recovery planar echo sequences; the time interval from the midpoint of the 180-degree inversion prepulse to the midpoint of the 90-degree pulse is usually referred to as TI. The aforementioned TE can characterize the time interval from the midpoint of the pulse that generates the macroscopic transverse magnetization vector to the midpoint of the echo; in spin echo sequences, TE can represent the time interval from the midpoint of the 90-degree pulse to the midpoint of the spin echo; in gradient echo sequences, TE can represent the time interval from the midpoint of the small-angle pulse to the midpoint of the gradient echo.

[0061] S200. Determine the region of interest in the diffusion-weighted image group.

[0062] Specifically, after acquiring a diffusion-weighted image set, the computer equipment can determine the region of interest (ROI) of the area to be scanned. One diffusion-weighted image in the set can be selected as a reference image, and the ROI is determined on the reference image. Optionally, the reference image can be a diffusion-weighted image in the set where the ROI has a relatively high contrast with the surrounding tissue. The image corresponding to the ROI can be a partial image of any diffusion-weighted image in the set, any complete diffusion-weighted image, a subset of diffusion-weighted images in the set, or even all diffusion-weighted images in the set. Determining the ROI in a diffusion-weighted image set can be understood as determining the ROI of each diffusion-weighted image in the set; that is, mapping the ROI to each diffusion-weighted image in the set to obtain a set of ROI images.

[0063] In this embodiment, there are corresponding region of interest image groups under the multi-b-value multi-TI-value protocol, the multi-b-value multi-TE-value protocol, and the multi-b-value multi-TI-value multi-TE-value protocol.

[0064] S300. Perform diffusion relaxation coupling spectrum imaging analysis on at least one voxel in the region of interest to obtain the coupling spectrum corresponding to the voxel.

[0065] Specifically, the computer device can perform diffusion relaxation coupling spectral imaging analysis on at least one voxel in an image group of regions of interest under different protocols, obtaining the coupling spectrum corresponding to that voxel. The diffusion relaxation coupling spectral imaging analysis method can carry spatial location information. The coupling spectrum can contain coupling spectral information carrying spatial location information. The image corresponding to the region of interest can include multiple voxels; in this embodiment, each voxel in the image corresponding to the region of interest can be analyzed one by one.

[0066] Prior to step S300, the magnetic resonance image processing method may further include: reconstructing a diffusion relaxation coefficient map of the region of interest (ROI) based on a multi-b-value ROI image set, and determining the range of diffusion relaxation coefficient values ​​based on the diffusion relaxation coefficient map. The diffusion relaxation coefficient may include the apparent diffusion coefficient, lateral relaxation time, and longitudinal relaxation time. The reconstruction method may be an algebraic method, an iterative method, a Fourier back-projection method, a convolutional back-projection method, etc. In this embodiment, to improve the accuracy of the obtained diffusion relaxation coefficients, the diffusion relaxation coefficient map can be fitted using the least squares method.

[0067] The above diffusion relaxation coupling spectral imaging analysis can be understood as multi-component analysis, and the specific analysis process can be illustrated with an example. For instance, for each voxel in the region of interest image group, a voxel can include multiple components. Then, at specific TE and b values, the signal intensity m on the region of interest image in the region of interest image group can be expressed as:

[0068]

[0069] Where D represents the apparent diffusion coefficient, T2 represents the lateral relaxation time, I represents the total number of apparent diffusion coefficients within the range of apparent diffusion coefficient values, J represents the total number of lateral relaxation times within the range of lateral relaxation times, i represents the index associated with the apparent diffusion coefficient (1≤i≤I), j represents the index associated with the lateral relaxation time (1≤j≤J), x and y represent the two-dimensional position information of the region of interest (ROI) image in the ROI image group, and f i,j (x,y,D,T2) represents the two-dimensional distribution function corresponding to the i-th D and j-th T2 at positions x and y. Each region of interest (ROI) image in the region of interest image group has corresponding b-values ​​and TE-values. Therefore, the computer device can select f(x,y,D,T2) in the formula (1) for calculating the ROI image with multiple b-values ​​and multiple TE-values. f(x,y,D,T2) carries known two-dimensional position information. In this embodiment, diffusion relaxation coupling spectrum imaging analysis can obtain f(x,y,D,T2). Then, the computer device can process f(x,y,D,T2) to obtain a two-dimensional spectrum, i.e., a coupling spectrum. This two-dimensional spectrum carries all discrete D and T2 values, and any point in the two-dimensional spectrum has a corresponding f(x,y,D,T2) for both D and T2. Figure 3 The image shown is a visualized two-dimensional spectrum.

[0070] In addition, the computer equipment can perform diffusion relaxation coupling spectrum imaging analysis on image groups of regions of interest at specific TI and b values ​​to obtain coupling spectra corresponding to voxels. In this case, the coupling spectrum carries D and T1, where T1 is the longitudinal relaxation time. Simultaneously, the computer equipment can also perform diffusion relaxation coupling spectrum imaging analysis on image groups of regions of interest at specific TE, TI, and b values ​​to obtain coupling spectra corresponding to voxels. In this case, the coupling spectrum carries D, T1, and T2.

[0071] S400. Based on the coupling spectrum, determine the distribution of target components contained in the region of interest.

[0072] Understandably, computer equipment can preprocess the coupling spectra corresponding to each voxel in the image of the region of interest (ROI) to determine the distribution of target components contained within the ROI. Each voxel in the ROI can contain multiple components, such as water, fat, protein, N-acetylaspartic acid (NAA), choline (Cho), creatine (Cr), lactate (Lac), inositol (MI), glutamate (Glx), lipids (Lip), alanine (Ala), leucine (AAs), acetate (Ace), or succinate (SUCC), etc. The preprocessing described above can include normalization, component analysis, statistical processing, etc.

[0073] In the aforementioned magnetic resonance image processing method, the computer equipment can acquire a multi-b-value diffusion-weighted image group, determine the region of interest (ROI) within the diffusion-weighted image group, perform diffusion relaxation coupling spectrum imaging analysis on at least one voxel within the ROI, obtain the coupling spectrum corresponding to the voxel, and determine the distribution of target components contained within the ROI based on the coupling spectrum. This method can perform component analysis on voxels within the ROI using diffusion relaxation coupling spectrum imaging analysis that can carry spatial location information, thereby determining the distribution of target components contained within each voxel carrying location information and the proportion of target components. This can estimate microstructural compartments with higher sensitivity and specificity, comprehensively and accurately reflecting the microstructural characteristics of tissues, and achieving multidimensional sub-voxel microstructural quantification. Compared with existing voxel-level analysis, this makes the quantitative analysis results more accurate, and medical personnel can also improve the accuracy of disease diagnosis results through more accurate target component distribution, thereby solving specific clinical problems and improving the precision of clinical diagnosis.

[0074] In some scenarios, to improve the accuracy of disease diagnosis, the image corresponding to the region of interest in the imaging image can be determined first. Specifically, for example... Figure 4 As shown, the step of determining the region of interest in the diffusion-weighted image group in S200 above includes:

[0075] S201. Select a reference diffusion-weighted image from the group of diffusion-weighted images with multiple b values.

[0076] Specifically, the computer device can select one diffusion-weighted image as a reference diffusion-weighted image from the diffusion-weighted image group corresponding to the multi-b-value multi-TI-value protocol. It can also select any one or more diffusion-weighted images as reference diffusion-weighted images from the diffusion-weighted image group corresponding to the multi-b-value multi-TI-value multi-TE-value protocol. Of course, it can also select any one or more diffusion-weighted images as reference diffusion-weighted images from the diffusion-weighted image group corresponding to the multi-b-value multi-TI-value multi-TE-value protocol.

[0077] The reference diffusion-weighted image is the region of interest (ROI) in the diffusion-weighted image group that meets a set contrast condition, such as a higher contrast between the ROI and surrounding tissue. The ROI can be a tumor, and it needs to be delineated on an image where the contrast between the tumor and normal tissue is more pronounced. Generally, for tumors in the body, b=800 or b=1000. For tumors in the head, the values ​​are even higher. If the ROI is knee cartilage or the entire prostate, it doesn't matter if it's a tumor; you only need to find the image where the cartilage or prostate is most easily visible. For example, b=0 can be chosen.

[0078] In this embodiment, the selected reference diffusion-weighted image can be one of the diffusion-weighted images in a group of diffusion-weighted images. This diffusion-weighted image is the image corresponding to the minimum of multiple TI values ​​with a specific b-value, or the image corresponding to the minimum of multiple TE values ​​with a specific b-value. It should be noted that the specific b-value can be set to, for example, b=0, b=800, b=200, b=1600, etc. There are no specific restrictions on the selection of the b-value; it can be determined based on the image contrast or signal-to-noise ratio of the detection site. Similarly, there are no specific restrictions on the TI or TE values ​​corresponding to the reference image, as long as the contrast of the region of interest is higher than that of the surrounding tissue. For example, the reference diffusion-weighted image can be the image corresponding to the minimum of multiple TI values ​​and multiple TE values ​​with a low b-value. The aforementioned low b-value is determined relatively based on the specific disease to be diagnosed, and the determined specific b-value is based on the ability to distinguish the lesion at the diagnostic site.

[0079] S202. Delineate the region of interest in the reference diffusion-weighted image.

[0080] Specifically, medical staff can roughly determine the region of interest (ROI) in the reference diffusion-weighted image based on practical experience. The computer equipment can receive and respond to the delineation command, delineating the ROI in the reference diffusion-weighted image. Typically, the ROI has a higher signal intensity. The delineation command can carry the location information of the ROI in the reference diffusion-weighted image. The delineated shape can be rectangular, circular, elliptical, or other manually drawn shapes; there are no specific limitations.

[0081] In this embodiment, after the computer device delineates the region of interest in the reference diffusion-weighted image, it can obtain an image of the region of interest in the reference diffusion-weighted image.

[0082] S203. Map the delineated region of interest to other diffusion-weighted images in the multi-b-value diffusion-weighted image group.

[0083] Specifically, a computer device can map a region of interest (ROI) image onto other diffusion-weighted images in a multi-b-valued diffusion-weighted image set, excluding the reference diffusion-weighted image, to obtain the ROI for each diffusion-weighted image in the set. Mapping can be understood as an image overlay process, that is, the addition of image matrices. If the ROI image and the diffusion-weighted images are not equal, the ROI image can be padded with zeros to make it the same size as the diffusion-weighted images.

[0084] The above-mentioned magnetic resonance image processing method can select a reference diffusion-weighted image in a multi-b-value diffusion-weighted image group, delineate the region of interest in the reference diffusion-weighted image, and map the delineated region of interest to other diffusion-weighted images in the multi-b-value diffusion-weighted image group. This allows for accurate acquisition of the target component distribution of the region of interest in the diffusion-weighted image group, further solving specific clinical problems and improving the accuracy of clinical diagnosis.

[0085] As one example, such as Figure 5 As shown, the step in S400 above, which determines the distribution of target components within the region of interest based on the coupling spectrum, can be achieved through the following steps:

[0086] S401. Based on the coupling spectrum, determine the proportion of the target component contained in each voxel.

[0087] Specifically, the computer device can divide the coupling spectrum of a voxel into multiple regions based on the range of apparent diffusion coefficients, the range of transverse relaxation times, or the range of longitudinal relaxation times corresponding to different components in the voxel. Then, it can determine the corresponding two-dimensional distribution function f based on the diffusion relaxation coefficient of each region. 2DFor the two-dimensional distribution function f 2D The calculation process is performed to obtain the proportion of the target component contained in each voxel. This target component can be water, fat, protein, N-acetylaspartic acid (NAA), choline (Cho), creatine (Cr), lactate (Lac), inositol (MI), glutamate (Glx), lipids (Lip), alanine (Ala), leucine (AAs), acetate (Ace), or succinate (SUCC), etc., with no limitation on the specific component. The calculation process can include arithmetic operations such as summation and triggering, as well as exponential operations, logarithmic operations, and combinations thereof.

[0088] like Figure 5A The diagram shown illustrates the division of a single voxel coupling spectrum into multiple regions in one embodiment of this application. The horizontal axis represents the longitudinal relaxation time, and the vertical axis represents the apparent diffusion coefficient. Based on prior quantitative parameter values, the coupling spectrum can be divided into four regions, including AD, with each region corresponding to a target component.

[0089] The same processing described above can be performed on different voxels within the region of interest to obtain the proportion of the target component contained in each voxel. If the magnetic resonance protocol is a multi-b-value, multi-TI-value protocol, then the corresponding two-dimensional distribution function f 2D To carry information about D and T1; if the magnetic resonance protocol is a multi-b-valued, multi-TE-valued protocol, then the corresponding two-dimensional distribution function f 2D To carry information about D and T2; if the magnetic resonance protocol is a multi-b-value, multi-TI-value, and multi-TE-value protocol, then the corresponding three-dimensional distribution function f 3D To carry information about D, T1, and T2.

[0090] S402, Map the proportion of the target component contained in each voxel to at least one of the diffusion-weighted image groups with multiple b values, and determine the distribution of the target component contained in the region of interest in the diffusion-weighted image group.

[0091] Specifically, the computer device can map the proportion of the target component contained in each voxel to at least one diffusion-weighted image in a multi-b-value diffusion-weighted image set based on the two-dimensional position information of each voxel, thereby obtaining the distribution of the target component contained in the region of interest. At least one diffusion-weighted image may include a reference diffusion-weighted image. The target component distribution can be the specific spatial distribution of each component in each voxel. The target component distribution may include positional information and information on the proportions of different components.

[0092] The aforementioned magnetic resonance imaging processing method can determine the distribution of target components contained in the region of interest by acquiring the coupled spectrum. It can estimate microstructural compartments with higher sensitivity and specificity, comprehensively and accurately reflect the microstructural characteristics of the tissue, and achieve quantitative analysis of multidimensional subvoxel microstructure. Compared with existing voxel-level analysis, it makes the obtained target component distribution more accurate. Furthermore, medical staff can improve the accuracy of disease diagnosis results by using more accurate target component distribution to solve specific clinical problems.

[0093] As one embodiment, in order to obtain the distribution of the target component, the proportion of the target component can be determined first, such as... Figure 6 As shown, the step in S401 above, which determines the proportion of the target component contained in each voxel based on the coupling spectrum, can be achieved through the following steps:

[0094] S411. Based on the prior quantitative parameter values, the coupled spectrum is divided into multiple regions, each region corresponding to a target component.

[0095] Specifically, the aforementioned prior quantitative parameter values ​​can be determined from a database based on clinical experience. These prior quantitative parameter values ​​define the ranges of D, T1, and T2 for different components within each voxel. The computer can then divide the coupling spectrum into multiple regions based on these prior quantitative parameter values. The size of each region is arbitrary and not limited, but the size of each region can be smaller than the overall area of ​​the coupling spectrum.

[0096] S412. Sum the two-dimensional distribution functions corresponding to the coupling spectrum in each region to determine the proportion of the target component contained in each voxel.

[0097] Understandably, the computer device can perform calculations based on the two-dimensional distribution function of the coupling spectrum corresponding to all points in each divided region to obtain the proportion of the target component contained in each voxel. In this embodiment, this calculation can be a summation process. That is, the summation of the two-dimensional distribution function of the coupling spectrum corresponding to all points in each region equals the total content of the component corresponding to that region, i.e., the proportion of the component.

[0098] The aforementioned magnetic resonance imaging processing method can divide the coupled spectrum into multiple regions based on prior quantitative parameter values, with each region corresponding to a target component. The two-dimensional distribution function of the coupled spectrum within each region is summed to determine the proportion of the target component contained in each voxel. This allows for the determination of the target component distribution based on the proportion of the target component contained in each voxel at different spatial locations. This enables the estimation of microstructural compartments with higher sensitivity and specificity, comprehensively and accurately reflecting the microstructural characteristics of tissues and achieving multidimensional sub-voxel microstructural quantification. Compared to existing voxel-level analysis, this method yields a more accurate distribution of the target component. Furthermore, medical personnel can improve the accuracy of disease diagnosis results through more accurate target component distribution to address specific clinical problems.

[0099] As one embodiment, the aforementioned prior quantitative parameter values ​​can be determined as follows: a quantitative parameter mapping map of the region of interest is reconstructed based on a multi-b value diffusion-weighted image group, and the prior quantitative parameter values ​​are determined based on the quantitative mapping map of the region of interest.

[0100] Specifically, computer equipment can determine a group of regions of interest (ROI) images with multiple b-values ​​based on a group of diffusion-weighted images with multiple b-values. Then, it can reconstruct a diffusion relaxation coefficient map of the ROI, which is essentially a quantitative parameter mapping map of the ROI, using this group of ROI images. The quantitative parameters can be the apparent diffusion coefficient, lateral relaxation time, and longitudinal relaxation time. Furthermore, the corresponding prior quantitative parameter values ​​are determined by the range of values ​​for the quantitative parameters in the ROI quantitative parameter mapping map or by empirical values ​​from a historical database.

[0101] The prior quantitative parameter values ​​are the longitudinal relaxation time, the transverse relaxation time, and / or the apparent diffusion coefficient. In this embodiment, prior quantitative parameter values ​​can be determined for each voxel, and these values ​​can be the minimum / maximum values ​​of the longitudinal relaxation time, the transverse relaxation time, and / or the minimum / maximum values ​​of the apparent diffusion coefficient.

[0102] If the diffusion-weighted image group with multiple b values ​​is a diffusion-weighted image group with multiple b values ​​and multiple TE values, then the prior quantitative parameter value can be the value of longitudinal relaxation time and apparent diffusion coefficient; if the diffusion-weighted image group with multiple b values ​​is a diffusion-weighted image group with multiple b values ​​and multiple TI values, then the prior quantitative parameter value can be the value of transverse relaxation time and apparent diffusion; if the diffusion-weighted image group with multiple b values ​​is a diffusion-weighted image group with multiple b values, multiple TI values ​​and multiple TE values, then the prior quantitative parameter value can be the value of transverse relaxation time, longitudinal relaxation time and apparent diffusion coefficient.

[0103] In this embodiment, for a single voxel, its DWI signal intensity at a specific TE and b value is:

[0104]

[0105] Where D represents the apparent diffusion coefficient, T2 represents the lateral relaxation time, and f is a coefficient related to D and T2. Based on the scanned DWI image matrix, using the least squares fitting method, ADC mapping can be obtained from images acquired at multiple b-values ​​under the same TE; T2 mapping can be obtained from images acquired at multiple TEs under the same b-value. After obtaining the T2 mapping and ADC mapping of the imaging site, the maximum range of T2 and ADC for the imaging site can be determined. The maximum range of T2 and ADC for the imaging site are the prior quantitative parameter values.

[0106] This embodiment can determine the prior quantitative parameter values ​​corresponding to the longitudinal relaxation time, the transverse relaxation time, and / or the apparent diffusion coefficient, and then divide the coupled spectrum into different component regions based on the prior quantitative parameter values, thereby enabling the rapid acquisition of the proportion of different components in each voxel.

[0107] In addition, after all the above steps, the magnetic resonance image processing method may further include: rendering the region of interest according to the distribution of target components contained in the region of interest.

[0108] It is understandable that computer devices can render the region of interest in an image group by obtaining the distribution of target components contained in the region of interest, and obtain a spatial location distribution map of the region of interest.

[0109] If a diffusion-weighted image of the brain exists, the region of interest (ROI) can be rendered based on the distribution of target components contained within it to obtain a corresponding spatial distribution map, such as... Figure 7 The image shows the spatial distribution and proportion of a component in a pituitary tumor. The highlighted central area represents the proportion of that component, with different gray values ​​indicating different proportions. The components include water, fat, protein, N-acetylaspartic acid (NAA), choline (Cho), creatine (Cr), lactate (Lac), inositol (MI), glutamate (Glx), lipids (Lip), alanine (Ala), leucine (AAs), acetate (Ace), or succinate (SUCC). The spatial distribution and proportion of this component can provide a basis for clinical diagnosis.

[0110] This embodiment can render the region of interest (ROI) based on the distribution of target components contained within it, enabling medical staff to intuitively and quickly determine the specific lesion location from the ROI image and determine the severity of the patient's disease based on the signal intensity at the lesion location. This accelerates clinical diagnosis and assists medical staff in assessing the extent of the patient's lesions.

[0111] For example, in the case of arthritis in the joints of the human body, Figure 8 The images show the component proportions of knee cartilage at different degrees of arthritis severity. The first column corresponds to healthy subjects; the second to subjects with mild arthritis; and the third to subjects with moderate arthritis. For each column, from top to bottom, the images show the knee cartilage image, longitudinal relaxation time (T2 map), apparent diffusion coefficient (ADC map), and spatial distribution of component A (f...). A (map) diagram and spatial distribution of component B (f B The image also shows the voxel percentage under different quantitative parameters. In this embodiment, the voxel percentage comparison charts (frequency histograms) corresponding to the T2 map and the ADC map corresponding to three test subjects with different degrees of arthritis severity are also shown. A The map corresponds to a comparison chart of voxel percentages (frequency histogram) and f. B A comparison chart of the voxel percentages corresponding to the map (frequency histogram). From Figure 8 As shown in the second and third rows of images, the longitudinal relaxation time map, apparent diffusion coefficient map, and corresponding voxel percentage map of the femoral cartilage alone cannot distinguish the degree of femoral cartilage lesions, i.e., it cannot effectively differentiate between healthy, mild arthritis, and moderate arthritis. However, by using the diffusion relaxation coupling spectral imaging analysis method in this embodiment, spatial distribution maps of components A and B can be obtained (e.g., ...). Figure 8 (See the fourth and fifth rows of images). The images show significant differences in the levels of the two components among three different patient groups: healthy individuals, those with mild arthritis, and those with moderate arthritis. These differences also exist at different spatial locations. Furthermore, the frequency histograms reveal significant differences in the frequencies of the components in the femoral cartilage samples with varying degrees of lesions. The method described in this application significantly improves the diagnostic accuracy of knee arthritis and enables the grading and assessment of its severity.

[0112] It is understood that the embodiments of this application only use the diagnosis of knee osteoarthritis as an example for illustration, and the method shown in this embodiment is not limited to a specific location or specific parameters. In other embodiments, it can also be used for analysis of various parts of the body. Exemplarily, the method of the embodiments of this application can be applied to the detection and diagnosis of tumors, the differentiation between benign and malignant tumors, the grading and staging of malignant tumors, and the presence of metastasis in tumors. It can be applied to the evaluation of the efficacy of tumor treatment and prognosis, and can also be applied to the assessment of various geriatric diseases such as Oppenheimer's disease.

[0113] The above-mentioned magnetic resonance image processing method can determine the prior quantitative parameter values ​​corresponding to the longitudinal relaxation time, transverse relaxation time and / or apparent diffusion coefficient, and then divide the coupled spectrum into different component regions according to the prior quantitative parameter values ​​to quickly obtain the proportion of different components in each voxel.

[0114] To facilitate understanding by those skilled in the art, the magnetic resonance image processing method provided in this application is described using a computer device as an example. Specifically, the method includes:

[0115] (1) Obtain a diffusion-weighted image group with multiple b values, where each diffusion-weighted image group with a b value corresponds to multiple TI values ​​and / or multiple TE values;

[0116] (2) Select a reference diffusion-weighted image from the group of diffusion-weighted images with multiple b-values; wherein, the reference image is a diffusion-weighted image in the group of diffusion-weighted images where the region of interest has a different contrast (e.g., a significant difference) relative to the surrounding tissue; the reference diffusion-weighted image is a low b-value image and corresponds to the minimum of multiple TI values; or, the reference diffusion-weighted image is a low b-value image and corresponds to the minimum of multiple TE values;

[0117] (3) Delineate the region of interest in the reference diffusion-weighted image;

[0118] (4) Map the delineated region of interest to other diffusion-weighted images in the multi-b-value diffusion-weighted image group;

[0119] (5) Perform diffusion relaxation coupling spectrum imaging analysis on at least one voxel in the region of interest to obtain the coupling spectrum corresponding to the voxel;

[0120] (6) A quantitative parameter mapping map of the region of interest is reconstructed from the multi-b-valued diffusion-weighted image group;

[0121] (7) Determine the prior quantitative parameter values ​​based on the quantitative mapping map of the region of interest; the prior quantitative parameter values ​​are the longitudinal relaxation time, the transverse relaxation time and / or the apparent diffusion coefficient.

[0122] (8) Based on the prior quantitative parameter values, the coupled spectrum is divided into multiple regions, each region corresponding to a target component;

[0123] (9) Sum the two-dimensional distribution function of the coupling spectrum in each region to determine the proportion of the target component contained in each voxel;

[0124] (10) Map the proportion of the target component contained in each voxel to at least one of the multi-b-valued diffusion-weighted image groups, and determine the distribution of the target component contained in the region of interest in the diffusion-weighted image group.

[0125] The specific execution process of (1) to (10) above can be found in the description of the above embodiments. The implementation principle and technical effect are similar, and will not be repeated here.

[0126] It should be understood that, although Figure 2 , Figures 4-6 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figure 2 , Figures 4-6 At least some of the steps in the process may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but may be executed at different times. The execution order of these steps or stages is not necessarily sequential, but may be executed in turn or alternately with other steps or at least some of the steps or stages in other steps.

[0127] In one embodiment, such as Figure 9 As shown, a magnetic resonance imaging processing device is provided, comprising: an image group acquisition module 11, a region of interest determination module 12, an imaging analysis module 13, and a component distribution determination module 14, wherein:

[0128] Image group acquisition module 11 is used to acquire diffusion-weighted image groups with multiple b values, where each diffusion-weighted image group with a b value corresponds to multiple TI values ​​and / or multiple TE values;

[0129] Region of Interest (ROI) determination module 12 is used to determine the region of interest in a diffusion-weighted image group;

[0130] Imaging analysis module 13 is used to perform diffusion relaxation coupling spectrum imaging analysis on at least one voxel in the region of interest to obtain the coupling spectrum corresponding to the voxel.

[0131] The component distribution determination module 14 is used to determine the distribution of target components contained in the region of interest based on the coupled spectrum.

[0132] The magnetic resonance image processing device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, so it will not be described again here.

[0133] In one embodiment, the region of interest determination module 12 includes: a reference image selection unit, a region delineation unit, and a delineated region mapping unit, wherein:

[0134] A reference image selection unit is used to select a reference diffusion-weighted image from a group of diffusion-weighted images with multiple b values;

[0135] A region delineation unit is used to delineate the region of interest in a reference diffusion-weighted image; and,

[0136] The delineation region mapping unit is used to map the delineated region of interest to other diffusion-weighted images in a multi-b value diffusion-weighted image group.

[0137] The reference image is a diffusion-weighted image in the diffusion-weighted image group in which the region of interest shows a difference (e.g., a significant difference) in contrast relative to the surrounding tissue. For example, the reference diffusion-weighted image is a low b-value image that corresponds to the minimum of multiple TI values; or, the reference diffusion-weighted image is a low b-value image that corresponds to the minimum of multiple TE values.

[0138] The magnetic resonance image processing device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, so it will not be described again here.

[0139] In one embodiment, the component distribution determination module 14 includes: a component ratio determination unit and a component ratio mapping unit, wherein,

[0140] The component ratio determination unit is used to determine the proportion of the target component contained in each voxel based on the coupling spectrum.

[0141] A component scaling unit is used to map the proportion of the target component contained in each voxel to at least one of a multi-b-valued diffusion-weighted image group, and to determine the distribution of the target component contained in the region of interest in the diffusion-weighted image group.

[0142] The magnetic resonance image processing device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, so it will not be described again here.

[0143] In one embodiment, the component ratio determination unit includes: a component region division subunit and a calculation subunit, wherein:

[0144] The component region division subunit is used to divide the coupled spectrum into multiple regions based on prior quantitative parameter values, with each region corresponding to a target component;

[0145] The computational sub-unit is used to sum the two-dimensional distribution function of the coupling spectrum in each region to determine the proportion of the target component contained in each voxel.

[0146] The prior quantitative parameter values ​​are determined as follows: a quantitative parameter mapping map of the region of interest is reconstructed from a multi-b-value diffusion-weighted image group, and the prior quantitative parameter values ​​are determined based on the quantitative mapping map of the region of interest; the prior quantitative parameter values ​​are the values ​​of the longitudinal relaxation time, the lateral relaxation time, and / or the apparent diffusion coefficient.

[0147] The magnetic resonance image processing device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, so it will not be described again here.

[0148] In one embodiment, the magnetic resonance image processing apparatus further includes a rendering module, wherein:

[0149] The rendering module is used to render the region of interest based on the distribution of target components contained within it.

[0150] The magnetic resonance image processing device provided in this embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, so it will not be described again here.

[0151] Specific limitations regarding the magnetic resonance image processing apparatus can be found in the limitations of the magnetic resonance image processing method described above, and will not be repeated here. Each module in the aforementioned magnetic resonance image processing apparatus can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in hardware or independently of the processor in the computer device, or stored in software in the memory of the computer device, so that the processor can call and execute the operations corresponding to each module.

[0152] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 10As shown, the computer device includes a processor, memory, and network interface connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database stores images scanned by the magnetic resonance imaging (MRI) device under different protocols. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer program implements an MRI image processing method.

[0153] Those skilled in the art will understand that Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0154] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0155] Obtain diffusion-weighted image groups with multiple b values, where each b-value diffusion-weighted image group corresponds to multiple TI values ​​and / or multiple TE values;

[0156] Determine the region of interest in a diffusion-weighted image group;

[0157] Perform diffusion relaxation coupling spectrum imaging analysis on at least one voxel in the region of interest to obtain the coupling spectrum corresponding to the voxel;

[0158] Based on the coupling spectrum, determine the distribution of target components contained in the region of interest.

[0159] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0160] Diagnostic data are determined based on the distribution of target components contained within the region of interest; and

[0161] The diagnostic data is output as determined. For example, the diagnostic data may be the grading of osteoarthritis of the femoral cartilage, such as healthy, mild arthritis, and moderate arthritis; the diagnostic data may be benign or malignant tumors; the diagnostic data may be the grade, stage, and presence of metastasis of malignant tumors; the diagnostic data may also be an assessment of the efficacy of tumor treatment; or the diagnostic data may be an assessment of the defined disease condition.

[0162] In one embodiment, a storage medium is provided having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0163] Obtain diffusion-weighted image groups with multiple b values, where each b-value diffusion-weighted image group corresponds to multiple TI values ​​and / or multiple TE values;

[0164] Determine the region of interest in a diffusion-weighted image group;

[0165] Perform diffusion relaxation coupling spectrum imaging analysis on at least one voxel in the region of interest to obtain the coupling spectrum corresponding to the voxel;

[0166] Based on the coupling spectrum, determine the distribution of target components contained in the region of interest.

[0167] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0168] The technical features of the above embodiments can be combined in any way. For the sake of brevity, 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.

[0169] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A magnetic resonance image processing method, characterized in that, The method includes: Obtain diffusion-weighted image groups with multiple b values, where each b-value diffusion-weighted image group corresponds to multiple TI values ​​and / or multiple TE values; Determine the region of interest in the diffusion-weighted image group; At least one voxel in the region of interest is subjected to diffusion relaxation coupling spectral imaging analysis to obtain the coupling spectrum corresponding to the voxel. Based on the prior quantitative parameter values, the coupling spectrum is divided into multiple regions, each region corresponding to a target component; the prior quantitative parameter values ​​are the longitudinal relaxation time, the transverse relaxation time, and / or the apparent diffusion coefficient. The two-dimensional distribution function corresponding to the coupling spectrum in each region is summed to determine the proportion of the target component contained in each voxel; The proportion of the target component contained in each voxel is mapped to at least one of the multi-b value diffusion-weighted image groups, and the distribution of the target component contained in the region of interest is determined in the diffusion-weighted image groups.

2. The method according to claim 1, characterized in that, Determining the region of interest in the diffusion-weighted image group includes: A reference diffusion-weighted image is selected from the group of diffusion-weighted images with multiple b values; Delineate the region of interest in the reference diffusion-weighted image; and, The delineated region of interest is mapped to other diffusion-weighted images in the multi-b value diffusion-weighted image group.

3. The method according to claim 1, characterized in that, The prior quantitative parameter values ​​are determined in the following manner: A quantitative parameter mapping map of the region of interest is reconstructed based on the diffusion-weighted image group with multiple b values. The prior quantitative parameter values ​​are determined based on the quantitative mapping map of the region of interest.

4. The method according to any one of claims 1-3, characterized in that, The method further includes: The region of interest is rendered based on the distribution of target components contained therein.

5. A magnetic resonance image processing device, characterized in that, The device includes: The acquisition module is used to acquire diffusion-weighted image groups with multiple b values, where each b-value diffusion-weighted image group corresponds to multiple TI values ​​and / or multiple TE values; A region of interest determination module is used to determine the region of interest in the diffusion-weighted image group; The imaging analysis module is used to perform diffusion relaxation coupling spectrum imaging analysis on at least one voxel in the region of interest to obtain the coupling spectrum corresponding to the voxel. The component distribution determination module is used to divide the coupled spectrum into multiple regions based on prior quantitative parameter values, with each region corresponding to a target component. It then sums the two-dimensional distribution functions corresponding to the coupled spectra within each region to determine the proportion of the target component contained in each voxel, and maps the proportion of the target component contained in each voxel to at least one of the multi-b-valued diffusion-weighted image groups. Finally, it determines the target component distribution contained in the region of interest within the diffusion-weighted image group. The prior quantitative parameter values ​​are the values ​​of the longitudinal relaxation time, the lateral relaxation time, and / or the apparent diffusion coefficient.

6. The apparatus according to claim 5, characterized in that, The region of interest determination module includes: A reference image selection unit is used to select a reference diffusion-weighted image from the group of diffusion-weighted images with multiple b values; A region delineation unit is used to delineate the region of interest in the reference diffusion-weighted image; and, The delineation region mapping unit is used to map the delineated region of interest to other diffusion-weighted images in the multi-b-value diffusion-weighted image group.

7. The apparatus according to claim 5, characterized in that, The prior quantitative parameter values ​​are determined as follows: a quantitative parameter mapping map of the region of interest is reconstructed based on a multi-b value diffusion-weighted image group, and the prior quantitative parameter values ​​are determined based on the quantitative mapping map of the region of interest. The a priori quantitative parameter values ​​are the longitudinal relaxation time, the transverse relaxation time, and / or the apparent diffusion coefficient.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it achieves the following: Obtain diffusion-weighted image groups with multiple b values, where each b-value diffusion-weighted image group corresponds to multiple TI values ​​and / or multiple TE values; Determine the region of interest in the diffusion-weighted image group; At least one voxel in the region of interest is subjected to diffusion relaxation coupling spectral imaging analysis to obtain the coupling spectrum corresponding to the voxel. Based on the prior quantitative parameter values, the coupling spectrum is divided into multiple regions, each region corresponding to a target component; the prior quantitative parameter values ​​are the longitudinal relaxation time, the transverse relaxation time, and / or the apparent diffusion coefficient. The two-dimensional distribution function corresponding to the coupling spectrum in each region is summed to determine the proportion of the target component contained in each voxel; The proportion of the target component contained in each voxel is mapped to at least one of the multi-b value diffusion-weighted image groups, and the distribution of the target component contained in the region of interest is determined in the diffusion-weighted image groups.

9. The computer device according to claim 8, characterized in that, When the processor executes the computer program, it also implements: Diagnostic data are determined based on the distribution of target components contained within the region of interest; and Output the determined diagnostic data.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

Images