Medical image processing method, device, equipment and storage medium

By dividing PET scan data into multiple groups and constructing a guiding kernel matrix for filtering, the problem of noise affecting spatial information in existing technologies is solved, achieving low-noise and high-spatial-information PET image reconstruction and improving the clinical application effect of low-dose PET imaging.

CN112614201BActive Publication Date: 2024-11-12SHENZHEN INST OF ADVANCED TECH
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202011519164.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-12-21
Publication Date
2024-11-12
Estimated Expiration
2040-12-21

AI Technical Summary

Technical Problem

Current technology cannot reduce the noise level of PET images without affecting the spatial information they carry, leading to a decline in PET image quality and increasing the risk of missed diagnoses and misdiagnoses.

Method used

Based on the principle of time equalization, the PET scan data is divided into at least two groups. The composite frame image of each group is determined and a guide kernel matrix is ​​constructed. The guide kernel filter is used to perform guide kernel filtering on each frame image to supplement spatial information and generate the target PET image.

Benefits of technology

While reducing noise levels, the spatial information of PET images is maintained or improved, thus enhancing the image reconstruction quality of low-dose PET imaging technology and promoting its clinical application.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112614201B_ABST
    Figure CN112614201B_ABST
Patent Text Reader

Abstract

The embodiment of the present invention discloses a medical image processing method, device, equipment and storage medium, the method comprising: dividing the PET scan data of the target object into at least two groups according to the time averaging principle; determining the composite frame image corresponding to each group of PET scan data to obtain at least two composite frame images; determining the guide kernel matrix corresponding to the at least two composite frame images; using the guide kernel matrix to perform guide kernel filtering on each frame image corresponding to the PET scan data to obtain the target PET image of the corresponding frame. The problem that the prior art cannot reduce the noise level of the PET image without affecting the spatial information carried by the PET image is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of image processing, and in particular to a medical image processing method, apparatus, device and storage medium. Background Art

[0002] Positron Emission Computed Tomography (PET) is a functional imaging instrument and a relatively advanced clinical imaging technology in the field of nuclear medicine. PET imaging technology is currently the only technology that uses anatomical forms to perform functional, metabolic and receptor imaging. It is non-invasive and can provide three-dimensional and functional images of the whole body. Before PET imaging of patients, they need to be injected with radioactive drugs (tracers) with a short half-life. Since the distribution of tracers changes rapidly, patients must be scanned at a higher scanning speed. However, a higher scanning speed is prone to insufficient counts when the tracer dose is low, and insufficient counts will reduce the image quality of PET images, such as high noise levels. When the noise level of PET images is high, high noise will mask important microlesions, and the boundaries of organs and tissues will become blurred, which will increase the chances of missed diagnosis and misdiagnosis.

[0003] Therefore, it is necessary to reduce the noise of PET images. However, existing noise reduction methods not only reduce the noise level of PET images, but also reduce the spatial information carried by the PET images. That is, the existing technology cannot reduce the noise level of PET images without affecting the spatial information carried by the PET images. Summary of the invention

[0004] The embodiments of the present invention provide a medical image processing method, apparatus, device and storage medium, which solve the problem that the prior art cannot reduce the noise level of a PET image without affecting the spatial information carried by the image.

[0005] In a first aspect, an embodiment of the present invention provides a medical image processing method, comprising:

[0006] Dividing the PET scan data of the target object into at least two groups according to the time averaging principle;

[0007] Determine a composite frame image corresponding to each set of PET scan data to obtain at least two composite frame images;

[0008] Determine a guide kernel matrix corresponding to the at least two composite frame images;

[0009] The guided kernel matrix is ​​used to perform guided kernel filtering on each frame image corresponding to the PET scan data to obtain a target PET image of the corresponding frame.

[0010] In a second aspect, an embodiment of the present invention further provides a medical image processing device, comprising:

[0011] A grouping module, used for dividing the PET scan data of the target object into at least two groups according to a time averaging principle;

[0012] An image reconstruction module, used for determining a composite frame image corresponding to each set of PET scan data to obtain at least two composite frame images;

[0013] A matrix determination module, used to determine the guiding kernel matrix corresponding to the at least two composite frame images;

[0014] The filtering module is used to perform guided kernel filtering on each frame image corresponding to the PET scanning data using the guided kernel matrix to obtain a target PET image of the corresponding frame.

[0015] In a third aspect, an embodiment of the present invention further provides an image processing device, the image processing device comprising:

[0016] one or more processors;

[0017] A storage device for storing one or more programs;

[0018] When the one or more programs are executed by the one or more processors, the one or more processors implement the medical image processing method as described in any embodiment.

[0019] In a fourth aspect, an embodiment of the present invention further provides a storage medium comprising computer executable instructions, which, when executed by a computer processor, are used to execute the medical image processing method as described in any embodiment.

[0020] The technical solution of the medical image processing method provided by the embodiment of the present invention divides the PET scanning data of the target object into at least two groups according to the time averaging principle; determines the composite frame image corresponding to each group of PET scanning data to obtain at least two composite frame images, so that the composite frame image carries spatial information; determines the guiding kernel matrix corresponding to the at least two composite frame images, so that the guiding kernel matrix carries the spatial information carried by all the composite frame images; uses the guiding kernel matrix to perform guided kernel filtering on each frame image corresponding to the PET scanning data to obtain the target PET image of the corresponding frame, thereby filtering each frame image by the guiding kernel matrix, and supplementing the spatial information of the corresponding frame image by the guiding kernel matrix during the filtering process, so that the target PET image has a lower noise level and richer spatial information, thereby improving the image reconstruction quality of the scanning data collected by the low-dose PET imaging technology, and thereby improving the promotion speed of the low-dose PET imaging technology in clinical practice. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 is a flowchart of a medical image processing method provided by Embodiment 1 of the present invention;

[0023] Figure 2 is a schematic diagram of a medical image processing method provided by Embodiment 1 of the present invention;

[0024] Figure 3 is a structural block diagram of a medical image processing device provided by Embodiment 2 of the present invention;

[0025] Figure 4 It is a structural block diagram of an image processing device provided in Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described through implementation methods with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] Embodiment 1

[0028] Figure 1 It is a flow chart of the medical image processing method provided in the first embodiment of the present invention. The technical solution of this embodiment is applicable to determining a guiding kernel matrix carrying spatial information through at least two composite frame images, so that when filtering each frame image of the PET image data, the guiding kernel matrix also supplements the spatial information of the corresponding frame image, so that the target PET image generated by filtering has both a low noise level and rich spatial information. The method can be executed by the medical image processing device provided in the embodiment of the present invention, and the device can be implemented in software and / or hardware, and configured in the processor of the image processing device for application. The method specifically includes the following steps:

[0029] S101 . Divide the PET scan data of the target object into at least two groups according to a time averaging principle.

[0030] The so-called time averaging principle means that the scanning time is evenly divided into N time periods according to the time-activity curve, and the PET scanning data corresponding to each time period is taken as a group, where N is a natural number greater than or equal to 2.

[0031] For example, Figure 2 As shown, the PET imaging device scanned the target object 24 times within one hour, that is, generated scanning data of 24 frames of PET images. If the current time averaging principle is to divide the scanning time into 3 time periods, then the scanning time of this imaging scan can be divided into: the first time period (the first 20 minutes), the second time period (the middle 20 minutes) and the third time period (the last 20 minutes). After verification, it was found that the PET imaging device scanned 16 times in the first time period, 4 times in the second time period, and 4 times in the third time period, then the scanning data of the PET imaging device, that is, the PET scanning data, can be divided into: a first group of PET scanning data including 1-16 frames of images, a second group of PET scanning data including 17-20 frames of images, and a third group of PET scanning data including 21-24 frames of images.

[0032] It should be noted that the PET scanning data in this embodiment is data obtained by performing attenuation correction processing on the original scanning data collected by the PET imaging device.

[0033] S102: Determine the composite frame image corresponding to each set of PET scan data to obtain at least two composite frame images.

[0034] Image reconstruction is performed on each set of PET scan data to obtain a composite frame image corresponding to each set of PET scan data, thereby obtaining at least two composite frame images. Figure 2 As shown, the PET scan data of the 1-16 frames are reconstructed to obtain the first composite frame image, the PET scan data of the 17-20 frames are reconstructed to obtain the second composite frame image, and the PET scan data of the 21-24 frames are reconstructed to obtain the third composite frame image. It can be understood that each composite frame image carries the information of all the image frames of its corresponding group, especially the spatial information.

[0035] The image reconstruction algorithm is preferably but not limited to the MLEM (Maximum Likelihood Expectation Maximization, MLEM for short) algorithm, that is, Figure 2 The EM in is as follows:

[0036]

[0037] Where y is the PET scan data, which can be considered as a collection of independent Poisson random variables; P is the system matrix; r represents background events such as random noise and scattering; n is the total number of pixels in the image, M is the number of detector pairs contained in the PET imaging device, and T M is a unit vector of length M.

[0038] In this embodiment, the PET scan data is preferably divided into three groups, and accordingly, the PET scan data corresponds to three composite frame images. Dividing the PET scan data into three groups can make the reconstructed three composite frame images contain more spatial information and more image detail information.

[0039] S103: Determine a guiding kernel matrix corresponding to the at least two composite frame images.

[0040] This embodiment preferably determines the guiding kernel matrix corresponding to the at least two composite frame images based on the k-nearest neighbor algorithm. Specifically, a window is created, suitable neighbor data is searched in the window, such as the nearest neighbor data to construct a kernel matrix, and the matrix is ​​normalized to obtain a normalized kernel matrix, and the kernel matrix is ​​preferably a sparse matrix to improve the speed of image filtering processing using the kernel matrix.

[0041] It can be understood that, since each image used to reconstruct the composite frame image carries spatial information, each composite frame image carries rich spatial information, and therefore the guiding kernel matrix determined based on at least two composite frame images also carries rich spatial information. Since the spatial information of the PET image comes from the pixel brightness information of the PET image, or in other words, the spatial information of the PET image is determined by the pixel brightness information of the PET image, the guiding kernel matrix of this embodiment carries the pixel brightness information of the at least two composite frame images.

[0042] In one embodiment, before determining the guiding kernel matrix corresponding to the at least two composite frame images, the at least two composite frame images are respectively subjected to a preset filtering process to update each composite frame image, wherein the preset filtering process is preferably but not limited to a Gaussian filtering process.

[0043] S104 , using the guided kernel matrix to perform guided kernel filtering on each frame of image corresponding to the PET scan data, so as to obtain a target PET image of the corresponding frame.

[0044] After the guide kernel matrix is ​​determined, the scan data of the current frame image is read from the PET scan data, and the read scan data is reconstructed to obtain the current frame image; the preset filtering times are obtained, and the current frame image is guided kernel filtered using the guide kernel matrix to obtain an intermediate PET image; when it is detected that an intermediate PET image is generated, it is determined whether the current filtering times is greater than 0; if so, the intermediate PET image is converted into corresponding scan data, and the scan data is reconstructed to update the current frame image, and the current frame image is guided kernel filtered using the guide kernel matrix to update the intermediate PET image; if not, the intermediate PET image is used as the target PET image.

[0045] Among them, this embodiment preferably uses the MLEM algorithm to reconstruct the scan data to obtain the current frame image.

[0046] The preset number of filtering times is a preset number of filtering times. In one embodiment, the preset number of filtering times can be used for guiding kernel filtering processing of each frame image corresponding to PET scanning data obtained by scanning different scanning parts of the target object with different models of PET imaging devices under different scanning conditions. In one embodiment, the preset number of filtering times can be used for guiding kernel filtering processing of each frame image corresponding to PET scanning data obtained by scanning the target object with the same or similar scanning conditions with different models of PET imaging devices. The scanning conditions include the number of scanning times and the time interval between two adjacent scans.

[0047] Among them, the method for determining the preset number of filtering times includes the following steps: reading the scanning data of the current frame image from the test PET scanning data, and performing image reconstruction on the read scanning data to obtain the current frame image; using the guiding kernel matrix corresponding to the test PET scanning data to perform guided kernel filtering on the current frame image to obtain an intermediate PET image; when it is detected that an intermediate PET image is generated, determining whether the current test number is greater than 0; if so, recording the preset image quality parameters and the corresponding filtering times of the intermediate PET image; and converting the intermediate PET image into the corresponding scanning data, and performing image reconstruction on the scanning data to update the current frame image, and using the guiding kernel matrix to perform guided kernel filtering on the current frame image to obtain the intermediate PET image; if not, determining the lowest value among all the recorded preset image quality parameters, and using the filtering times corresponding to the lowest value as the preset filtering times.

[0048] It is understandable that the number of tests needs to be greater than the preset number of filters. If all the recorded preset image quality parameters show a downward trend, and the value of the preset image quality parameter extracted last time is the lowest, it can be inferred that the currently set number of tests is less than or equal to the preset number of filters, and in a high probability, the currently set number of tests is less than the preset number of filters. At this time, it is necessary to reset the number of tests and use the above-mentioned method for determining the preset number of filters to re-determine the preset number of filters.

[0049] In some embodiments, the intermediate PET image is projected to a preset projection direction to obtain a corresponding projection image; then the preset image quality parameters of the projection image are used as the preset image quality parameters of the intermediate PET image. Since the data volume of the two-dimensional projection image is much smaller than the data volume of the three-dimensional intermediate PET image, the data volume of the preset image quality parameters corresponding to the two-dimensional projection image is also smaller than the data volume of the preset image quality parameters corresponding to the three-dimensional intermediate PET image. Therefore, when the image processing speed of the image processing device cannot meet the analysis requirements of the preset image quality parameters of the three-dimensional intermediate PET image, the preset image quality parameters of the projection image can be used as the preset image quality parameters of the intermediate PET image. Among them, the preset projection direction is preferably but not limited to the forward projection direction.

[0050] In some embodiments, see Figure 2 , read the scan data of the current frame image from the PET scan data, and reconstruct the scan data to obtain the current frame image; use the guide kernel matrix to perform guided kernel filtering on the current frame image to obtain an intermediate PET image, and determine the projection image of the intermediate PET image in a preset projection direction; when it is detected that a projection image is generated, determine whether the preset image quality parameters of the projection image meet the preset conditions; if so, use the projected image or the intermediate PET image as the target PET image, and output the target PET image; if not, determine the scan data corresponding to the intermediate PET image, and reconstruct the scan data to update the current frame image, and use the guide kernel matrix to perform guided kernel filtering on the current frame image to update the intermediate PET image, and determine the projection image of the updated intermediate PET image in the preset projection direction. When guided kernel filtering is performed on each frame image, it is detected whether the preset image quality parameters of the intermediate PET image meet the preset conditions, and it is suitable for images corresponding to PET scan data of any part of any target object collected by any PET imaging device under any imaging conditions, and has good universality, but it can be understood that in actual use, especially in clinical application, the implementation of this embodiment depends on image processing equipment with superior performance.

[0051] Of course, in order to increase the speed of detecting whether the preset image quality parameters of the intermediate PET image meet the preset conditions each time the kernel filtering is guided, the intermediate PET image generated each time can be projected to a preset projection direction to obtain a corresponding projection image, and then the preset image quality parameters of the projection image can be detected to see whether they meet the preset conditions, and the result can be used as the evaluation result of the preset image quality parameters of the corresponding intermediate PET image.

[0052] It can be understood that the scanning data corresponding to at least two composite frame images used to determine the guiding kernel matrix in this embodiment are derived from the current PET scanning data, so the guiding kernel matrix does not have the problem of mismatching with any frame image. Compared with the guiding kernel matrix determined according to scanning data of other modalities in the prior art, the guiding kernel matrix of this embodiment can supplement accurate spatial information for each frame image while ensuring the filtering effect of each frame image, so as to improve the reconstruction filtering effect of each frame image, thereby improving the image quality of the target PET image.

[0053] The technical solution of the medical image processing method provided by the embodiment of the present invention divides the PET scanning data of the target object into at least two groups according to the time averaging principle; determines the composite frame image corresponding to each group of PET scanning data to obtain at least two composite frame images, so that the composite frame image carries spatial information; determines the guiding kernel matrix corresponding to the at least two composite frame images, so that the guiding kernel matrix carries the spatial information carried by all the composite frame images; uses the guiding kernel matrix to perform guided kernel filtering on each frame image corresponding to the PET scanning data to obtain the target PET image of the corresponding frame, thereby filtering each frame image by the guiding kernel matrix, and supplementing the spatial information of the corresponding frame image by the guiding kernel matrix during the filtering process, so that the target PET image has a lower noise level and richer spatial information, thereby improving the image reconstruction quality of the scanning data collected by the low-dose PET imaging technology, and thereby improving the promotion speed of the low-dose PET imaging technology in clinical practice.

[0054] Embodiment 2

[0055] Figure 3 : is a block diagram of a medical image processing device provided by an embodiment of the present invention. The device is used to execute the medical image processing method provided by any of the above embodiments, and the device can be implemented in software or hardware. The device includes:

[0056] A grouping module 11, used to divide the PET scan data of the target object into at least two groups according to a time averaging principle;

[0057] An image reconstruction module 12, used for determining a composite frame image corresponding to each set of PET scan data to obtain at least two composite frame images;

[0058] A matrix determination module 13, used to determine the guiding kernel matrix corresponding to the at least two composite frame images;

[0059] The filtering module 14 is used to perform guided kernel filtering on each frame image corresponding to the PET scan data using a guided kernel matrix to obtain a target PET image of the corresponding frame.

[0060] Optionally, the matrix determination module is used to determine the guiding kernel matrix corresponding to the at least two composite frame images based on a k-nearest neighbor algorithm; wherein the guiding kernel matrix includes pixel brightness information of the at least two composite frame images.

[0061] Optionally, the filtering module is used to read the scanning data of the current frame image from the PET scanning data, and perform image reconstruction on the scanning data to obtain the current frame image; obtain a preset filtering number, and use a guiding kernel matrix to perform guided kernel filtering on the current frame image to obtain an intermediate PET image; when it is detected that an intermediate PET image is generated, determine whether the current filtering number is greater than 0; if so, convert the intermediate PET image into corresponding scanning data, and perform image reconstruction on the scanning data to update the current frame image, and use the guiding kernel matrix to perform guided kernel filtering on the current frame image to update the intermediate PET image; if not, use the intermediate PET image as the target PET image.

[0062] Optionally, the module for determining the preset number of filtering times is used to read the scanning data of the current frame image from the test PET scanning data, and perform image reconstruction on the scanning data to obtain the current frame image; use the guiding kernel matrix corresponding to the test PET scanning data to perform guided kernel filtering on the current frame image to obtain an intermediate PET image; when it is detected that an intermediate PET image is generated, determine whether the current number of tests is greater than 0; if so, record the preset image quality parameters and the corresponding number of filtering times of the intermediate PET image; and convert the intermediate PET image into the corresponding scanning data, and perform image reconstruction on the scanning data to update the current frame image, and use the guiding kernel matrix to perform guided kernel filtering on the current frame image to obtain the intermediate PET image; if not, determine the lowest value among all the recorded preset image quality parameters, and use the number of filtering times corresponding to the lowest value as the preset number of filtering times.

[0063] Optionally, when determining the preset image quality parameters of the intermediate PET image, the filtering module first determines a projection image of the intermediate PET image in a preset direction; and then uses the preset image quality parameters of the projection image as the preset image quality parameters of the intermediate PET image.

[0064] Optionally, the filtering module is used to read the scanning data of the current frame image from the PET scanning data, and perform image reconstruction on the scanning data to obtain the current frame image; use the guiding kernel matrix to perform guided kernel filtering on the current frame image to obtain an intermediate PET image, and determine the projection image of the intermediate PET image in a preset projection direction; when it is detected that a projection image is generated, determine whether the preset image quality parameters of the projection image meet preset conditions; if so, use the projection image as the target PET image; if not, determine the scanning data corresponding to the intermediate PET image, and perform image reconstruction on the scanning data to update the current frame image, and use the guiding kernel matrix to perform guided kernel filtering on the current frame image to update the intermediate PET image, and determine the projection image of the updated intermediate PET image in the preset projection direction.

[0065] The technical solution of the medical image processing device provided by the embodiment of the present invention divides the PET scanning data of the target object into at least two groups according to the time averaging principle; determines the composite frame image corresponding to each group of PET scanning data to obtain at least two composite frame images, so that the composite frame image carries spatial information; determines the guiding kernel matrix corresponding to the at least two composite frame images, so that the guiding kernel matrix carries the spatial information carried by all composite frame images; uses the guiding kernel matrix to perform guided kernel filtering on each frame image corresponding to the PET scanning data to obtain the target PET image of the corresponding frame, so that in the process of filtering each frame image through the guiding kernel matrix, the spatial information of the corresponding frame image is supplemented through the guiding kernel matrix, so that the target PET image has a lower noise level and richer spatial information, thereby improving the image reconstruction quality of the scanning data collected by the low-dose PET imaging technology, and thus improving the promotion speed of the low-dose PET imaging technology in clinical practice.

[0066] The medical image processing device provided in the embodiment of the present invention can execute the medical image processing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0067] Embodiment 3

[0068] Figure 4 A schematic diagram of the structure of an image processing device provided in Embodiment 3 of the present invention is shown in FIG. Figure 4 As shown, the device includes a processor 201, a memory 202, an input device 203, and an output device 204; the number of processors 201 in the device can be one or more. Figure 4 A processor 201 is taken as an example; the processor 201, the memory 202, the input device 203 and the output device 204 in the device can be connected by a bus or other means. Figure 4 The example of connecting through bus is taken in the following.

[0069] The memory 202, as a computer-readable storage medium, can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the medical image processing method in the embodiment of the present invention (for example, the grouping module 11, the image reconstruction module 12, the matrix determination module 13 and the filtering module 14). The processor 201 executes various functional applications and data processing of the device by running the software programs, instructions and modules stored in the memory 202, that is, realizes the above-mentioned medical image processing method.

[0070] The memory 202 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 202 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 202 may further include a memory remotely arranged relative to the processor 201, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0071] The input device 203 may be used to receive input digital or character information, and to generate key signal input related to user settings and function control of the device.

[0072] The output device 204 may include a display device such as a display screen, for example, a display screen of a user terminal.

[0073] Embodiment 4

[0074] An embodiment of the present invention further provides a storage medium containing computer executable instructions, wherein the computer executable instructions are used to perform a medical image processing method when executed by a computer processor, the method comprising:

[0075] Dividing the PET scan data of the target object into at least two groups according to the time averaging principle;

[0076] Determine a composite frame image corresponding to each set of PET scan data to obtain at least two composite frame images;

[0077] Determine a guide kernel matrix corresponding to the at least two composite frame images;

[0078] The guided kernel matrix is ​​used to perform guided kernel filtering on each frame image corresponding to the PET scan data to obtain a target PET image of the corresponding frame.

[0079] Of course, the computer executable instructions of a storage medium containing computer executable instructions provided by an embodiment of the present invention are not limited to the operations of the method described above, but can also execute related operations in the medical image processing method provided by any embodiment of the present invention.

[0080] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for enabling an image processing device (which can be a personal computer, server, or network device, etc.) to perform the medical image processing method described in each embodiment of the present invention.

[0081] It is worth noting that in the embodiment of the above-mentioned medical image processing device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0082] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A medical image processing method, characterized in that: include: Dividing the PET scan data of the target object into at least two groups according to the time averaging principle; Determine a composite frame image corresponding to each set of PET scan data to obtain at least two composite frame images; Determine a guide kernel matrix corresponding to the at least two composite frame images; Using the guided kernel matrix to perform guided kernel filtering on each frame of image corresponding to the PET scan data to obtain a target PET image of the corresponding frame; The step of determining the guiding kernel matrix corresponding to the at least two composite frame images comprises: Determine the guide kernel matrix corresponding to the at least two composite frame images based on a k-nearest neighbor algorithm; Wherein, the guiding kernel matrix includes pixel brightness information of the at least two composite frame images; The step of using the guided kernel matrix to perform guided kernel filtering on each frame of the image corresponding to the PET scan data to obtain a target PET image of the corresponding frame includes: Reading scan data of a current frame image from the PET scan data, and performing image reconstruction on the scan data to obtain a current frame image; Obtaining a preset number of filtering times, and using the guiding kernel matrix to perform guided kernel filtering on the current frame image to obtain an intermediate PET image; When it is detected that the intermediate PET image is generated, determining whether the current filtering number is greater than 0; If yes, converting the intermediate PET image into corresponding scan data, reconstructing the scan data to update the current frame image, and performing guided kernel filtering on the current frame image using the guided kernel matrix to update the intermediate PET image; If not, the intermediate PET image is used as the target PET image; or, Reading scan data of a current frame image from the PET scan data, and performing image reconstruction on the scan data to obtain a current frame image; Using the guided kernel matrix to perform guided kernel filtering on the current frame image to obtain an intermediate PET image, and determining a projection image of the intermediate PET image in a preset projection direction; When it is detected that the projection image is generated, determining whether a preset image quality parameter of the projection image meets a preset condition; If yes, taking the projection image or the intermediate PET image as the target PET image; If not, determine the scanning data corresponding to the intermediate PET image, reconstruct the scanning data to update the current frame image, use the guided kernel matrix to perform guided kernel filtering on the current frame image to update the intermediate PET image, and determine the projection image of the updated intermediate PET image in a preset projection direction.

2. The method according to claim 1, characterized in that: The PET scan data is divided into three groups.

3. The method according to claim 1, characterized in that The method for determining the preset filtering times includes: Reading scan data of a current frame image from the test PET scan data, and performing image reconstruction on the scan data to obtain a current frame image; Performing guided kernel filtering on the current frame image using a guided kernel matrix corresponding to the test PET scan data to obtain an intermediate PET image; When it is detected that an intermediate PET image is generated, determining whether the current test number is greater than 0; If yes, then record the preset image quality parameter and the corresponding filtering times of the intermediate PET image; and convert the intermediate PET image into corresponding scan data, and perform image reconstruction on the scan data to update the current frame image, and perform guided kernel filtering on the current frame image using the guided kernel matrix to obtain the intermediate PET image; If not, then determine the lowest value among all the recorded preset image quality parameters, and use the number of filtering times corresponding to the lowest value as the preset number of filtering times.

4. The method according to claim 3, characterized in that The method for determining the preset image quality parameters of the intermediate PET image includes: Determining a projection image of the intermediate PET image in a preset direction; The preset image quality parameters of the projection image are used as the preset image quality parameters of the intermediate PET image.

5. A medical image processing device, characterized in that: include: A grouping module, used for dividing the PET scan data of the target object into at least two groups according to a time averaging principle; An image reconstruction module, used for determining a composite frame image corresponding to each set of PET scan data to obtain at least two composite frame images; A matrix determination module, used to determine the guiding kernel matrix corresponding to the at least two composite frame images; A filtering module, configured to perform guided kernel filtering on each frame of image corresponding to the PET scan data using the guided kernel matrix to obtain a target PET image of the corresponding frame; The matrix determination module is used to determine the guiding kernel matrix corresponding to the at least two composite frame images based on the k-nearest neighbor algorithm; wherein the guiding kernel matrix includes the pixel brightness information of the at least two composite frame images The filtering module is used to read the scanning data of the current frame image from the PET scanning data, and perform image reconstruction on the scanning data to obtain the current frame image; obtain a preset filtering number, and use a guiding kernel matrix to perform guided kernel filtering on the current frame image to obtain an intermediate PET image; when it is detected that an intermediate PET image is generated, determine whether the current filtering number is greater than 0; if so, convert the intermediate PET image into corresponding scanning data, and perform image reconstruction on the scanning data to update the current frame image, and use the guiding kernel matrix to perform guided kernel filtering on the current frame image to update the intermediate PET image; if not, use the intermediate PET image as the target PET image; or, read the current frame image from the PET scanning data The method comprises the steps of: obtaining scanning data of the intermediate PET image, reconstructing the scanning data to obtain a current frame image, performing guided kernel filtering on the current frame image using a guided kernel matrix to obtain an intermediate PET image, and determining a projection image of the intermediate PET image in a preset projection direction; when it is detected that a projection image is generated, determining whether a preset image quality parameter of the projection image meets a preset condition; if so, taking the projection image as a target PET image; if not, determining the scanning data corresponding to the intermediate PET image, reconstructing the scanning data to update the current frame image, performing guided kernel filtering on the current frame image using a guided kernel matrix to update the intermediate PET image, and determining a projection image of the updated intermediate PET image in a preset projection direction.

6. An image processing device, characterized in that: The image processing device comprises: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the medical image processing method as described in any one of claims 1 to 4.

7. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions are used to perform the medical image processing method according to any one of claims 1 to 4 when executed by a computer processor.

Citation Information

Patent Citations

  • Dynamic PET image denoising method based on combined image guiding

    CN103955899A

  • CT image noise reducing method, device and computer equipment

    CN107622481A

  • Medical scanning imaging method and device, storage medium and computer equipment

    CN110136076A

  • Method and Apparatus of High-Resolution Image Reconstruction Based on Multi-Frame Low-Resolution Images

    US20130064472A1

  • Standardized uptake value (SUV) guided reconstruction control for improved outcome robustness in positron emission tomography (PET) imaging

    US20190325619A1