Image processing method, medical image processing system, apparatus, device, and medium

Abstract

The application discloses an image processing method, a medical image processing system, a device, equipment and a medium. The image processing method comprises the following steps: acquiring a spectral image of a target object; performing material decomposition on the spectral image based on the constituent material of the target object to obtain a material concentration map; and obtaining material content in the target object and / or deformation data of the target object according to the material concentration map. The method disclosed by the application performs material decomposition on the spectral image through the image processing mode, recognizes the material content in the target object and / or the deformation data of the target object, and improves the recognition effect and efficiency.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to an image processing method, a medical image processing system, apparatus, device, and medium. Background Technology

[0002] The material separation characteristics of energy-dispersive CT imaging technology can overcome the limitations of traditional single-parameter CT imaging, accurately identifying and quantifying multiple substances and tissue characteristics in a single scan, thus improving the accuracy of material analysis. However, most existing technologies can only identify and analyze single structures or tissues in energy-dispersive images through technical means, resulting in low image processing efficiency. Summary of the Invention

[0003] In view of the above problems, this application provides an image processing method that can decompose energy spectrum images into substances and identify the substance content and / or deformation data of the target object, thereby improving the recognition effect and efficiency.

[0004] In a first aspect, this application provides an image processing method, which includes: acquiring an energy spectrum image of a target object; performing material decomposition on the energy spectrum image based on the constituent materials of the target object to obtain a material concentration map; and obtaining the material content in the target object and / or the deformation data of the target object based on the material concentration map.

[0005] In the technical solution of this application embodiment, based on the constituent substances of the target object, the energy spectrum image of the target object is decomposed to obtain a substance concentration map. Then, based on the substance concentration map, the substance content or deformation data of the target object is obtained. This method can decompose the energy spectrum image to identify the substance content and / or deformation data of the target object, thereby improving the identification effect and efficiency.

[0006] In some embodiments, the energy spectrum image is decomposed based on the constituent substances of the target object, including: performing two-substance decomposition on the energy spectrum image based on the constituent substances of the target object to obtain a two-substance concentration map; and / or performing three-substance decomposition on the energy spectrum image based on the constituent substances of the target object to obtain a three-substance concentration map, wherein the two-substance and three-substance include a common substance.

[0007] In some embodiments, obtaining the substance content in the target object and / or the deformation data of the target object based on the substance concentration map includes: obtaining the substance content in the target object based on a three-substance concentration map; and / or obtaining the deformation data of the target object based on a two-substance concentration map and a three-substance concentration map.

[0008] In some embodiments, obtaining deformation data of a target object based on a two-substance concentration map and a three-substance concentration map includes: fusing the common substance concentration map in the two-substance concentration map and the three-substance concentration map to obtain a fused substance concentration map; and obtaining deformation data of the target object based on the fused substance concentration map.

[0009] In some embodiments, obtaining the substance content in a target object based on a three-substance concentration map includes: segmenting the target object in two substance concentration maps to obtain a first segmentation identifier; determining a first feature point of the target object based on the first segmentation identifier; and obtaining the substance content in the target object based on the first feature point and the three-substance concentration map.

[0010] In some embodiments, obtaining deformation data of a target object based on a fusion substance concentration map includes: segmenting the target object in the fusion substance concentration map to obtain a second segmentation identifier; determining a second feature point of the target object based on the second segmentation identifier; and obtaining deformation data of the target object based on the second feature point and the fusion substance concentration map.

[0011] In some embodiments, obtaining deformation data of a target object based on a second feature point and a fused substance concentration map includes: determining the region image where the target sub-object is located from the fused substance concentration map based on the second feature point; and obtaining the deformation data of the target object based on the region image where the target sub-object is located.

[0012] In some embodiments, the method further includes: obtaining a first morphological parameter of the target object based on a first segmentation identifier and a first feature point.

[0013] In some embodiments, the method further includes: obtaining a second morphological parameter of the target object based on a second segmentation identifier and a second feature point.

[0014] On the other hand, this application provides a medical image processing system capable of executing the method of any of the above embodiments. The system includes: an image acquisition module for acquiring medical energy spectrum CT images of a target object; and an image processing module connected to the image acquisition module for performing material decomposition on the medical energy spectrum CT images based on the constituent substances of the target object to obtain a material concentration map, and obtaining the material content in the target object and / or the deformation data of the target object based on the material concentration map.

[0015] In some embodiments, the target object includes the spine and / or intervertebral disc; the substance concentration map includes at least one of a water concentration map, a calcium concentration map, a fat concentration map, and a hydroxyapatite concentration map; the image processing module includes: a substance decomposition unit, used to perform water-calcium decomposition on the medical spectral CT image based on the constituent substances of the target object, and / or to perform water-fat-hydroxyapatite decomposition on the medical spectral CT image to obtain a substance concentration map; an output unit, connected to the substance decomposition unit, used to output the substance content in the spine and / or the deformation data of the intervertebral disc according to the substance concentration map, wherein the substance content in the spine characterizes the degree of osteoporosis of the spine.

[0016] In some embodiments, the substance content in the target object includes at least one of hydroxyapatite content and fat content; the output unit is also configured to output the hydroxyapatite content and / or fat content according to the substance concentration map.

[0017] On the other hand, this application provides an image processing apparatus, the apparatus comprising: an acquisition module for acquiring an energy spectrum image of a target object; a decomposition module for decomposing the energy spectrum image into substances based on the constituent substances of the target object to obtain a substance concentration map; and a obtaining module for obtaining the substance content in the target object and / or the deformation data of the target object based on the substance concentration map.

[0018] On the other hand, this application provides an electronic device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method of any of the above embodiments.

[0019] On the other hand, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method of any of the above embodiments.

[0020] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0021] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiments below. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0022] Figure 1 A flowchart of an image processing method according to an embodiment of this application is shown;

[0023] Figure 2 A block diagram of a medical image processing system according to an embodiment of this application is shown;

[0024] Figure 3 A flowchart of the medical image processing system according to an embodiment of this application is shown;

[0025] Figure 4 A schematic diagram of a spine identification image according to an embodiment of this application is shown;

[0026] Figure 5 This illustration shows a schematic diagram of intervertebral disc tissue identification images according to an embodiment of this application;

[0027] Figure 6 A flowchart illustrating the identification of vertebral osteoporosis and intervertebral disc degenerative diseases according to an embodiment of this application is shown.

[0028] Figure 7 A block diagram of an image processing apparatus according to an embodiment of this application is shown;

[0029] Figure 8 A schematic diagram of an electronic device according to an embodiment of this application is shown. Detailed Implementation

[0030] The embodiments of the technical solution of this application will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of this application and are therefore merely examples, and should not be used to limit the scope of protection of this application.

[0031] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0032] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0033] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0034] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0035] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0036] In the description of the embodiments of this application, the technical terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the embodiments of this application and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of this application.

[0037] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0038] The material separation characteristics of energy-dispersive CT imaging technology can overcome the limitations of traditional single-parameter CT imaging, accurately identifying and quantifying multiple substances and tissue characteristics in a single scan, thus improving the accuracy of material analysis. However, most existing technologies can only identify and analyze single structures or tissues in energy-dispersive images through technical means, resulting in low image processing efficiency.

[0039] For example, spectral CT images, through dual-base decomposition algorithms (such as water / iodine, calcium / water base pairs), can not only accurately quantify bone mineral density values, but also clearly present the spatial relationship between the intervertebral disc and surrounding nerve structures through multi-planar reconstruction technology. Therefore, bone quality assessment and intervertebral disc degeneration analysis can be completed simultaneously in a single scan, thereby improving the processing efficiency of spectral CT images and effectively reducing the workload.

[0040] A method for confirming vertebral bone mineral density involves acquiring the classification and localization results of at least one vertebra from a CT image, further segmenting it, identifying the region of interest (VOI) of the cancellous bone in each vertebra, and confirming the bone mineral density of each vertebra based on the VOI and the corresponding hydroxyapatite sequence value.

[0041] A method for detecting key points in the spine involves segmenting the vertebral bodies in a spinal medical image and locating key points, which are the anterior, posterior, left, and right edges of the superior endplate, as well as the anterior, posterior, left, and right edges of the inferior endplate.

[0042] A multimodal image fusion method for the spine based on CT and MRI (magnetic resonance imaging) images can generate multimodal fused images that clearly display the hard and soft tissue structures of the spine. The methods mentioned above can only analyze the material of one object in the spectral CT image, resulting in low image processing efficiency.

[0043] In the technical solution of this application embodiment, based on the constituent substances of the target object, the energy spectrum image of the target object is decomposed to obtain a substance concentration map. Then, based on the substance concentration map, the substance content or deformation data of the target object is obtained. This method can decompose the energy spectrum image to identify the substance content and / or deformation data of the target object, thereby improving the identification effect and efficiency.

[0044] Figure 1 A flowchart of an image processing method according to an embodiment of this application is shown.

[0045] like Figure 1 As shown, the image processing method 100 provided in this application includes steps S110 to S130.

[0046] Step S110: Obtain the energy spectrum image of the target object.

[0047] For example, the target object can be a human or animal tissue structure, such as bones, liver, or other organs, or it can be a non-human tissue, such as inorganic nanomaterials or electronic components; for example, an energy spectrum image of the spinal region can be acquired, and the energy spectrum image can include high-energy images and low-energy images from energy spectrum CT.

[0048] Step S120: Based on the constituent substances of the target object, perform substance decomposition on the energy spectrum image to obtain a substance concentration map.

[0049] For example, the constituent substances of the target object can be the basic constituent elements of the corresponding target object, such as water, iodine, calcium, etc., so that the energy spectrum image can be decomposed into two basic substances, such as the decomposition method based on water-iodine and calcium-water basic substances, to obtain the concentration maps of the decomposed substances.

[0050] Step S130: Based on the substance concentration map, obtain the substance content in the target object and / or the deformation data of the target object.

[0051] For example, based on the substance concentration map, i.e. the concentration map of the decomposed substances, the substance content in the target object can be obtained, such as the calcium content, water content, etc. in the target object; or the deformation data of the target object can be obtained, for example, based on the substance concentration map obtained by analyzing the energy spectrum image, the relevant data results of the deformation of the target object caused by the change of the concentration of the constituent substances can be identified.

[0052] In the technical solution of this application embodiment, based on the constituent substances of the target object, the energy spectrum image of the target object is decomposed to obtain a substance concentration map. Then, based on the substance concentration map, the substance content or deformation data of the target object is obtained. This method can decompose the energy spectrum image to identify the substance content and / or deformation data of the target object, thereby improving the identification effect and efficiency.

[0053] The image processing method based on this application can either identify the material content or deformation data of the target object separately, or simultaneously obtain both the material content and deformation data of the target object, thereby performing dual-modal recognition of the target object in an integrated manner, improving recognition efficiency and comprehensiveness.

[0054] For example, based on the constituent substances of the target object, the energy spectrum image is decomposed into substances. For instance, based on the constituent substances of the target object, the energy spectrum image is decomposed into two substances to obtain a concentration map of two substances; or based on the constituent substances of the target object, the energy spectrum image is decomposed into three substances to obtain a concentration map of three substances, wherein the two substances and the three substances include a common substance.

[0055] Specifically, the high-energy and low-energy images of the target object can be decomposed into two substances based on the constituent materials of the target object, such as water-iodine or calcium-water based substances, to obtain concentration maps of the two substances after decomposition, such as water concentration map and iodine concentration map; or the high-energy and low-energy images of the target object can be decomposed into three substances based on the constituent materials of the target object to obtain concentration maps of the three substances after decomposition, that is, the concentration maps of the three substances respectively. Thus, different substance decomposition methods can be used to analyze the target object in the same sequence of energy spectrum images.

[0056] In the technical solution of this application embodiment, based on the constituent substances of the target object, the energy spectrum image is decomposed into two substances to obtain a two-substance concentration map, and the energy spectrum image is decomposed into three substances to obtain a three-substance concentration map. This allows for the analysis of the target object using different substance decomposition methods in the same sequence of energy spectrum images, thereby improving the efficiency of image processing.

[0057] For example, the substance content or deformation data of the target object can be obtained from the substance concentration map. For instance, the substance content of the target object can be obtained from the three-substance concentration map; or the deformation data of the target object can be obtained from the two-substance concentration map and the three-substance concentration map.

[0058] Specifically, based on the three-substance concentration map, the substance content in the target object is obtained. For example, the content of each of the three substances in the target object can be obtained from the three-substance concentration map obtained from the decomposition of the three substances. Based on the two-substance concentration map and the three-substance concentration map, the deformation data of the target object is obtained. For example, based on the substance concentration map obtained from the decomposition of the two substances and the substance concentration map obtained from the decomposition of the three substances, the deformation-related results of the target object caused by the change in the concentration of the substances in it are obtained, i.e., deformation data.

[0059] In the technical solution of this application embodiment, the substance content in the target object can be obtained according to the three-substance concentration map; the deformation data of the target object can be obtained according to the two-substance concentration map and the three-substance concentration map, so that the target object can be subjected to dual-modal analysis in the same sequence of energy spectrum images, thereby improving the efficiency of image processing and effectively reducing the workload.

[0060] The following details how to obtain the substance content of a target object based on a three-substance concentration diagram.

[0061] For example, the substance content in the target object is obtained based on the three substance concentration maps. For instance, the target object in the two substance concentration maps is first segmented to obtain a first segmentation identifier; then, based on the first segmentation identifier, a first feature point of the target object is determined; and based on the first feature point and the three substance concentration maps, the substance content in the target object is obtained.

[0062] Specifically, for a target object in a two-substance concentration map, a segmentation network using deep learning technology is first used to segment the target object, outputting an identifier image of the target object, and obtaining a segmentation identifier for each segment, i.e., the first segmentation identifier. Then, a deep learning center point localization and classification network is used to output the center point coordinates (first feature point) and center point category of each target object's segment. Deep learning can improve recognition accuracy and efficiency. Based on the first feature point and the three-substance concentration map, the substance content in the target object is obtained. For example, based on the center point coordinates of the target object and the concentration map obtained from the decomposition of the three substances, the content of the three substances at the coordinate points of each target object's segment can be statistically calculated, taking the center point of the segment as the reference, thereby calculating the average content of each substance in any sub-region; or based on the center point coordinates of the target object's segment and the three-substance concentration map, the content of the three substances at the coordinate points of each target object's segment can be statistically calculated, taking the center point of the segment as the reference, and the ratio of any two substances in any sub-region can be calculated.

[0063] In the technical solution of this application embodiment, the target object is first segmented in the two substance concentration maps to obtain a first segmentation identifier. Then, based on the first segmentation identifier, the first feature point of the target object is determined. Based on the first feature point and the three substance concentration maps, the substance content in the target object is obtained. Thus, different substance decomposition methods can be used to analyze the target object in the same sequence of energy spectrum images, thereby improving the efficiency and accuracy of image processing.

[0064] Next, we will explain in detail how to obtain the deformation data of the target object based on the two-substance concentration map and the three-substance concentration map.

[0065] For example, deformation data of the target object can be obtained based on the two-substance concentration map and the three-substance concentration map. For instance, the common substance concentration map in the two-substance concentration map and the three-substance concentration map can be merged to obtain a merged substance concentration map. Then, the deformation data of the target object can be obtained based on the merged substance concentration map.

[0066] Specifically, the common substance concentration map obtained from the decomposition of two substances (denoted as the first common substance concentration map) and the common substance concentration map obtained from the decomposition of three substances (denoted as the second common substance concentration map) can be fused using a linear fusion method to obtain a fused substance concentration map, as shown in formula (1):

[0067] P 融合 =ωP1+(1-ω)P2 (1)

[0068] Where P is the fusion vector, ω is the weight, P1 is the first common substance concentration map, and P2 is the second common substance concentration map. An average fusion method can be used, which involves adding the corresponding pixels of the two common substance concentration maps to obtain an accumulated substance concentration map. The pixel values ​​of the accumulated substance concentration map (the number corresponding to the accumulated image) are then divided by 2 to obtain the fused substance concentration map. Finally, based on the fused substance concentration map, the deformation data of the target object is obtained.

[0069] For example, deformation data of the target object is obtained based on the fusion substance concentration map. For instance, the target object in the fusion substance concentration map is first segmented to obtain a second segmentation identifier; then, based on the second segmentation identifier, a second feature point of the target object is determined; and based on the second feature point and the fusion substance concentration map, the deformation data of the target object is obtained.

[0070] For example, deformation data of the target object is obtained based on the second feature point and the fused substance concentration map. For instance, firstly, the region image where the target sub-object is located is determined from the fused substance concentration map based on the second feature point; then, the deformation data of the target object is obtained based on the region image where the target sub-object is located.

[0071] Specifically, for the target object in the fused material concentration map, a segmentation network using deep learning techniques, such as ResUnet or VBNet, is first used to segment the target object, obtaining a segmentation identifier (second segmentation identifier). Based on each segmentation identifier, the center point coordinates (second feature points) of each target object's segment are calculated. Then, the fused material concentration map and the second segmentation identifier are input, and based on the center point of each target object's segment, the (sub)region where the target sub-object is located is sampled (the target sub-object includes at least one segment corresponding to the second segmentation identifier and two segment corresponding to the first segmentation identifier). A deep learning dual-channel classification network is used on this region, including but not limited to CNN, VGG, and Attention network structures, to identify the data related to the deformation results of the target object.

[0072] In the technical solution of this application embodiment, the common substance concentration map of two substance concentration maps and three substance concentration maps are fused to obtain a fused substance concentration map. Then, the target object in the fused substance concentration map is segmented to obtain a second segmentation identifier. Based on the second segmentation identifier, the second feature point of the target object is determined. According to the second feature point, the region image where the target sub-object is located is determined from the fused substance concentration map. Then, according to the region image where the target sub-object is located, the deformation data of the target object is obtained. Thus, in the same sequence of energy spectrum images, different substance decomposition methods are used to analyze the target object, improving the efficiency and accuracy of image processing.

[0073] For example, the method further includes: obtaining the first morphological parameters of the target object based on the first segmentation identifier and the first feature point.

[0074] Specifically, based on the first segmentation identifier of each segment of the target object and the coordinates of the center point (first feature point) of each segment, the coordinates, volume, diameter, area and other parameters of each segment of the target object can be calculated, i.e., the first morphological parameters.

[0075] For example, the method further includes obtaining a second morphological parameter of the target object based on the second segmentation identifier and the second feature point.

[0076] Specifically, the center point coordinates of each segment are calculated based on the second segmentation identifier of each segment of the target object, and the volume, perimeter and other morphological parameters of each segment of the target object are obtained based on this, namely the second morphological parameters.

[0077] In the technical solution of this application embodiment, the first morphological parameters of the target object can be obtained based on the first segmentation identifier and the first feature point, and the second morphological parameters of the target object can be obtained based on the second segmentation identifier and the second feature point. Thus, after processing the energy spectrum image, while outputting the material content and deformation data of the target object, the morphological parameters of the target object are also output for additional reference, thereby improving the comprehensiveness of image processing and analysis.

[0078] Figure 2 A block diagram of a medical image processing system according to an embodiment of this application is shown.

[0079] like Figure 2 As shown, the medical image processing system 200 provided in this application embodiment is capable of executing the method of any of the above embodiments. The system 200 includes: an image acquisition module 210 and an image processing module 220.

[0080] For example, the image acquisition module 210 is used to acquire medical spectral CT images of the target object. The image acquisition module 210 may be a CT scanner with spectral imaging function, and the spectral CT images may include high-energy images and low-energy images based on the same sequence. The image processing module 220 is connected to the image acquisition module 210 and is used to perform material decomposition on the medical spectral CT images based on the constituent substances of the target object to obtain a material concentration map, and obtain the material content in the target object or the deformation data of the target object based on the material concentration map. For details, please refer to the implementation description of the above method, which will not be repeated here.

[0081] For example, the target object can be a human or animal tissue structure, such as bones, liver, or other organs, or it can be a non-human tissue, such as inorganic nanomaterials or electronic components; for example, an energy spectrum image of the spine or intervertebral disc can be acquired, and the energy spectrum image can include high-energy images and low-energy images from energy spectrum CT.

[0082] For example, the constituent substances of the target object can be the basic constituent elements of the corresponding target object, such as water, iodine, calcium, etc., so that the energy spectrum image can be decomposed into two basic substances, such as the decomposition method based on water-iodine and calcium-water basic substances, to obtain the concentration maps of the decomposed substances.

[0083] For example, based on the substance concentration map, i.e. the concentration map of the decomposed substances, the substance content in the target object can be obtained, such as the calcium content, water content, etc. in the target object; or the deformation data of the target object can be obtained, for example, based on the substance concentration map obtained by analyzing the energy spectrum image, the relevant data results of the deformation of the target object caused by the change of the concentration of the constituent substances can be identified.

[0084] In the technical solution of this application embodiment, a medical energy spectrum CT image of the target object is acquired by an image acquisition module, and a material decomposition of the medical energy spectrum CT image is performed by an image processing module based on the constituent substances of the target object to obtain a material concentration map. Based on the material concentration map, the material content or deformation data of the target object is obtained. This method can decompose the medical energy spectrum CT image into substances, identify the material content and / or deformation data of the target object, improve the identification effect and efficiency, and effectively reduce the workload.

[0085] In one example, the medical image processing system can perform water-calcium decomposition and water-fat-hydroxyapatite decomposition on medical spectral CT images, respectively, and then classify osteoporosis and identify degenerative diseases based on the concentration maps of the decomposed substances, combined with... Figure 3 Describe it.

[0086] Figure 3 A flowchart of the medical image processing system according to an embodiment of this application is shown.

[0087] like Figure 3 As shown, the input is a spectral CT image (the medical spectral CT image of the target object), which is then subjected to water-calcium decomposition and water-fat-hydroxyapatite decomposition respectively. Then, based on the decomposed calcium map, the vertebral bodies are segmented, located, and classified. Based on the hydroxyapatite image (HAP map), bone density is calculated, and osteoporosis classification is diagnosed. After fusing the two decomposed water maps, the intervertebral discs are segmented and degenerative diseases are identified based on the fused water map, thus outputting the patient's degree of vertebral osteoporosis and whether they have intervertebral disc degenerative diseases. The details are explained below.

[0088] For example, continue to refer to Figure 2 The target object includes the spine or intervertebral disc; the substance concentration map includes at least one of water concentration map, calcium concentration map, fat concentration map, and hydroxyapatite concentration map; the image processing module includes: a substance decomposition unit 221, used to perform water-calcium decomposition on the medical energy spectrum CT image based on the constituent substances of the target object, or to perform water-fat-hydroxyapatite decomposition on the medical energy spectrum CT image to obtain a substance concentration map; an output unit 222, connected to the substance decomposition unit 221, used to output the substance content in the spine or the deformation data of the intervertebral disc according to the substance concentration map, wherein the substance content in the spine characterizes the degree of osteoporosis of the spine.

[0089] For example, the content of substances in the target object includes at least one of hydroxyapatite content and fat content; the output unit 222 is also used to output the hydroxyapatite content or fat content according to the substance concentration map.

[0090] Specifically, a CT scanner (image acquisition module 410) can be used to capture spectral CT images of the chest and abdomen, and medical spectral CT images of the spine or intervertebral disc can be obtained. Then, based on the constituent elements (compositional substances) of the chest and abdomen, the material decomposition unit 421 performs water-calcium decomposition on the medical spectral CT images to obtain a water concentration map (first water map) and a calcium concentration map; and performs water-fat-hydroxyapatite decomposition to obtain a water concentration map (second water map), a fat concentration map (fat map), and a hydroxyapatite concentration map (HAP map).

[0091] Then, through output unit 422, the center point coordinates and morphological parameters of each vertebra are obtained based on the calcium concentration map (refer to the implementation description of the medical image processing system above for details). Based on the HAP map and the center point coordinates of the vertebrae, the hydroxyapatite content of each vertebra is output, and the average hydroxyapatite content of any region is calculated as the bone mineral density value of the corresponding vertebra. Based on this, the cancellous bone region of the vertebrae is sampled. The system will classify osteoporosis according to the standards in the clinical diagnostic guidelines for osteoporosis. Specifically, if the bone mineral density value is <80 mg / cm³, the osteoporosis will be classified as osteoporosis. 3 If the bone mineral density value is between 80-120 mg / cm³, then the vertebra is considered to be in a state of osteoporosis; if the bone mineral density value is between 80-120 mg / cm³, then the vertebra is considered to be in a state of osteoporosis. 3 Within the specified range, it is considered low bone mass; if the bone mineral density value is >120 mg / cm³, it is considered low bone mass. 3If the bone mass is determined to be normal, the system will automatically output the bone mineral density value and corresponding classification results for each vertebra. At the same time, based on the HAP map and fat map, the system can obtain the hydroxyapatite content and fat content of each vertebra, calculate the ratio of hydroxyapatite to fat in the cancellous bone region of the vertebra, and provide doctors with the proportion of trabecular bone and bone marrow in the vertebra as an auxiliary parameter for bone mineral density measurement. In addition, bone mineral density measurements are performed on different sub-regions (ROIs) of the spine to form a hydroxyapatite distribution histogram, which provides doctors with the uniformity of bone mineral density distribution and can predict the risk of vertebral fractures in advance.

[0092] Through the output unit 422, based on the fused water map of the first and second water maps, degenerative diseases of the intervertebral disc are identified. The identification results include degenerative and non-degenerative lesions, as well as the morphological parameters of the intervertebral disc. Doctors can refer to the vertebral osteoporosis and intervertebral disc degenerative lesions for diagnosis.

[0093] In the technical solution of this application embodiment, the material decomposition unit performs water-calcium decomposition and water-fat-hydroxyapatite decomposition on the medical energy spectrum CT image based on the composition of the target object to obtain a material concentration map. The output unit 422 outputs the hydroxyapatite content or fat content according to the material concentration map to assist doctors in making a diagnosis by referring to the osteoporosis of the vertebral body and the degenerative lesions of the intervertebral disc. This allows for an integrated and comprehensive detection and diagnosis of the vertebral body and intervertebral disc, providing effective assistance to doctors in diagnosis and improving image processing efficiency.

[0094] In one example, the decomposition and identification of medical spectral CT images of the chest and abdomen using a medical image processing system can be combined with... Figure 4 and Figure 5 Provide a detailed description.

[0095] For example, the image processing module 220 performs material decomposition on the medical spectral CT image based on the constituent substances of the target object, including: for example, performing two-substance decomposition on the medical spectral CT image based on the constituent substances of the target object to obtain a two-substance concentration map; or performing three-substance decomposition on the medical spectral CT image based on the constituent substances of the target object to obtain a three-substance concentration map, wherein the two-substance and three-substance include a common substance.

[0096] Specifically, high-energy and low-energy images from chest and abdominal spectral CT can be decomposed using a two-substance decomposition based on water and calcium to obtain two concentration maps of the decomposed substances, namely a water concentration map (first water map) and a calcium concentration map (calcium map); or high-energy and low-energy images from chest and abdominal spectral CT can be decomposed using a three-substance decomposition based on water, fat, and hydroxyapatite to obtain three concentration maps of the decomposed substances, namely a water concentration map (second water map), a fat concentration map (fat map), and a hydroxyapatite concentration map (HAP map). Thus, different substance decomposition methods can be used to analyze the target object in the same sequence of spectral images.

[0097] In the technical solution of this application embodiment, the image processing module performs two-substance decomposition on the medical energy spectrum CT image based on the constituent substances of the target object to obtain a two-substance concentration map, and performs three-substance decomposition on the energy spectrum image to obtain a three-substance concentration map. This allows for the analysis of the target object using different substance decomposition methods in the same sequence of energy spectrum images, thereby improving the efficiency of image processing.

[0098] For example, the image processing module 220 obtains the substance content or deformation data of the target object based on the substance concentration map. For instance, it obtains the substance content of the target object based on the three substance concentration map; or it obtains the deformation data of the target object based on the two substance concentration map and the three substance concentration map.

[0099] Specifically, the image processing module 220 obtains the substance content in the target object based on the three-substance concentration map. For example, it can obtain the hydroxyapatite content or fat content of the spine (target object one) based on the HAP map and fat map in the three-substance concentration map obtained from the decomposition of water-fat-hydroxyapatite. Based on the two-substance concentration map and the three-substance concentration map, it obtains the deformation data of the target object. For example, it can obtain the degenerative deformation result (deformation data) of the intervertebral disc (target object two) based on the first water map obtained from the decomposition of water-calcium (two-substance decomposition) and the second water map obtained from the decomposition of water-fat-hydroxyapatite (three-substance decomposition).

[0100] In the technical solution of this application embodiment, the substance content in the target object can be obtained by the image processing module based on the three-substance concentration map; the deformation data of the target object can be obtained based on the two-substance concentration map and the three-substance concentration map, thereby performing dual-modal analysis on the target object in the same sequence of energy spectrum images, improving the efficiency of image processing and effectively reducing the workload.

[0101] For example, the image processing module 220 obtains the substance content in the target object based on the three substance concentration maps. For example, it first segments the target object in the two substance concentration maps to obtain a first segmentation identifier; then, based on the first segmentation identifier, it determines the first feature point of the target object; and based on the first feature point and the three substance concentration maps, it obtains the substance content in the target object.

[0102] Figure 4 A schematic diagram of a spine identification image according to an embodiment of this application is shown.

[0103] Specifically, such as Figure 4 As shown, the image processing module 220 first uses a deep learning spine segmentation network to segment the spine in the calcium map (two-substance concentration map), outputting a spine identification image to obtain the segmentation identifier of each vertebra (the first segmentation identifier, i.e., the red part in the figure); then, based on the calcium map, a deep learning center point localization and classification network is used to output the coordinates of the vertebral center point (the first feature point) and the center point category of each vertebra. Deep learning can improve the accuracy and efficiency of recognition; based on the first feature point and the three-substance concentration map, the substance content in the target object is obtained. For example, based on the vertebral center point coordinates and the HAP map (three-substance concentration map), with the vertebral center point as the reference, the hydroxyapatite content of each vertebral coordinate point can be counted, thereby calculating the average hydroxyapatite content of any sub-region; or based on the vertebral center point coordinates, the HAP map and the fat map, with the vertebral center point as the reference, the hydroxyapatite content and fat content of each vertebral coordinate point can be counted, and the ratio of hydroxyapatite to fat in any sub-region can be calculated.

[0104] In the technical solution of this application embodiment, the target object in the two substance concentration maps is first segmented by the image processing module to obtain the first segmentation identifier. Then, based on the first segmentation identifier, the first feature point of the target object is determined. Based on the first feature point and the three substance concentration maps, the substance content in the target object is obtained. Thus, different substance decomposition methods can be used to analyze the target object in the same sequence of energy spectrum images, thereby improving the efficiency and accuracy of image processing.

[0105] Next, we will specifically introduce how the image processing module 220 obtains the deformation data of the target object based on the two-substance concentration map and the three-substance concentration map.

[0106] For example, the image processing module 220 obtains the deformation data of the target object based on the two-substance concentration map and the three-substance concentration map. For example, the common substance concentration map in the two-substance concentration map and the three-substance concentration map is first fused to obtain a fused substance concentration map; then, the deformation data of the target object is obtained based on the fused substance concentration map.

[0107] Specifically, the first water map (two-substance concentration map) and the second water map (three-substance concentration map) can be merged using a linear fusion method to obtain a fused water map (fused substance concentration map), as shown in formula (1):

[0108] P 融合 =ωP1+(1-ω)P2 (1)

[0109] Where P is the fusion vector, ω is the weight, P1 is the first water map, and P2 is the second water map. An average fusion method can be used, which involves adding the corresponding pixels of the two water maps to obtain an accumulated water map. The pixel value of the accumulated water map (the number corresponding to the accumulated image) is divided by 2 to obtain the fused water map. Then, based on the fused water map (fused substance concentration map), the deformation data of the intervertebral disc (target object) is obtained.

[0110] For example, the image processing module 220 obtains the deformation data of the target object based on the fused substance concentration map. For example, the target object in the fused substance concentration map is first segmented to obtain a second segmentation identifier; then, based on the second segmentation identifier, a second feature point of the target object is determined; and based on the second feature point and the fused substance concentration map, the deformation data of the target object is obtained.

[0111] For example, the image processing module 220 obtains the deformation data of the target object based on the second feature point and the fused substance concentration map. For example, firstly, the region image where the target sub-object is located is determined from the fused substance concentration map based on the second feature point; then, the deformation data of the target object is obtained based on the region image where the target sub-object is located.

[0112] Figure 5 A schematic diagram of intervertebral disc tissue identification images according to an embodiment of this application is shown.

[0113] Specifically, such as Figure 5 As shown, the left image is the fusion water map, and the right image is the intervertebral disc tissue labeling image after network processing. First, a deep learning intervertebral disc segmentation network, including but not limited to ResUnet and VBNet, is used to segment the intervertebral disc tissue (target object) in the fusion water map, obtaining intervertebral disc segmentation labels (second segmentation labels, i.e., the green part in the right image). Based on each intervertebral disc segmentation label, the coordinates of the center point of each intervertebral disc (second feature point) are calculated. Then, the fusion water map and intervertebral disc segmentation labels are input, and based on the center point of each intervertebral disc, the (sub)region containing the intervertebral disc and the upper and lower vertebral bodies (target sub-objects) is sampled. A deep learning dual-channel classification network, including but not limited to CNN, VGG, and Attention, is used on this region to identify data related to the deformation results of the intervertebral disc (e.g., results related to intervertebral disc degenerative deformation).

[0114] In the technical solution of this application embodiment, the common substance concentration map in the two substance concentration maps and the three substance concentration maps are fused by the image processing module to obtain a fused substance concentration map. Then, the target object in the fused substance concentration map is segmented to obtain a second segmentation identifier. Based on the second segmentation identifier, the second feature point of the target object is determined. According to the second feature point, the region image where the target sub-object is located is determined from the fused substance concentration map. Then, according to the region image where the target sub-object is located, the deformation data of the target object is obtained. Thus, in the same sequence of energy spectrum images, different substance decomposition methods are used to analyze the target object, thereby improving the efficiency and accuracy of image processing.

[0115] For example, the image processing module 220 is also used to obtain the first morphological parameters of the target object based on the first segmentation identifier and the first feature points.

[0116] Specifically, the image processing module 220 can calculate the coordinates, volume, diameter, area, and other parameters of each vertebra based on the segmentation identifier (first segmentation identifier) ​​of each vertebra and the center point coordinates (first feature point) of each vertebra, i.e., the first morphological parameters.

[0117] For example, the image processing module 220 is also used to obtain the second morphological parameters of the target object based on the second segmentation identifier and the second feature point.

[0118] Specifically, the image processing module 220 calculates the center point coordinates of each intervertebral disc based on the segmentation identifier (second segmentation identifier) ​​of each intervertebral disc, and obtains the volume, circumference and other morphological parameters of each intervertebral disc, i.e., the second morphological parameters.

[0119] In the technical solution of this application embodiment, the first morphological parameter of the target object can be obtained by the image processing module based on the first segmentation identifier and the first feature point, and the second morphological parameter of the target object can be obtained based on the second segmentation identifier and the second feature point. Thus, after processing the energy spectrum image, the morphological parameter of the target object is output at the same time as the material content and deformation data of the target object, for additional reference, thereby improving the comprehensiveness of image processing and analysis.

[0120] Figure 6 A flowchart illustrating the identification of vertebral osteoporosis and intervertebral disc degenerative diseases according to an embodiment of this application is shown.

[0121] like Figure 6 As shown, for spectral CT images of the chest and abdomen, a medical image processing system can, based on material decomposition technology, complete vertebral osteoporosis screening and intervertebral disc degeneration diagnosis in one step, including the following steps:

[0122] Step 1: Input high-energy and low-energy images of the chest and abdomen from spectral CT.

[0123] Step 2: Decomposition of matter

[0124] (1) Perform two-substance decomposition based on water-calcium on the energy spectrum image to obtain the decomposed water concentration map (called the first water map) and calcium concentration map (called the calcium map);

[0125] (2) The energy spectrum image is decomposed into three substances based on water-fat-hydroxyapatite to obtain the decomposed water concentration map (called the second water map), fat concentration map (called the fat map), and hydroxyapatite concentration map (called the HAP map).

[0126] Step 3: Segmentation and Detection

[0127] Process 1: Vertebral body segmentation, classification, and key point localization

[0128] (1) Input a calcium map, use a deep learning spine segmentation network to output a spine labeling image;

[0129] (2) Input the calcium map, use the deep learning center point localization and classification network to output the coordinates of the vertebral body center point and the center point category;

[0130] (3) Based on the segmentation markers of each vertebra and the coordinates of the center point of each vertebra, calculate the coordinates, volume, diameter, area and other parameters of each vertebra;

[0131] Step 2: Vertebral Bone Mineral Density Measurement

[0132] (1) Input the HAP image and the key points of the vertebral body center obtained in Process 1. Using the vertebral body center point as the reference, sample the cancellous bone region of the vertebral body in the HAP image. For each point in this region, count its hydroxyapatite content, and then calculate the average hydroxyapatite content of this region, which is used as the corresponding vertebral body bone mineral density value;

[0133] (2) Input the HAP map and fat map, and the key point of the vertebral body center obtained in process one. Sample the cancellous bone region of the vertebral body in the HAP map and fat map with the vertebral body center point. Statistically calculate the hydroxyapatite content and fat content of each point in the region, and calculate the ratio of hydroxyapatite to fat in the region as an auxiliary parameter for bone density measurement.

[0134] Process 3: Intervertebral disc segmentation and identification of degenerative diseases

[0135] (1) Input the first water map obtained by water-calcium decomposition and the second water map obtained by water-fat-hydroxyapatite, and fuse them. The fusion can be carried out by linear fusion, see formula (1), or by average fusion, that is, the corresponding pixels of the two water maps are added together to obtain the cumulative water map. The pixel value of the cumulative water map is divided by 2 (the number corresponding to the cumulative image) to obtain the fused water map.

[0136] (2) Input the fused water map, use a deep learning intervertebral disc segmentation network, including but not limited to ResUnet, VBNet and other network structures, and output the intervertebral disc tissue identification image after network processing;

[0137] (3) Calculate the volume, center point coordinates, circumference and other morphological parameters of each intervertebral disc according to the segmentation markers of each intervertebral disc;

[0138] (4) Input the fusion water map and intervertebral disc segmentation markers. Based on the center point of each intervertebral disc, sample the sub-region containing the fusion water map and segmentation markers of the intervertebral disc and the upper and lower vertebral bodies. Use a deep learning dual-channel classification network on the sub-region images to identify intervertebral disc degenerative diseases. The identification results include degenerative and non-degenerative lesions. The classification network includes, but is not limited to, CNN, VGG, Attention and other network structures.

[0139] Step 4: Output Results

[0140] Vertebral body and intervertebral disc morphological parameter output: Based on the vertebral body and intervertebral disc identifiers obtained in step three, detailed morphological parameters for each vertebral body and intervertebral disc are automatically output. For the vertebral body, the output parameters include its volume, diameter, and area (calculated in process one based on the vertebral body segmentation identifiers and center point coordinates). For the intervertebral disc, the output morphological parameters include volume (obtained in the intervertebral disc morphological parameter calculation step in process three), center point coordinates (also from the calculation results in process three), and circumference (obtained through calculation in process three).

[0141] Vertebral body bone mineral density (BMD) output and classification: In the second step of vertebral body BMD measurement, the BMD value for each vertebra has been calculated. At this point, the system will classify the BMD according to the standards in the clinical diagnostic guidelines for osteoporosis. Specifically, if the BMD value is <80 mg / cm³, the BMD will be classified accordingly. 3 If the bone mineral density value is between 80-120 mg / cm³, then the vertebra is considered to be in a state of osteoporosis; if the bone mineral density value is between 80-120 mg / cm³, then the vertebra is considered to be in a state of osteoporosis. 3 Within the specified range, it is considered low bone mass; if the bone mineral density value is >120 mg / cm³, it is considered low bone mass. 3 The bone density was determined to be normal. The system will automatically output the bone mineral density value of each vertebra and the corresponding classification result.

[0142] Output of intervertebral disc degenerative disease identification results: Through the deep learning dual-channel classification network in process three, the degenerative disease status of each intervertebral disc has been identified. The system will output the degenerative disease identification result for each intervertebral disc, clearly labeling it as a degenerative lesion or a non-degenerative lesion.

[0143] Therefore, the medical image processing system based on this application can use different material decomposition methods to detect and diagnose vertebral bodies and intervertebral discs in medical spectral CT images of the chest and abdomen. It can complete the measurement of bone density and the diagnosis of intervertebral disc degenerative diseases in one stop, assist doctors in making rapid diagnoses, and effectively reduce their workload.

[0144] Figure 7 A block diagram of an image processing apparatus according to an embodiment of this application is shown.

[0145] This application provides an image processing apparatus 700, the apparatus 700 including:

[0146] The acquisition module 710 is used to acquire the energy spectrum image of the target object.

[0147] The decomposition module 720 is used to decompose the energy spectrum image based on the constituent substances of the target object to obtain a substance concentration map.

[0148] Module 730 is used to obtain the substance content and / or deformation data of the target object based on the substance concentration map.

[0149] For example, the decomposition module 720 is further configured to: decompose the energy spectrum image into two substances based on the constituent substances of the target object to obtain a concentration map of two substances; and / or decompose the energy spectrum image into three substances based on the constituent substances of the target object to obtain a concentration map of three substances, wherein the two substances and the three substances include a common substance.

[0150] For example, module 730 is further configured to: obtain the substance content in the target object based on the three-substance concentration map; and / or obtain the deformation data of the target object based on the two-substance concentration map and the three-substance concentration map.

[0151] For example, obtaining deformation data of a target object based on a two-substance concentration map and a three-substance concentration map includes: fusing the common substance concentration map in the two-substance concentration map and the three-substance concentration map to obtain a fused substance concentration map; and obtaining deformation data of the target object based on the fused substance concentration map.

[0152] For example, obtaining the substance content in a target object based on a three-substance concentration map includes: segmenting the target object in two substance concentration maps to obtain a first segmentation identifier; determining a first feature point of the target object based on the first segmentation identifier; and obtaining the substance content in the target object based on the first feature point and the three-substance concentration map.

[0153] For example, obtaining deformation data of a target object based on a fusion substance concentration map includes: segmenting the target object in the fusion substance concentration map to obtain a second segmentation identifier; determining a second feature point of the target object based on the second segmentation identifier; and obtaining deformation data of the target object based on the second feature point and the fusion substance concentration map.

[0154] For example, obtaining deformation data of a target object based on a second feature point and a fused substance concentration map includes: determining the region image where the target sub-object is located from the fused substance concentration map based on the second feature point; and obtaining the deformation data of the target object based on the region image where the target sub-object is located.

[0155] For example, the image processing apparatus 700 further includes: a first morphological parameter obtaining module, used to obtain the first morphological parameters of the target object based on the first segmentation identifier and the first feature points.

[0156] For example, the image processing apparatus 700 further includes a second morphological parameter obtaining module, used to obtain a second morphological parameter of the target object based on a second segmentation identifier and a second feature point.

[0157] Figure 8 A schematic diagram of an electronic device according to an embodiment of this application is shown.

[0158] This application provides an electronic device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the method in any of the above embodiments.

[0159] like Figure 8 As shown, for ease of understanding, embodiments of this application illustrate a specific electronic device 800.

[0160] Electronic device 800 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0161] like Figure 8 As shown, device 800 includes a computing unit 801, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 802 or a computer program loaded from storage unit 808 into random access memory (RAM) 803. RAM 803 may also store various programs and data required for the operation of electronic device 800. The computing unit 801, ROM 802, and RAM 803 are interconnected via bus 804. Input / output (I / O) interface 805 is also connected to bus 804.

[0162] Multiple components in electronic device 800 are connected to I / O interface 805. These components include: input unit 806, such as a keyboard or mouse; output unit 807, such as various types of displays or speakers; storage unit 808, such as a disk or optical disk; and communication unit 809, such as a network interface card (NIC), modem, or wireless transceiver. Communication unit 809 allows electronic device 800 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0163] The computing unit 801 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods described above. For example, in some embodiments, any one or more of the methods described above can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of any one or more of the methods described above can be performed. Alternatively, in other embodiments, the computing unit 801 can be configured to perform any one or more of the methods described above by any other suitable means (e.g., by means of firmware).

[0164] This application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method in any of the above embodiments.

[0165] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequential list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this application, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0166] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0167] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and they should all be covered within the scope of the claims and specification of this application. In particular, as long as there is no structural conflict, the various technical features mentioned in the embodiments can be combined in any way. This application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. An image processing method, characterized in that, The method includes: Obtain the energy spectrum image of the target object; Based on the constituent substances of the target object, the energy spectrum image is decomposed to obtain a substance concentration map. The decomposition of the energy spectrum image based on the constituent substances of the target object includes two-substance decomposition of the energy spectrum image based on the constituent substances of the target object to obtain a two-substance concentration map, and three-substance decomposition of the energy spectrum image based on the constituent substances of the target object to obtain a three-substance concentration map. The two-substance and three-substance include a common substance. Based on the three-substance concentration map, the substance content in the target object is obtained, and based on the two-substance concentration map and the three-substance concentration map, the deformation data of the target object is obtained; The step of obtaining the substance content in the target object based on the three substance concentration maps includes segmenting the target object in the two substance concentration maps to obtain a first segmentation identifier, determining a first feature point of the target object based on the first segmentation identifier, and obtaining the substance content in the target object based on the first feature point and the three substance concentration maps. The step of obtaining the deformation data of the target object based on the two-substance concentration map and the three-substance concentration map includes fusing the common substance concentration map in the two-substance concentration map and the three-substance concentration map to obtain a fused substance concentration map, segmenting the target object in the fused substance concentration map to obtain a second segmentation identifier, determining a second feature point of the target object based on the second segmentation identifier, determining the region image of the target sub-object in the fused substance concentration map based on the second feature point, and obtaining the deformation data of the target object based on the region image of the target sub-object.

2. The image processing method according to claim 1, characterized in that, The method further includes: Based on the first segmentation identifier and the first feature point, the first morphological parameters of the target object are obtained.

3. The image processing method according to claim 1, characterized in that, The method further includes: Based on the second segmentation identifier and the second feature point, the second morphological parameters of the target object are obtained.

4. A medical image processing system, characterized in that, The system is capable of performing the method according to any one of claims 1-3, the system comprising: The image acquisition module is used to acquire medical spectral CT images of the target object; An image processing module, connected to the image acquisition module, is used to perform material decomposition on the medical energy spectrum CT image based on the constituent substances of the target object to obtain a material concentration map. The material decomposition based on the constituent substances of the target object includes performing two-substance decomposition on the energy spectrum image based on the constituent substances of the target object to obtain a two-substance concentration map, and performing three-substance decomposition on the energy spectrum image based on the constituent substances of the target object to obtain a three-substance concentration map. The two-substance and three-substance components include a common substance. Based on the three-substance concentration map, the material content in the target object is obtained, and based on the two-substance concentration map and the three-substance concentration map, the deformation data of the target object is obtained. The step of obtaining the substance content in the target object based on the three substance concentration maps includes segmenting the target object in the two substance concentration maps to obtain a first segmentation identifier, determining a first feature point of the target object based on the first segmentation identifier, and obtaining the substance content in the target object based on the first feature point and the three substance concentration maps. The step of obtaining the deformation data of the target object based on the two-substance concentration map and the three-substance concentration map includes fusing the common substance concentration map in the two-substance concentration map and the three-substance concentration map to obtain a fused substance concentration map, segmenting the target object in the fused substance concentration map to obtain a second segmentation identifier, determining a second feature point of the target object based on the second segmentation identifier, determining the region image of the target sub-object in the fused substance concentration map based on the second feature point, and obtaining the deformation data of the target object based on the region image of the target sub-object.

5. The medical image processing system according to claim 4, characterized in that, The target objects include the spine and intervertebral discs; the substance concentration maps include water concentration maps, calcium concentration maps, fat concentration maps, and hydroxyapatite concentration maps; The image processing module includes: The substance decomposition unit is used to perform water-calcium decomposition and water-fat-hydroxyapatite decomposition on the medical energy spectrum CT image based on the composition of the target object, respectively, to obtain the substance concentration map. The output unit, connected to the substance decomposition unit, is used to output the substance content in the spine and the deformation data of the intervertebral disc according to the substance concentration map, wherein the substance content in the spine characterizes the degree of osteoporosis of the spine.

6. The medical image processing system according to claim 5, characterized in that, The substance content in the target object includes hydroxyapatite content and fat content; the output unit is also used to output the hydroxyapatite content and the fat content according to the substance concentration diagram.

7. An image processing apparatus, characterized in that, The device includes: The acquisition module is used to acquire the energy spectrum image of the target object; The decomposition module is used to decompose the energy spectrum image based on the constituent substances of the target object to obtain a substance concentration map. The decomposition of the energy spectrum image based on the constituent substances of the target object includes two-substance decomposition of the energy spectrum image based on the constituent substances of the target object to obtain a two-substance concentration map, and three-substance decomposition of the energy spectrum image based on the constituent substances of the target object to obtain a three-substance concentration map. The two-substance and three-substance decompositions include a common substance. The module is used to obtain the substance content in the target object based on the three-substance concentration map, and to obtain the deformation data of the target object based on the two-substance concentration map and the three-substance concentration map; The step of obtaining the substance content in the target object based on the three substance concentration maps includes segmenting the target object in the two substance concentration maps to obtain a first segmentation identifier, determining a first feature point of the target object based on the first segmentation identifier, and obtaining the substance content in the target object based on the first feature point and the three substance concentration maps. The step of obtaining the deformation data of the target object based on the two-substance concentration map and the three-substance concentration map includes fusing the common substance concentration map in the two-substance concentration map and the three-substance concentration map to obtain a fused substance concentration map, segmenting the target object in the fused substance concentration map to obtain a second segmentation identifier, determining a second feature point of the target object based on the second segmentation identifier, determining the region image of the target sub-object in the fused substance concentration map based on the second feature point, and obtaining the deformation data of the target object based on the region image of the target sub-object.

8. An electronic 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 implements the steps of the method according to any one of claims 1-3.

9. A computer-readable 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 1-3.

Images