Tissue feature value determination method and apparatus, computer device, and storage medium
By using a tissue segmentation model to segment and map medical images, the problem of inconsistent organ and tissue boundary identification was solved, thereby improving the accuracy and reliability of organ and tissue feature values.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI UNITED IMAGING HEALTHCARE
- Filing Date
- 2022-05-24
- Publication Date
- 2026-07-21
AI Technical Summary
In existing technologies, organ and tissue boundary identification relies on manual labeling, which leads to inconsistent identification results among different staff members and reduces the reliability of organ and tissue feature values.
A pre-trained tissue segmentation model is used to segment medical images to obtain organ regions. Based on tissue parameters, feature values of organ regions are determined, and mapping and registration are performed in combination with morphological and functional images to improve accuracy.
It achieves accurate classification of organs and tissues, improves the accuracy and reliability of characteristic value determination, and outputs structured reports to reflect characteristic value information.
Smart Images
Figure CN114972247B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of medical image processing technology, and in particular to a method, apparatus, computer device, and storage medium for determining tissue characteristic values. Background Technology
[0002] Medical image segmentation is very useful in determining organ and tissue feature values. Accurate organ and tissue boundary identification and tissue segmentation directly affect the determination results of organ and tissue feature values. Therefore, accurate organ and tissue boundary identification is very important.
[0003] In existing technologies, the boundary identification of organs and tissues is generally done manually by staff.
[0004] However, the manual marking method used by staff may lead to different staff members having different results in identifying the boundaries of organs and tissues in the same medical image. This also results in different determinations of the final organ and tissue feature values, reducing the reliability of the organ and tissue feature values. Summary of the Invention
[0005] Therefore, it is necessary to provide a method, apparatus, computer device, and storage medium for determining organizational characteristic values to address the aforementioned technical problems.
[0006] Firstly, this application provides a method for determining organizational characteristic values, including:
[0007] Acquire medical images;
[0008] Based on a pre-trained tissue segmentation model, medical images are segmented to obtain multiple organ partitions.
[0009] For each organ partition, obtain the corresponding tissue parameters and determine the feature values of the organ partition based on the tissue parameters.
[0010] In one embodiment, the medical image includes a morphological image; the medical image is segmented into multiple organ partitions based on a pre-trained tissue segmentation model, including:
[0011] Based on a pre-trained tissue segmentation model, tissue segmentation is performed on morphological images to obtain multiple organ partitions.
[0012] Accordingly, for each organ region, the corresponding tissue parameters for that organ region are obtained, including:
[0013] Obtain the functional image corresponding to the morphological image;
[0014] Tissue parameters corresponding to each organ region are obtained based on morphological and functional imaging images.
[0015] In one embodiment, tissue parameters corresponding to each organ region are obtained based on morphological and functional imaging maps, including:
[0016] The morphological images and functional images are mapped and registered to determine the target areas corresponding to each organ region in the functional images.
[0017] The tissue parameters within the target area are determined as the tissue parameters corresponding to each organ partition.
[0018] In one embodiment, the medical image map includes a functional image map; the medical image map is segmented into multiple organ partitions based on a pre-trained tissue segmentation model, including:
[0019] Based on the tissue segmentation model, functional images are segmented into multiple organ partitions.
[0020] Accordingly, for each organ region, the corresponding tissue parameters for that organ region are obtained, including:
[0021] Obtain the tissue parameters corresponding to each organ partition based on the tissue parameters of the region where each organ partition is located.
[0022] In one embodiment, the method further includes:
[0023] Statistical analysis of feature values for each organ region, and output of a structured report.
[0024] In one embodiment, the medical image is a liver image, and the organ partition includes at least one of the whole liver, liver segments, and liver lesion areas.
[0025] In one embodiment, the functional image map is at least one of a diffusion functional image map, a perfusion functional image map, and a fat functional image map.
[0026] Secondly, this application also provides an apparatus for determining tissue characteristic values, comprising:
[0027] The acquisition module is used to acquire medical image images;
[0028] The segmentation module is used to segment medical images into multiple organ partitions based on a pre-trained tissue segmentation model.
[0029] The determination module is used to obtain the tissue parameters corresponding to each organ partition and determine the feature values of the organ partition based on the tissue parameters.
[0030] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:
[0031] Acquire medical images;
[0032] Based on a pre-trained tissue segmentation model, medical images are segmented to obtain multiple organ partitions.
[0033] For each organ partition, obtain the corresponding tissue parameters and determine the feature values of the organ partition based on the tissue parameters.
[0034] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:
[0035] Acquire medical images;
[0036] Based on a pre-trained tissue segmentation model, medical images are segmented to obtain multiple organ partitions.
[0037] For each organ partition, obtain the corresponding tissue parameters and determine the feature values of the organ partition based on the tissue parameters.
[0038] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:
[0039] Acquire medical images;
[0040] Based on a pre-trained tissue segmentation model, medical images are segmented to obtain multiple organ partitions.
[0041] For each organ partition, obtain the corresponding tissue parameters and determine the feature values of the organ partition based on the tissue parameters.
[0042] The aforementioned method, apparatus, computer equipment, storage medium, and computer program product for determining tissue feature values acquire medical images, segment the images into multiple organ regions based on a pre-trained tissue segmentation model, and then obtain the corresponding tissue parameters for each organ region. Feature values for each organ region are then determined based on these tissue parameters. This method enables accurate segmentation of different tissue regions within organs, thereby improving the accuracy and reliability of the feature values determined based on the segmentation results. Attached Figure Description
[0043] Figure 1 This is a flowchart illustrating a method for determining organizational feature values in one embodiment;
[0044] Figure 2This is a flowchart illustrating the process of obtaining tissue parameters corresponding to organ partitions in one embodiment;
[0045] Figure 3 This is a flowchart illustrating the process of obtaining tissue parameters corresponding to organ partitions in another embodiment;
[0046] Figure 4 This is a structural block diagram of an organization feature value determination device in one embodiment;
[0047] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation
[0048] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0049] In one embodiment, such as Figure 1 As shown, a method for determining organizational feature values is provided. This embodiment illustrates the application of this method to a terminal. It is understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:
[0050] S110. Obtain medical images.
[0051] The medical image refers to the image and / or image of the organ or tissue of the object to be analyzed, acquired by medical imaging equipment. Optionally, the medical image can be a CT (Computerized Tomography) image or an NMR (Nuclear Magnetic Resonance) image. Optionally, the medical image can be a 2D image or a 3D image. In this embodiment, there are no specific limitations on the specific type or formation method of the medical image, as long as the tissue area of the organ or tissue is clearly presented. Optionally, the organ or tissue can be heart tissue, liver tissue, spleen and stomach tissue, etc.
[0052] Optionally, the terminal implementing the above-mentioned tissue feature value determination method can be the above-mentioned medical imaging device to directly acquire the medical image of the object to be analyzed; it can also communicate with the medical imaging device through a network to acquire the medical image of the object to be analyzed collected by the medical imaging device; or it can acquire the medical image of the object to be analyzed from the local storage area of the terminal or an external storage device.
[0053] S120. Based on a pre-trained tissue segmentation model, medical images are segmented to obtain multiple organ partitions.
[0054] In this context, multiple organ partitions refer to different regions within the same organ or tissue. For example, these could be different regions of the heart, such as the left atrium, right atrium, left ventricle, and right ventricle; or different regions of the liver, such as the left lobe, right lobe, entire liver segment, or diseased areas. Tissue segmentation is the process of dividing an organ or tissue into its various tissue regions on a medical image. The resulting tissue regions are the organ partitions described above. The tissue segmentation model is a network model trained using a large number of pre-divided medical images as training samples, based on a deep learning algorithm. Optionally, the tissue segmentation model can be V-Net, 3DU-Net, or other models.
[0055] Specifically, the terminal inputs the acquired medical image into the trained tissue segmentation model, which then performs tissue segmentation on the input medical image to obtain the location of each tissue region of the organ on the medical image, thus obtaining the aforementioned multiple organ partitions.
[0056] Optionally, multiple organ partitions can be displayed on the medical image using different display methods. For example, the boundaries of each organ partition can be drawn, and the corresponding tissue area can be labeled with text. Different colors or line types can be used to draw the boundaries, or the corresponding organ partitions can be displayed directly on the medical image using different colors, with different colors representing different tissue areas. Alternatively, each organ partition can be displayed separately.
[0057] S130. For each organ partition, obtain the tissue parameters corresponding to the organ partition, and determine the feature values of the organ partition based on the tissue parameters.
[0058] Among them, the tissue parameters corresponding to the organ partition are the tissue parameters of the tissue region corresponding to the organ partition obtained by measurement, and the feature values are the parameter values obtained after data processing of the tissue parameters.
[0059] Optionally, tissue parameters can be the content of target substances in the corresponding tissue area, such as iron content, oxygen content, fat content, etc., or the blood flow conditions in the corresponding tissue area, such as blood flow rate, blood flow velocity, etc.
[0060] Optionally, the feature values of the organ partitions may include the maximum or minimum value obtained by comparing the size of tissue parameters at different locations within a preset time period on the corresponding tissue region, or the average value obtained by summing and averaging, or the variance obtained by calculating the variance.
[0061] Optionally, after determining the various organ partitions on the medical image, the terminal obtains the tissue parameters of the corresponding tissue regions of each organ partition, and then processes the obtained tissue parameters to obtain the feature values of each organ partition, i.e., organ tissue feature values.
[0062] In this embodiment, the terminal acquires a medical image and performs tissue segmentation on the image based on a pre-trained tissue segmentation model to obtain multiple organ partitions. Then, it obtains the tissue parameters corresponding to each organ partition and determines the feature values of each partition based on these parameters. This method enables accurate segmentation of different tissue regions within an organ, thereby improving the accuracy of the feature values determined based on the segmentation results and consequently increasing the reliability of the obtained feature values.
[0063] In practical applications, functional imaging images include tissue parameters from the entire acquired organ tissue. Therefore, to obtain organ tissue feature values more efficiently, the aforementioned medical imaging images include functional imaging images. The above-mentioned S120, which involves segmenting the medical imaging images based on a pre-trained tissue segmentation model to obtain multiple organ partitions, includes:
[0064] Based on the tissue segmentation model, functional images are segmented into multiple organ regions.
[0065] Among them, the tissue segmentation model uses a large number of functional image maps of pre-divided tissue regions as training samples and is trained based on deep learning algorithms to obtain a network model.
[0066] Optionally, the aforementioned medical imaging image is a liver imaging image, and correspondingly, the organ region includes at least one of the whole liver, liver segments, and liver lesion areas. Correspondingly, the functional imaging image may include at least one of diffusion function imaging image, perfusion function imaging image, and fat function imaging image. The tissue parameter included in the diffusion function imaging image may be the diffusion coefficient, the tissue parameter included in the perfusion function imaging image may be the blood flow coefficient or blood flow velocity coefficient, and the tissue parameter included in the fat function imaging image may be the fat content or fat fraction of the liver.
[0067] Optionally, the terminal inputs the acquired functional image map into the trained tissue segmentation model, and the tissue segmentation model performs tissue segmentation on the input functional image map to obtain the location of each tissue region of the organ on the functional image map, thus obtaining the above-mentioned multiple organ partitions.
[0068] Accordingly, in S130 above, obtaining the tissue parameters corresponding to each organ partition includes:
[0069] Obtain the tissue parameters corresponding to each organ partition based on the tissue parameters of the region where the organ partition is located.
[0070] Optionally, since the functional imaging map includes tissue parameters of the entire organ tissue, after dividing the functional imaging map into various organ regions, the tissue parameters of the corresponding organ regions can be obtained, and then the tissue parameters of each organ region can be directly obtained based on the functional imaging map.
[0071] When performing organ segmentation on functional imaging images, issues such as insufficient image clarity and blurred tissue boundaries can arise. This problem can be addressed in three ways: First, during image acquisition, increasing the magnetic field strength, employing advanced acquisition methods such as multi-echo steady-state acquisition, and using more efficient k-space filling methods such as spiral or radial patterns can enable the device to acquire higher-resolution images in a shorter time. Second, preprocessing operations such as filtering, contrast enhancement, and edge detection can be performed on the images using traditional methods. Third, incorporating channel attention modules, deep supervision modules, and contrastive learning modules into the deep learning model for segmentation can enhance the ability to identify organ boundaries, thus enabling accurate organ segmentation on functional imaging images.
[0072] In this embodiment, the medical image map includes a functional image map. The terminal performs tissue segmentation on the functional image map based on a tissue segmentation model to obtain multiple organ partitions. Since the functional image map includes tissue parameters of the entire organ tissue, the tissue parameters corresponding to each organ partition can be directly obtained based on the segmentation results of the model, which improves the efficiency of obtaining tissue parameters and correspondingly improves the efficiency of determining feature values.
[0073] Functional imaging images include tissue parameters of the entire organ, which can effectively improve the efficiency of acquiring tissue parameters. To improve the accuracy of organ partitioning, medical imaging images include morphological images. Morphological images can clearly show the boundaries of each tissue region. Therefore, partitioning organs based on morphological images can greatly improve the accuracy of partitioning. The above-mentioned S120, performing tissue segmentation on the medical imaging image based on a pre-trained tissue segmentation model, yields multiple organ partitions, including:
[0074] Based on the tissue segmentation model, functional images are segmented into multiple organ regions.
[0075] The tissue segmentation model is a network model trained using a large number of morphological images of pre-divided tissue regions as training samples and based on a deep learning algorithm.
[0076] Optionally, the terminal inputs the acquired morphological image into the trained tissue segmentation model, and the tissue segmentation model performs tissue segmentation on the input morphological image to obtain the location of each tissue region of the organ on the morphological image, thus obtaining the above-mentioned multiple organ partitions.
[0077] For organ segmentation of morphological images, existing V-Net, 3DU-Net or other deep learning models can be used. Alternatively, channel attention modules, deep supervision modules, and contrastive learning modules can be added to deep learning models to enhance the ability to identify organ boundaries and improve the accuracy of organ segmentation.
[0078] Correspondingly, such as Figure 2 As shown, in S130 above, obtaining the tissue parameters corresponding to each organ partition includes:
[0079] S210. Obtain the functional image corresponding to the morphological image.
[0080] Among them, morphological images and functional images that have a corresponding relationship are two types of images of the same organ or tissue.
[0081] Specifically, after acquiring the morphological image, the terminal can further acquire the functional image corresponding to the morphological image based on the above correspondence.
[0082] S220. Obtain the tissue parameters corresponding to each organ region based on the morphological and functional imaging images.
[0083] Optionally, since the functional imaging image includes tissue parameters of the entire organ tissue, and the positions of each tissue region on the morphological imaging image and the functional imaging image of the same organ tissue correspond, after the terminal divides each organ partition based on the morphological imaging image, it can obtain the tissue parameters corresponding to the corresponding regions of each organ partition on the corresponding functional imaging image, that is, as the tissue parameters corresponding to each organ partition.
[0084] Specifically, such as Figure 3 As shown, the process of obtaining the tissue parameters corresponding to each organ partition specifically includes:
[0085] S310. Map and register the morphological images with the functional images to determine the target areas corresponding to each organ region in the functional images.
[0086] Specifically, the terminal maps and registers the morphological image and the functional image to determine the conversion relationship of pixel coordinates between the morphological image and the functional image, thereby obtaining the coordinate regions of each organ partition in the morphological image, and based on the above conversion relationship, converting the coordinate regions of each organ partition in the morphological image to the coordinate regions in the functional image. Each coordinate region formed in the functional image is the target region corresponding to the organ partition in the functional image.
[0087] S320. Determine the tissue parameters within the target area as the tissue parameters corresponding to each organ partition.
[0088] Specifically, the terminal obtains the tissue parameters of the target region from the functional imaging map and determines that the group parameter data of the target region is the tissue parameter corresponding to the corresponding organ partition.
[0089] In this embodiment, the medical image includes a morphological image. The terminal performs tissue segmentation on the morphological image based on a pre-trained tissue segmentation model to obtain multiple organ partitions, and then obtains the corresponding functional image based on the morphological image. The tissue parameters corresponding to each organ partition are then obtained based on the morphological and functional image. The morphological image clearly presents the boundaries of each tissue region. Dividing the organ partitions based on the morphological image effectively improves the accuracy of the division, and overall improves the accuracy of the final determined feature values.
[0090] In one embodiment, after obtaining the feature values of each organ partition, the method further includes: statistically analyzing the feature values of each organ partition and outputting a structured report.
[0091] The structured report is used to reflect the characteristic value information of each organ region in an organ tissue. Optionally, the structured report can be in tabular form (as shown in Table 1 below), plain text form, or a combination of charts and graphs. This embodiment does not specifically limit the specific presentation format of the structured report.
[0092] Table 1. Structural Report of Liver Tissue
[0093]
[0094] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.
[0095] In one embodiment, such as Figure 4 The present invention provides an apparatus for determining tissue feature values, comprising: an acquisition module 401, a segmentation module 402, and a determination module 403, wherein:
[0096] The acquisition module 401 is used to acquire medical image images;
[0097] The segmentation module 402 is used to segment medical images into multiple organ partitions based on a pre-trained tissue segmentation model.
[0098] The determination module 403 is used to obtain the tissue parameters corresponding to each organ partition and determine the feature values of the organ partition based on the tissue parameters.
[0099] In one embodiment, the medical image includes a morphological image; the segmentation module 402 is specifically used for:
[0100] Based on a pre-trained tissue segmentation model, tissue segmentation is performed on morphological images to obtain multiple organ partitions.
[0101] Accordingly, module 403 is specifically used for:
[0102] Obtain the functional images corresponding to the morphological images; obtain the tissue parameters corresponding to each organ region based on the morphological and functional images.
[0103] In one embodiment, the determining module 403 is specifically used for:
[0104] The morphological images and functional images are mapped and registered to determine the target regions corresponding to each organ region in the functional images; the tissue parameters within the target regions are then determined as the tissue parameters corresponding to each organ region.
[0105] In one embodiment, the medical image map includes a functional image map; the segmentation module 402 is specifically used for:
[0106] Based on the tissue segmentation model, functional images are segmented into multiple organ partitions.
[0107] Accordingly, module 403 is specifically used for:
[0108] Obtain the tissue parameters corresponding to each organ partition based on the tissue parameters of the region where each organ partition is located.
[0109] In one embodiment, the above-described apparatus further includes an output module, which is specifically used for:
[0110] Statistical analysis of feature values for each organ region, and output of a structured report.
[0111] In one embodiment, the medical image is a liver image, and the organ partition includes at least one of the whole liver, liver segments, and liver lesion areas.
[0112] In one embodiment, the functional image map is at least one of a diffusion functional image map, a perfusion functional image map, and a fat functional image map.
[0113] Each module in the aforementioned tissue characteristic value determination device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.
[0114] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a method for determining organizational characteristics. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.
[0115] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0116] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:
[0117] Acquire medical images; perform tissue segmentation on the medical images based on a pre-trained tissue segmentation model to obtain multiple organ partitions; for each organ partition, acquire the corresponding tissue parameters, and determine the feature values of the organ partition based on the tissue parameters.
[0118] In one embodiment, the medical image includes a morphological image; the processor, when executing the computer program, also performs the following steps:
[0119] Based on a pre-trained tissue segmentation model, morphological images are segmented to obtain multiple organ partitions; functional images corresponding to the morphological images are obtained; and tissue parameters corresponding to each organ partition are obtained based on the morphological and functional images.
[0120] In one embodiment, the processor further performs the following steps when executing the computer program:
[0121] Morphological images and functional images are mapped and registered to determine the target regions corresponding to each organ region in the functional images; the tissue parameters within the target regions are then determined as the tissue parameters corresponding to each organ region.
[0122] In one embodiment, the medical image includes a functional image; the processor, when executing the computer program, also performs the following steps:
[0123] Based on the tissue segmentation model, the functional image is segmented into multiple organ partitions; the tissue parameters corresponding to each organ partition are obtained according to the tissue parameters of the region where each organ partition is located.
[0124] In one embodiment, the processor further performs the following steps when executing the computer program:
[0125] Statistical analysis of feature values for each organ region, and output of a structured report.
[0126] In one embodiment, the medical image is a liver image, and the organ partition includes at least one of the whole liver, liver segments, and liver lesion areas.
[0127] In one embodiment, the functional image map is at least one of a diffusion functional image map, a perfusion functional image map, and a fat functional image map.
[0128] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:
[0129] Acquire medical images; perform tissue segmentation on the medical images based on a pre-trained tissue segmentation model to obtain multiple organ partitions; for each organ partition, acquire the corresponding tissue parameters, and determine the feature values of the organ partition based on the tissue parameters.
[0130] In one embodiment, the medical image includes a morphological image; the computer program, when executed by a processor, also performs the following steps:
[0131] Based on a pre-trained tissue segmentation model, morphological images are segmented to obtain multiple organ partitions; functional images corresponding to the morphological images are obtained; and tissue parameters corresponding to each organ partition are obtained based on the morphological and functional images.
[0132] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:
[0133] The morphological images and functional images are mapped and registered to determine the target regions corresponding to each organ region in the functional images; the tissue parameters within the target regions are then determined as the tissue parameters corresponding to each organ region.
[0134] In one embodiment, the medical image includes a functional image; the computer program, when executed by a processor, further performs the following steps:
[0135] Based on the tissue segmentation model, the functional image is segmented into multiple organ partitions; the tissue parameters corresponding to each organ partition are obtained according to the tissue parameters of the region where each organ partition is located.
[0136] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:
[0137] Statistical analysis of feature values for each organ region, and output of a structured report.
[0138] In one embodiment, the medical image is a liver image, and the organ partition includes at least one of the whole liver, liver segments, and liver lesion areas.
[0139] In one embodiment, the functional image map is at least one of a diffusion functional image map, a perfusion functional image map, and a fat functional image map.
[0140] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:
[0141] Acquire medical images; perform tissue segmentation on the medical images based on a pre-trained tissue segmentation model to obtain multiple organ partitions; for each organ partition, acquire the corresponding tissue parameters, and determine the feature values of the organ partition based on the tissue parameters.
[0142] In one embodiment, the medical image includes a morphological image; the computer program, when executed by a processor, also performs the following steps:
[0143] Based on a pre-trained tissue segmentation model, morphological images are segmented to obtain multiple organ partitions; functional images corresponding to the morphological images are obtained; and tissue parameters corresponding to each organ partition are obtained based on the morphological and functional images.
[0144] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:
[0145] The morphological images and functional images are mapped and registered to determine the target regions corresponding to each organ region in the functional images; the tissue parameters within the target regions are then determined as the tissue parameters corresponding to each organ region.
[0146] In one embodiment, the medical image includes a functional image; the computer program, when executed by a processor, further performs the following steps:
[0147] Based on the tissue segmentation model, the functional image is segmented into multiple organ partitions; the tissue parameters corresponding to each organ partition are obtained according to the tissue parameters of the region where each organ partition is located.
[0148] In one embodiment, when the computer program is executed by the processor, it further performs the following steps:
[0149] Statistical analysis of feature values for each organ region, and output of a structured report.
[0150] In one embodiment, the medical image is a liver image, and the organ partition includes at least one of the whole liver, liver segments, and liver lesion areas.
[0151] In one embodiment, the functional image map is at least one of a diffusion functional image map, a perfusion functional image map, and a fat functional image map.
[0152] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.
[0153] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0154] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.
Claims
1. A method of determining a tissue characteristic value, characterized by, The method includes: Acquire medical images of the organ tissues of the object to be analyzed; the medical images are liver images; the medical images include morphological images. A pre-trained tissue segmentation model is used to segment different regions on the organ tissue based on the boundaries of each organ partition in the morphological image, resulting in multiple organ partitions; the multiple organ partitions include at least one of the left lobe of the liver, the right lobe of the liver, the entire liver segment, and the liver lesion region; Obtain the functional image corresponding to the morphological image; the functional image includes at least one of diffusion functional image, perfusion functional image and fat functional image, the tissue parameter included in the diffusion functional image includes diffusion coefficient, the tissue parameter included in the perfusion functional image includes blood flow coefficient or blood flow velocity coefficient, and the tissue parameter included in the fat functional image includes liver fat content or fat fraction. Based on the conversion relationship between the pixel coordinates of the morphological image and the functional image, the coordinate regions of each organ partition in the morphological image are converted to the target regions in the functional image. The tissue parameters within the target area are determined as the tissue parameters corresponding to each organ partition. The tissue parameters at different locations within a preset time period on the corresponding tissue area of the organ partition are compared and processed to obtain feature values. The tissue parameters include the content of the target substance and / or blood flow in the corresponding tissue area.
2. The method of claim 1, wherein, The content of the target substance includes: iron content, oxygen content, and fat content.
3. The method of claim 1, wherein, The blood flow conditions include: blood flow rate and blood flow velocity.
4. The method of claim 1, wherein, The medical imaging image includes a functional imaging image; the method further includes: Based on a pre-trained tissue segmentation model, the functional image is segmented to obtain multiple organ partitions; the multiple organ partitions include at least one of the left lobe of the liver, the right lobe of the liver, the entire liver segment, and the liver lesion area. The tissue parameters corresponding to each organ partition are obtained based on the tissue parameters of the region where each organ partition is located.
5. The method according to claim 4, characterized in that, The tissue segmentation model includes a channel attention module, a deep supervision module, and a contrastive learning module, which are used to enhance the ability to identify organ boundaries.
6. The method according to claim 4, characterized in that, The method further includes: Statistically analyze the feature values of each organ region and output a structured report.
7. A device for determining tissue characteristic values, characterized in that, The device includes: The acquisition module is used to acquire medical images of the organ tissues of the object to be analyzed; the medical images are liver images; the medical images include morphological images. The segmentation module is used to segment different regions on the organ tissue based on the boundaries of each organ partition in the morphological image using a pre-trained tissue segmentation model to obtain multiple organ partitions; the multiple organ partitions include at least one of the left lobe of the liver, the right lobe of the liver, the entire liver segment, and the liver lesion region; A determination module is used to acquire a functional image map corresponding to the morphological image map; the functional image map includes at least one of a diffusion functional image map, a perfusion functional image map, and a fat functional image map, wherein the tissue parameters included in the diffusion functional image map include a diffusion coefficient, the tissue parameters included in the perfusion functional image map include a blood flow coefficient or a blood flow velocity coefficient, and the tissue parameters included in the fat functional image map include the fat content or fat fraction of the liver; based on the conversion relationship between the pixel coordinates of the morphological image map and the functional image map, the coordinate regions of each organ partition in the morphological image map are converted to the target regions in the functional image map; the tissue parameters in the target regions are determined as the tissue parameters corresponding to each organ partition, and the tissue parameters at preset time periods or different locations in the corresponding tissue regions of the organ partitions are compared and processed to obtain feature values; the tissue parameters include the target substance content and / or blood flow status of the corresponding tissue region.
8. The apparatus according to claim 7, characterized in that, The content of the target substance includes: iron content, oxygen content, and fat content.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method for determining tissue characteristic values according to any one of claims 1 to 6.
10. 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 for determining tissue characteristic values as described in any one of claims 1 to 6.