WO2021081841A1 - 基于vrds 4d医学影像的脾脏肿瘤识别方法及相关装置 - Google Patents

基于vrds 4d医学影像的脾脏肿瘤识别方法及相关装置 Download PDF

Info

Publication number
WO2021081841A1
WO2021081841A1 PCT/CN2019/114481 CN2019114481W WO2021081841A1 WO 2021081841 A1 WO2021081841 A1 WO 2021081841A1 CN 2019114481 W CN2019114481 W CN 2019114481W WO 2021081841 A1 WO2021081841 A1 WO 2021081841A1
Authority
WO
WIPO (PCT)
Prior art keywords
spleen
blood vessel
image data
preset
degree
Prior art date
Application number
PCT/CN2019/114481
Other languages
English (en)
French (fr)
Inventor
李斯图尔特平
李戴维伟
Original Assignee
未艾医疗技术(深圳)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 未艾医疗技术(深圳)有限公司 filed Critical 未艾医疗技术(深圳)有限公司
Priority to PCT/CN2019/114481 priority Critical patent/WO2021081841A1/zh
Priority to CN201980099987.8A priority patent/CN114365190A/zh
Publication of WO2021081841A1 publication Critical patent/WO2021081841A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/40—Analysis of texture
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features

Definitions

  • This application relates to the technical field of medical imaging devices, and in particular to a method and related devices for identifying spleen tumors based on VRDS 4D medical images.
  • the spleen is the body's largest immune organ. It is located in the upper left abdomen and accounts for 25% of the total lymphatic tissues of the body. It contains a large number of lymphocytes and macrophages. It is the center of the body's cellular and humoral immunity.
  • CT electronic computer tomography
  • MRI magnetic resonance imaging
  • DTI diffusion tensor imaging
  • PET positron emission computed tomography
  • the embodiments of the present application provide a spleen tumor identification method and related devices based on VRDS 4D medical images, which are beneficial to improve the accuracy and efficiency of spleen tumor identification by medical imaging devices.
  • the first aspect of the embodiments of the present application provides a method for identifying spleen tumors based on VRDS 4D medical imaging, including:
  • Target medical image data includes image data of the spleen and image data of blood vessels around the spleen
  • a second aspect of the embodiments of the present application provides a medical imaging device, including:
  • the acquiring unit is used to acquire a scanned image of the spleen of the target user
  • a processing unit configured to process the scanned image of the spleen to obtain target medical image data, wherein the target medical image data includes image data of the spleen and image data of blood vessels around the spleen;
  • a determining unit configured to determine the tumor characteristics of the spleen according to the image data of the spleen and the image data of the blood vessel;
  • the output unit is configured to perform 4D medical imaging according to the target medical image data and output the abnormal type of the spleen.
  • a third aspect of the embodiments of the present application provides a medical imaging device, including a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and are generated It is executed by the processor to execute the instructions of the steps in any one of the methods of the first aspect of the above claims.
  • the fourth aspect of the embodiments of the present application provides a computer-readable storage medium, where the computer-readable storage medium is used to store a computer program, and the stored computer program is executed by the processor to implement the first aspect of the claims. Any of the methods.
  • a scanned image of the spleen of the target user is acquired, and then the scanned image of the spleen is processed to obtain target medical image data, where the target medical image data includes the image data of the spleen and the blood vessels around the spleen Secondly, determine the tumor characteristics of the spleen based on the image data of the spleen and the image data of the blood vessels. Secondly, identify the abnormal types of the spleen based on the characteristics of the tumor of the spleen. Finally, perform 4D medical imaging based on the target medical image data and output the spleen The type of exception.
  • the medical imaging device in the present application can identify the abnormal type of the spleen by processing the scanned image of the spleen, and output the abnormal type of the spleen, avoiding the situation that the observation based on the human eye is not accurate enough, and is beneficial to improve the medical imaging device to perform the spleen Accuracy and efficiency of tumor recognition.
  • Fig. 1 is a schematic structural diagram of a VRDS 4D medical image intelligent analysis and processing system provided by an embodiment of the application;
  • FIG. 2 is a schematic flowchart of a method for identifying spleen tumors based on VRDS 4D medical imaging according to an embodiment of the application;
  • FIG. 3 is a schematic diagram of a medical imaging device provided by an embodiment of the application.
  • FIG. 4 is a schematic structural diagram of a medical imaging device in a hardware operating environment related to an embodiment of the application.
  • the medical imaging devices involved in the embodiments of this application refer to various instruments that use various media as information carriers to reproduce the internal structure of the human body as images.
  • the image information and the actual structure of the human body have spatial and temporal distributions.
  • DICOM data refers to the original image file data that reflects the internal structural characteristics of the human body collected by medical equipment, which can include electronic computed tomography CT, magnetic resonance MRI, diffusion tensor imaging DTI, and positron emission computed tomography PET-
  • image source refers to the Texture2D/3D image volume data generated by analyzing the original DICOM data.
  • VRDS refers to the Virtual Reality Doctor system (VRDS for short).
  • Figure 1 is a schematic structural diagram of a VRDS-based 4D medical image intelligent analysis and processing system 100 provided by an embodiment of the present application.
  • the system 100 includes a medical imaging device 110 and a network database 120.
  • the medical imaging device 110 can Including the local medical imaging device 111 and/or the terminal medical imaging device 112, the local medical imaging device 111 or the terminal medical imaging device 112 is used to identify spleen tumors based on the VRDS 4D medical image based on the original DICOM data.
  • the recognition, positioning, four-dimensional volume rendering, and abnormal analysis of human spleen tumors are carried out to achieve four-dimensional three-dimensional imaging effects
  • the four-dimensional medical image specifically refers to the medical image including the internal spatial structure characteristics and external spatial structure characteristics of the displayed tissue
  • the internal spatial structure feature refers to that the slice data inside the tissue is not lost, that is, the medical imaging device can present the internal structure of target organs, blood vessels and other tissues
  • the external spatial structure feature refers to the environmental features between tissues, including tissues
  • the characteristics of the spatial position between the tissue and the tissue including intersection, interval, fusion, etc., such as the edge structure characteristics of the intersection position between the spleen and the artery, etc.
  • the local medical imaging device 111 can also be used for the terminal medical imaging device 112 Edit the scanned image to form the transfer function result of the four-dimensional human body image.
  • the transfer function result can include the transfer function result of the surface of the internal organs and the tissue structure in the internal organs of the human body, and the transfer function result of the cube space, as shown in the transfer function.
  • the network database 120 may be, for example, a cloud medical imaging device, etc.
  • the network database 120 is used to store the image source generated by analyzing the original DICOM data and the transfer function result of the four-dimensional human body image edited by the local medical imaging device 111.
  • the scanned image may be from Multiple local medical imaging devices 111 are used to realize interactive diagnosis of multiple doctors.
  • HMDS head-mounted Displays Set
  • the operating actions refer to the user’s actions through the medical imaging device.
  • External ingestion equipment such as mouse, keyboard, tablet (portable android device, Pad), iPad (internet portable apple device), etc., operate and control the four-dimensional human image to achieve human-computer interaction.
  • the operation actions include at least the following One: (1) Change the color and/or transparency of a specific organ/tissue, (2) Position the zoom view, (3) Rotate the view, realize the multi-view 360-degree observation of the four-dimensional human body image, (4) "Enter” Observe the internal structure of human organs, render real-time clipping effects, and (5) move the view up and down.
  • FIG. 2 is a schematic flowchart of a method for identifying spleen tumors based on VRDS 4D medical imaging according to an embodiment of the present application, which is applied to the medical imaging device described in FIG. 1, as shown in FIG. 2, this embodiment
  • the provided spleen tumor identification methods based on VRDS 4D medical images include:
  • the scanned image includes any one of the following: CT image, MRI image, DTI image, PET-CT image.
  • the scanned image also includes an enhanced scanned image after the contrast agent has been injected.
  • the scanned image of the spleen to obtain target medical image data, where the target medical image data includes image data of the spleen and image data of blood vessels around the spleen.
  • processing the scanned image of the spleen to obtain target medical image data includes: generating an image source of the spleen according to the scanned image of the spleen; performing first preset processing on the image source to obtain a bitmap BMP data source; import the BMP data source into the preset VRDS medical network model to obtain the first medical image data, where the first medical image data includes the image data of the spleen and the first image data of the blood vessel; the first medical image data Import the preset cross blood vessel network model to obtain the second medical image data, where the second medical image data includes the image data of the spleen and the second image data of the blood vessel, and the second image data of the blood vessel includes the surface characteristics of the blood vessel, and the second image data of the blood vessel.
  • the surface features are obtained by filtering the smooth muscle and elastic fiber image data in the first image data of the blood vessel through the cross blood vessel network model; the second preset processing is performed on the second medical image data to obtain the target medical image data.
  • generating the image source of the spleen according to the scanned image of the spleen includes: the medical imaging device acquires multiple scanned images that reflect the internal structural features of the target user's human body collected by medical equipment; and from the multiple scanned images Filter out at least one scanned image containing the spleen, and use at least one scanned image as the target user's medical digital imaging and communication DICOM data; analyze the DICOM data to generate the target user's image source, the image source includes Texture 2D/3D image volume data .
  • the first preset processing includes at least one of the following operations: VRDS limited contrast adaptive histogram equalization, hybrid partial differential denoising, and VRDS Ai elastic deformation processing.
  • VRDS limited contrast adaptive histogram equalization includes: regional noise ratio limiting, global contrast limiting; the local histogram of the image source is divided into multiple partitions, for each partition, according to the accumulation of the neighborhood of the partition
  • the slope of the histogram determines the slope of the transformation function, and the degree of contrast magnification around the pixel value of the partition is determined according to the gradient of the transformation function, and then the limit cropping process is performed according to the degree of contrast magnification to generate the distribution of the effective histogram. It also generates effective and available neighborhood consignment values, and evenly distributes these cropped parts of the histogram to other areas of the histogram.
  • hybrid partial differential denoising includes: different from Gaussian low-pass filtering (indiscriminately weakening the high-frequency components of the image, denoising will also produce image edge blurring), the isoilluminance formed by objects in natural images
  • the line (including the edge) should be a smooth and smooth curve, that is, the absolute value of the curvature of these isoilluminance lines should be small enough.
  • the design uses VRDS Ai curvature drive and VRDS Ai high-order hybrid denoising to protect the edges of the image and avoid the step effect in the smoothing process.
  • the hybrid partial differential denoising model is used.
  • the VRDS Ai elastic deformation processing includes: superimposing positive and negative random distances on the original lattice to form a difference position matrix, and then the grayscale at each difference position forms a new lattice, which can realize the distortion of the image. Deform, and then rotate, distort, and translate the image.
  • the medical imaging device obtains the BMP data source by processing the original scanned image data, which increases the amount of information of the original data, and increases the depth information, and finally obtains data that meets the requirements of 4D medical image display.
  • the first image data of the blood vessel includes the fusion data of the intersection position of the artery and the vein
  • the second image data of the blood vessel includes the surface feature of the blood vessel.
  • the surface feature of the blood vessel is selected from the first image data of the blood vessel through the cross-vessel network model.
  • the smooth muscle and elastic fiber image data of the blood vessel are obtained.
  • the surface features of the blood vessel also include the data set of the artery and the data set of the vein, and the first data in the arterial data set and the second data in the vein data set are independent of each other.
  • the first data The second data is data associated with the intersection position
  • the second data is the data associated with the intersection position.
  • the VRDS medical network model is provided with the transfer function of the structural characteristics of the spleen and the transfer function of the structural characteristics of the blood vessel
  • the BMP data source obtains the first medical image data through the processing of the transfer function
  • the cross-vessel network model realizes the artery through the following operations
  • Data separation from veins (1) Extract the fusion data at the intersection; (2) Separate the fusion data based on a preset data separation algorithm for each fusion data to obtain independent arterial boundary point data and vein boundary point data; 3) Integrating multiple arterial boundary point data obtained after processing into the first data, and integrating multiple venous boundary point data obtained after processing into the second data.
  • the second preset processing includes at least one of the following operations: 2D boundary optimization processing, 3D boundary optimization processing, and data enhancement processing.
  • the 2D boundary optimization processing includes: multiple sampling to obtain low-resolution information and high-resolution information.
  • the low-resolution information can provide the contextual semantic information of the segmentation target in the entire image, that is, the characteristics that reflect the relationship between the target and the environment. These features are used to determine the object category, and the high-resolution information is used to provide more refined features, such as gradients, for the segmentation target.
  • the 3D boundary optimization processing includes: 3D convolution, 3D max pooling, and 3D upward convolution layer, the input data size is a1, a2, a3, the number of channels is c, the filter size is f, that is, the filter dimension is f*f*f*c, the number of filters is n, the final output of the 3-dimensional convolution is:
  • each layer contains two 3*3*3 convolution kernels, each of which is followed by an activation function (Relu), and then there is a maximum pooling of 2*2*2 in each dimension to merge the two Steps.
  • each layer is composed of 2*2*2 upward convolution, with a step size of 2 in each dimension, and then two 3*3*3 convolutions, and then Relu. Then in the analysis path, the shortcut connections of equal resolution layers provide the basic high-resolution features of the synthesized path. In the last layer, 1*1*1 convolution reduces the number of output channels.
  • the data enhancement processing includes any one of the following: data enhancement based on arbitrary angle rotation, data enhancement based on histogram equalization, data enhancement based on white balance, data enhancement based on mirroring operation, data enhancement based on random cut, and data enhancement based on Data enhancement to simulate different lighting changes.
  • the medical imaging device can process the BMP data source through the VRDS medical network model and the cross blood vessel network model, and combine boundary optimization and data enhancement processing to obtain target image data, which solves the problem that traditional medical imaging cannot achieve segmentation of arteries and arteries.
  • the overall separation of veins and the problems in the medical field where the surface features of blood vessels cannot be extracted have improved the authenticity, comprehensiveness and refinement of medical image display.
  • determining the tumor characteristics of the spleen according to the imaging data of the spleen and the imaging data of the blood vessel includes: determining the characteristics of the spleen according to the imaging data of the spleen and the imaging data of the blood vessel Tumor location; acquiring the characteristics of the spleen from the image data of the spleen, wherein the characteristics of the spleen include at least one of the following data: the size of the spleen, the size of the lymph nodes in the spleen, the lymph nodes in the spleen The degree of expansion of the tube; the comprehensive degree of expansion of the blood vessel is determined according to the image data of the blood vessel.
  • the image data of the blood vessel includes the diameter of the blood vessel and the surface characteristics of the blood vessel
  • determining the comprehensive expansion degree of the blood vessel according to the image data of the blood vessel includes: determining the surface expansion degree of the blood vessel according to the surface characteristics of the blood vessel; and acquiring the target
  • the physical parameters of the user where the physical parameters of the target user include at least one of the following: the height, weight, blood pressure, blood sugar, and heart rate of the target user; the first weight corresponding to the diameter of the blood vessel is determined according to the physical parameters of the target user and the value of the blood vessel
  • the second weight corresponding to the degree of surface expansion; the diameter of the blood vessel and the degree of surface expansion of the blood vessel are respectively weighted according to the first weight and the second weight to obtain the comprehensive expansion degree of the blood vessel.
  • the diameter of the blood vessel can be obtained in the following way: the blood vessel is segmented according to the first image data of the blood vessel to obtain the M-segment blood vessel, where M is a positive integer; the average diameter of each blood vessel in the M-segment blood vessel is determined, Among them, the difference between the average diameter of the i-th blood vessel and the average diameter of the (i+1)-th blood vessel is not less than the preset segment threshold, and i is a positive integer less than M; determine each segment of the M-segment blood vessel The weight corresponding to the blood vessel; the average diameter of each blood vessel in the M blood vessel is weighted according to the weight corresponding to each blood vessel in the M blood vessel to obtain the diameter of the blood vessel.
  • the determining the degree of surface expansion of the blood vessel according to the surface characteristics of the blood vessel includes: determining the target surface area of the blood vessel according to the surface characteristics of the blood vessel, obtaining the first surface feature of the target surface area of the blood vessel; obtaining the normal blood vessel The second surface feature of the target surface area; the first surface feature is compared with the second surface feature to determine the degree of surface expansion of the blood vessel.
  • the determining the target surface area of the blood vessel according to the surface characteristics of the blood vessel includes: analyzing the feature point distribution of the surface area of the blood vessel according to the surface characteristics of the blood vessel; Perform circular image interception with different centers to obtain N circular surface partitions, where N is an integer greater than 3; determine the number of feature points contained in each circular surface partition in the N circular surface partitions; A target circular surface partition is selected from the N circular surface partitions, wherein the number of feature points included in the target circular surface partition is greater than that included in other circular surface partitions in the N circular surface partitions The number of feature points.
  • the surface characteristics of the target surface area are compared with the surface characteristics of the normal blood vessel.
  • the complexity of feature comparison can be reduced, the comparison time can be shortened, and the comparison efficiency can be improved.
  • the image data of the blood vessel also includes the curvature of the blood vessel
  • the method for determining the curvature of the blood vessel may be:
  • the origin of the coordinate system is any position of the blood vessel, and the X axis, Y axis and Z axis of the coordinate system are perpendicular to each other and follow the right-hand spiral rule;
  • the spatial position corresponding to the first pixel is recorded, and whenever the first pixel is detected.
  • the gray value corresponding to the two pixel points does not belong to the gray value corresponding to the blood vessel cell data of the outermost layer of the blood vessel, and the gray value corresponding to the adjacent pixels of the second pixel point belongs to the blood vessel of the outermost layer of the blood vessel
  • the spatial position corresponding to the second pixel point is recorded;
  • the second image data of the blood vessel is segmented according to the spatial positions corresponding to all the first pixels and the spatial positions corresponding to all the second pixels, so as to obtain a plurality of outermost parts corresponding to a plurality of blood vessels.
  • Layer vascular cell data set each outermost vascular cell data set includes multiple outermost vascular cell data;
  • the characteristic curve projected on any plane of the currently processed outermost vascular cell data set select any point on the characteristic curve as the starting point; starting from the starting point, follow the positive and negative directions of the characteristic curve
  • the direction of the pixel points is continuously marked, and the marking is stopped when the target pixel point is marked.
  • the positive direction of the characteristic curve is the lateral positive direction of the second image data of the blood vessel
  • the reverse direction of the characteristic curve is the second image data of the blood vessel.
  • the target pixel is the pixel with the largest change in the curvature of the target blood vessel segment
  • the target blood vessel segment is the blood vessel between the starting point and the target spatial position of the target blood vessel
  • the target blood vessel Corresponding to the currently processed outermost blood vessel cell data set, the target spatial position is the position corresponding to the target pixel; acquiring the curvature corresponding to the target blood vessel segment; setting the curvature corresponding to the target blood vessel segment to be The target blood vessel corresponds to the degree of curvature.
  • the image data of the blood vessel includes the diameter of the blood vessel, the surface characteristics of the blood vessel and the curvature of the blood vessel
  • determining the comprehensive expansion degree of the blood vessel according to the image data of the blood vessel includes: determining the surface of the blood vessel according to the surface characteristics of the blood vessel Expansion degree; obtain the physical parameters of the target user, where the physical parameters of the target user include at least one of the following: the height, weight, blood pressure, blood sugar, and heart rate of the target user; determine the first corresponding to the diameter of the blood vessel according to the physical parameters of the target user Three weights, the fourth weight corresponding to the degree of surface expansion of the blood vessel, and the fifth weight corresponding to the degree of curvature of the blood vessel; according to the third weight, the fourth weight, and the fifth weight, the diameter of the blood vessel, the degree of surface expansion of the blood vessel, and the blood vessel The degree of curvature is weighted to obtain the comprehensive degree of expansion of the blood vessel.
  • the spleen is an immune organ with abundant blood supply and a low incidence of tumors. Benign tumors of the spleen are more common. Hemangiomas are the most common. Malignant tumors of the spleen are most common. Malignant lymphomas are the most common. Benign tumors of the spleen include splenic cysts, hemangioma, and lymphatic vessels. Tumors and splenic hamartomas. Malignant tumors of the spleen include angiosarcoma, lymphoma, and metastases. Benign tumors of the spleen are generally more regular in shape, with clear boundaries, and the outline of the spleen outside the lesion is relatively complete. Malignant tumors of the spleen generally have enlarged spleen , And the lymph nodes are also swollen.
  • splenic cyst is a tumor-like cystic lesion of the spleen tissue, which can be clinically divided into parasitic cysts and non-parasitic cysts, with no blood vessels and clear boundaries;
  • Hemangiomas are the most common type of benign tumors of the spleen. Cavernous hemangioma is the most common type of splenic hemangioma, which can be divided into cavernous hemangioma and capillary hemangioma according to the degree of vasodilation, and they can also be mixed with clear boundaries;
  • Lymphangioma is formed by obstruction and expansion of lymphatic vessels.
  • the pathology is composed of expanded lymphatic sinuses. It is divided into capillary lymphoma, cavernous lymphangioma and cystic lymphangioma, with clear boundaries;
  • Splenic hamartoma including white pulp splenic hamartoma, red pulp splenic hamartoma, and mixed splenic hamartoma.
  • White pulp splenic hamartoma is composed of abnormal lymphoid tissue
  • red pulp splenic hamartoma is composed of Disordered splenic sinus composition and mixed splenic hamartoma have both, with clear boundaries;
  • Lymphoma malignant tumor, manifested in splenic hilar area, aortic lymphadenopathy, severe splenomegaly, unclear borders, and medium-to-low enhancement after enhanced scan;
  • Angiosarcoma malignant tumor, splenic mass and splenic enlargement, lymph node enlargement, unclear borders, and medium-to-high enhancement after enhanced scan;
  • Metastases late stage malignant tumors, swollen lymph nodes, more simultaneous occurrence of metastases in the liver, unclear borders, and bull's eye sign or bull's eye sign after enhanced scan.
  • the identifying the abnormal type of the spleen according to the tumor characteristics of the spleen includes: when the size of the spleen does not exceed a first preset threshold and the size of lymph nodes does not exceed a second preset threshold, according to The tumor position of the spleen determines the definition of the tumor boundary of the spleen; judges whether the definition of the tumor boundary of the spleen exceeds the preset definition threshold; if the definition of the tumor boundary of the spleen exceeds the preset definition threshold, the abnormal type of the spleen is identified as a splenic cyst ; If the sharpness of the tumor boundary of the spleen does not exceed the preset sharpness threshold, judge whether the comprehensive expansion degree of the blood vessel exceeds the preset vascular expansion threshold; if the comprehensive expansion degree of the blood vessel exceeds the preset vascular expansion threshold, identify the abnormality of the spleen
  • the type is hemangioma.
  • lymphangioma If the comprehensive expansion of blood vessels does not exceed the preset threshold of vasodilatation, it is determined whether the expansion of lymphatic vessels exceeds the preset threshold of lymphatic expansion. If the expansion of lymphatics exceeds the preset threshold of lymphatic expansion Threshold, the abnormal type of spleen is identified as lymphangioma.
  • the size of the spleen does not exceed the threshold of normal spleen size and the size of lymph nodes does not exceed the threshold of normal lymph node size, it can be preliminarily judged to be benign tumors.
  • Benign tumors of the spleen include splenic cysts, hemangioma, lymphangioma, etc.
  • whether it is a splenic cyst can be determined by judging the definition of the tumor boundary of the spleen, whether it is a hemangioma by judging the degree of comprehensive expansion of blood vessels, and whether it is a lymphangioma by judging the degree of expansion of lymphatic vessels.
  • the target medical image data also includes enhanced medical image data obtained by processing an enhanced scan image of the spleen.
  • the image data determines the degree of tumor enhancement of the spleen after the enhanced scan; judges whether the degree of tumor enhancement of the spleen exceeds the preset threshold of enhancement; if the degree of tumor enhancement of the spleen exceeds the preset threshold of enhancement, the abnormal type of the spleen is identified as angiosarcoma; if If the enhancement degree of the spleen tumor does not exceed the preset enhancement degree threshold, the abnormal type of the spleen is identified as lymphoma.
  • the size of the spleen exceeds the threshold of normal spleen size or the size of lymph nodes exceeds the threshold of normal lymph node size, it can be preliminarily judged as malignant tumors.
  • Malignant tumors of the spleen include angiosarcoma, lymphoma, and metastases.
  • the degree of tumor enhancement can further determine the abnormal type of the spleen. If the enhancement degree of the tumor of the spleen exceeds the preset enhancement degree threshold, the abnormal type of the spleen is identified as angiosarcoma; if the tumor enhancement degree of the spleen does not exceed the preset enhancement threshold, the abnormal type of the spleen is identified as lymphoma.
  • 4D medical imaging refers to the presentation of 4-dimensional medical images.
  • performing 4D medical imaging according to the target medical image data includes: the medical imaging device selects enhancement data with a quality score greater than a preset score from the target medical image data as VRDS 4D imaging data; according to VRDS 4D imaging The data undergoes 4D medical imaging.
  • the quality score can be comprehensively evaluated from the following dimensions: average gradient, information entropy, visual information fidelity, peak signal-to-noise ratio PSNR, structural similarity SSIM, mean square error MSE, etc.
  • average gradient information entropy
  • visual information fidelity visual information fidelity
  • peak signal-to-noise ratio PSNR peak signal-to-noise ratio
  • structural similarity SSIM structural similarity
  • mean square error MSE mean square error
  • a scanned image of the spleen of the target user is acquired, and then the scanned image of the spleen is processed to obtain target medical image data.
  • the target medical image data includes the image data of the spleen and the surrounding image data of the spleen.
  • the imaging data of blood vessels secondly, determine the tumor characteristics of the spleen according to the imaging data of the spleen and the imaging data of the blood vessels, secondly, identify the abnormal types of the spleen according to the tumor characteristics of the spleen, and finally, perform 4D medical imaging based on the target medical imaging data and output Abnormal type of spleen.
  • the medical imaging device in the present application can identify the abnormal type of the spleen by processing the scanned image of the spleen, and output the abnormal type of the spleen, avoiding the situation that the observation based on the human eye is not accurate enough, and is beneficial to improve the medical imaging device to perform the spleen Accuracy and efficiency of tumor recognition.
  • FIG. 3 is a schematic diagram of a medical imaging apparatus 300 provided by an embodiment of the application.
  • the medical imaging apparatus 300 may include:
  • the obtaining unit 301 is configured to obtain a scanned image of the spleen of the target user
  • the processing unit 302 is configured to process the scanned image of the spleen to obtain target medical image data, where the target medical image data includes image data of the spleen and image data of blood vessels around the spleen;
  • the determining unit 303 is configured to determine the tumor characteristics of the spleen according to the image data of the spleen and the image data of the blood vessel;
  • the identification unit 304 is configured to identify the abnormal type of the spleen according to the characteristics of the tumor of the spleen;
  • the output unit 305 is configured to perform 4D medical imaging according to the target medical image data, and output the abnormal type of the spleen.
  • the processing unit 302 is specifically configured to: generate the image source of the spleen according to the scanned image of the spleen; perform the first preset processing on the image source to obtain the bitmap BMP data source;
  • the BMP data source imports a preset VRDS medical network model to obtain first medical image data, where the first medical image data includes the image data of the spleen and the first image data of the blood vessel;
  • the first medical image data is imported into the preset cross blood vessel network model to obtain the second medical image data, wherein the second medical image data includes the image data of the spleen and the second image data of the blood vessel.
  • the second image data includes the surface features of the blood vessel, and the surface features of the blood vessel are obtained by filtering the smooth muscle and elastic fiber image data in the first image data of the blood vessel through the cross blood vessel network model;
  • the medical image data executes a second preset processing to obtain the target medical image data.
  • the determining unit 303 is specifically configured to: determine the location of the tumor of the spleen according to the image data of the spleen and the image data of the blood vessel; and obtain the spleen from the image data of the spleen.
  • the characteristics of the spleen include at least one of the following data: the size of the spleen, the size of the lymph nodes in the spleen, and the degree of expansion of the lymph vessels in the spleen; The degree of comprehensive expansion of blood vessels.
  • the image data of the blood vessel includes the diameter of the blood vessel and the surface characteristics of the blood vessel
  • the determining unit 303 is specifically configured to: determine the surface of the blood vessel according to the surface characteristics of the blood vessel. Degree of expansion; acquiring the physical parameters of the target user, wherein the physical parameters of the target user include at least one of the following: the height, weight, blood pressure, blood sugar, and heart rate of the target user; according to the physical parameters of the target user
  • the first weight corresponding to the diameter of the blood vessel and the second weight corresponding to the degree of surface expansion of the blood vessel are determined; the diameter of the blood vessel and the blood vessel are respectively determined according to the first weight and the second weight. Perform weighting calculation on the surface expansion degree of, to obtain the comprehensive expansion degree of the blood vessel.
  • the medical imaging device 300 further includes a tube diameter acquiring unit 306, and the tube diameter acquiring unit 306 is configured to segment the blood vessel according to the first image data of the blood vessel to obtain M A segment of blood vessel, where M is a positive integer; determine the average diameter of each segment of the M-segment blood vessel, where the average diameter of the i-th blood vessel and the average diameter of the (i+1)-th blood vessel The diameter difference is not less than the preset segmentation threshold, and i is a positive integer less than M; the weight corresponding to each segment of the M blood vessel is determined; the M segment is determined according to the weight corresponding to each segment of the M blood vessel The average diameter of each segment of the blood vessel in the blood vessel is weighted to obtain the diameter of the blood vessel.
  • the determining unit 303 is specifically configured to: determine the target surface area of the blood vessel according to the surface characteristics of the blood vessel, obtain the first surface feature of the target surface area of the blood vessel; The second surface feature of the target surface area; the first surface feature is compared with the second surface feature to determine the degree of surface expansion of the blood vessel.
  • the determining unit 303 is specifically configured to: analyze the feature point distribution of the surface area of the blood vessel according to the surface characteristics of the blood vessel; and round the surface area of the blood vessel according to N different circle centers. Image interception to obtain N circular surface partitions, where N is an integer greater than 3; determine the number of feature points contained in each circular surface partition in the N circular surface partitions; from the N circular surface partitions A target circular surface partition is selected from the surface partitions, wherein the number of feature points included in the target circular surface partition is greater than the number of feature points included in other circular surface partitions in the N circular surface partitions .
  • the identification unit 304 is specifically configured to: when the size of the spleen does not exceed a first preset threshold and the size of the lymph node does not exceed a second preset threshold, according to the tumor of the spleen The position determines the definition of the tumor boundary of the spleen; it is determined whether the definition of the tumor boundary of the spleen exceeds a preset definition threshold; if the definition of the tumor boundary of the spleen exceeds the preset definition threshold, the The abnormal type of the spleen is splenic cyst; if the sharpness of the tumor boundary of the spleen does not exceed the preset sharpness threshold, it is determined whether the comprehensive expansion degree of the blood vessel exceeds the preset vascular expansion threshold, and if the comprehensive expansion of the blood vessel If the degree of expansion exceeds the preset threshold of vasodilatation degree, the abnormal type of the spleen is identified as an hemangioma, and if the comprehensive expansion degree of the blood vessel does not exceed
  • the target medical image data further includes enhanced medical image data obtained by processing an enhanced scan image of the spleen
  • the identification unit 304 is further configured to: when the size of the spleen exceeds the size of the spleen When the first preset threshold or the size of the lymph node exceeds the second preset threshold, determine the degree of tumor enhancement of the spleen after the enhanced scan according to the enhanced medical image data; determine whether the degree of tumor enhancement of the spleen exceeds A preset enhancement degree threshold; if the enhancement degree of the spleen tumor exceeds the preset enhancement degree threshold, the abnormal type of the spleen is identified as angiosarcoma; if the tumor enhancement degree of the spleen does not exceed the preset enhancement
  • the degree threshold is used to identify the abnormal type of the spleen as lymphoma.
  • FIG. 4 is a schematic structural diagram of a medical imaging device in a hardware operating environment involved in an embodiment of the application.
  • the medical imaging device in the hardware operating environment involved in the embodiment of the present application may include:
  • the processor 401 such as a CPU.
  • the memory 402 optionally, the memory may be a high-speed RAM memory, or a stable memory, such as a disk memory.
  • the communication interface 403 is used to implement connection and communication between the processor 401 and the memory 402.
  • FIG. 4 does not constitute a limitation, and may include more or less components than shown in the figure, or a combination of certain components, or different component arrangements. .
  • the memory 402 may include an operating system, a network communication module, and a program for spleen tumor recognition.
  • the operating system is a program that manages and controls the hardware and software resources of the medical imaging device, supports the operation of the spleen tumor recognition program and other software or programs.
  • the network communication module is used to implement communication between various components in the memory 402 and communication with other hardware and software in the medical imaging device.
  • the processor 401 is configured to execute the spleen tumor recognition program stored in the memory 402, and implement the following steps:
  • Target medical image data includes image data of the spleen and image data of blood vessels around the spleen
  • the present application also provides a computer-readable storage medium, the computer-readable storage medium is used to store a computer program, and the stored computer program is executed by the processor to implement the following steps:
  • Target medical image data includes image data of the spleen and image data of blood vessels around the spleen
  • the disclosed device can be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the modules is only a logical function division, and there may be other divisions in actual implementation, for example, multiple modules or components may be combined or may be Integrate into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or modules, and may be in electrical or other forms.
  • modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist alone physically, or two or more modules may be integrated into one module.
  • the above-mentioned integrated modules can be implemented in the form of hardware or software function modules.
  • the integrated module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solution of this application essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage
  • the medium includes a number of instructions to enable a computer device (which may be a personal computer, a medical imaging device, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned storage media include: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

一种基于VRDS 4D医学影像的脾脏肿瘤识别方法及相关装置，所述方法包括：获取目标用户的脾脏的扫描图像（201）；对所述脾脏的扫描图像进行处理，以得到目标医学影像数据，其中，所述目标医学影像数据包括所述脾脏的影像数据和所述脾脏周围的血管的影像数据（202）；根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤特性（203）；根据所述脾脏的肿瘤特性识别所述脾脏的异常类型（204）；根据所述目标医学影像数据进行4D医学成像，并输出所述脾脏的异常类型（205）。所述方法提高了医学成像装置进行脾脏肿瘤识别的准确度和效率。

Description

基于VRDS 4D医学影像的脾脏肿瘤识别方法及相关装置 技术领域
本申请涉及医学成像装置技术领域，具体涉及一种基于VRDS 4D医学影像的脾脏肿瘤识别方法及相关装置。
背景技术
脾脏是机体最大的免疫器官，位于左上腹部，占全身淋巴组织总量的25％，含有大量的淋巴细胞和巨噬细胞，是机体细胞免疫和体液免疫的中心。
目前，医生仍然采用观看阅读连续的二维切片扫描图像，例如，CT(电子计算机断层扫描)、MRI(磁共振成像)、DTI(弥散张量成像)、PET(正电子发射型计算机断层显像)等，以此对患者的脾脏肿瘤进行判断分析。然而，脾脏位于膈下，被周围的骨骼保护，所以脾脏肿瘤的早期症状不明显，不容易被发现，仅仅通过直接观看两维切片数据有时无法识别出肿瘤，严重影响到医生对脾脏肿瘤的诊断，从而延误了脾脏肿瘤的治疗。随着医学成像技术的飞速发展，人们对医学成像提出了新的需求。
发明内容
本申请实施例提供了一种基于VRDS 4D医学影像的脾脏肿瘤识别方法及相关装置，有利于提高医学成像装置进行脾脏肿瘤识别的准确度和效率。
本申请实施例第一方面提供了基于VRDS 4D医学影像的脾脏肿瘤识别方法，包括：
获取目标用户的脾脏的扫描图像；
对所述脾脏的扫描图像进行处理，以得到目标医学影像数据，其中，所述目标医学影像数据包括所述脾脏的影像数据和所述脾脏周围的血管的影像数据；
根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤特性；
根据所述脾脏的肿瘤特性识别所述脾脏的异常类型；
根据所述目标医学影像数据进行4D医学成像，并输出所述脾脏的异常类型。
本申请实施例第二方面提供了一种医学成像装置，包括：
获取单元，用于获取目标用户的脾脏的扫描图像；
处理单元，用于对所述脾脏的扫描图像进行处理，以得到目标医学影像数据，其中，所述目标医学影像数据包括所述脾脏的影像数据和所述脾脏周围的血管的影像数据；
确定单元，用于根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤特性；
识别单元，用于根据所述脾脏的肿瘤特性识别所述脾脏的异常类型；
输出单元，用于根据所述目标医学影像数据进行4D医学成像，并输出所述脾脏的异常类型。
本申请实施例第三方面提供了一种医学成像装置，包括处理器、存储器、通信接口以及一个或多个程序，其中，所述一个或多个程序被存储在所述存储器中，并且被生成由所述处理器执行，以执行权利要求上述第一方面任一项方法中的步骤的指令。
本申请实施例第四方面提供了一种计算机可读存储介质，所述计算机可读存储介质用于存储计算机程序，所述存储计算机程序被所述处理器执行，以实现权利要求上述第一方面任一项所述的方法。
可以看出，上述技术方案中，获取目标用户的脾脏的扫描图像，接着对脾脏的扫描图像进行处理，以得到目标医学影像数据，其中，目标医学影像数据包括脾脏的影像数据和脾脏周围的血管的影像数据，其次，根据脾脏的影像数据和血管的影像数据确定脾脏的肿瘤特性，其次，根据脾脏的肿瘤特性识别脾脏的异常类型，最后，根据目标医学影像数据进行4D医学成像，并输出脾脏的异常类型。可见，本申请中的医学成像装置能够通过处理脾脏的扫描图像，识别脾脏的异常类型，并输出该脾脏的异常类型，避免了基于人眼观察不够精准的情况，有利于提高医学成像装置进行脾脏肿瘤识别的准确度和效率。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
其中：
图1为本申请实施例提供的一种基于VRDS 4D医学影像智能分析处理系统的结构示意图；
图2为本申请实施例提供的一种基于VRDS 4D医学影像的脾脏肿瘤识别方法的流程示意图；
图3为本申请实施例提供的一种医学成像装置的示意图；
图4为本申请的实施例涉及的硬件运行环境的医学成像装置结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
以下分别进行详细说明。
本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”、“第三”是用于区别不同对象，而不是用于描述特定顺序。此外，术语“包括”和“具有”以及它们任何变形，意图在于覆盖不排他的包含。
本申请实施例所涉及到的医学成像装置是指利用各种不同媒介作为信息载体，将人体内部的结构重现为影像的各种仪器，其影像信息与人体实际结构有着空间和时间分布上的对应关系。“DICOM数据”是指通过医疗设备采集的反映人体内部结构特征的原始图像文件数据，可以包括电子计算机断层扫描CT、核磁共振MRI、弥散张量成像DTI、正电子发射型计算机断层显像PET-CT等信息，“图源”是指解析原始DICOM数据生成的Texture2D/3D图像体数据。“VRDS”是指虚拟现实医用系统(Virtual Reality Doctor system，简称为VRDS)。
首先，参见图1，图1是本申请实施例提供的一种基于VRDS 4D医学影像智能分析处理系统100的结构示意图，该系统100包括医学成像装置110和网络数据库120，其中医学成像装置110可以包括本地医学成像装置111和/或终端医学成像装置112，本地医学成像装置111或终端医学成像装置112用于基于原始DICOM数据，以本申请实施例所呈现的基于VRDS 4D医学影像的脾脏肿瘤识别方法为基础，进行人体脾脏肿瘤的识别、定位和四维体绘制、异常分析，实现四维立体成像效果(该4维医学影像具体是指医学影像包括所显示组织的内部空间结构特征及外部空间结构特征，所述内部空间结构特征是指组织内部的切片数据未丢失，即医学成像装置可以呈现目标器官、血管等组织的内部构造，外部空间结构特性是指组织与组织之间的环境特征，包括组织与组织之间的空间位置特性(包括交叉、间隔、融合)等，如脾脏与动脉之间的交叉位置的边缘结构特性等)，本地医学成像装置111相对于终端医学成像装置112还可以用于对扫描图像进行编辑，形成四维人体图像的传递函数结果，该传递函数结果可以包括人体内脏器官表面和人体内脏器官内的组织结构的传递函数结果，以及立方体空间的传递函数结果，如传递函数所需的立方编辑框与弧线编辑的数组数量、坐标、颜色、透明度等信息。网络数据库120例如可以是云医学成像装置等，该网络数据库120用于存储解析原始DICOM数据生成的图源，以及本地医学成像装置111编辑得到的四维人体图像的传递函数结果，扫描图像可以是来自于多个本地医学成像装置111以实现多个医生的交互诊断。
用户通过上述医学成像装置110进行具体的图像显示时，可以选择显示器或者虚拟现实VR的头戴式显示器(Head mounted Displays Set，HMDS)结合操作动作进行显示，操作动作是指用户通过医学成像装置的外部摄入设备，如鼠标、键盘、平板电脑(portable android device，Pad)、iPad(internet portable apple device)等，对四维人体图像进行的操作控制，以实现人机交互，该操作动作包括以下至少一种：(1)改变某个具体器官/组织的颜色和/或透明度，(2)定位缩放视图，(3)旋转视图，实现四维人体图像的多视角360度观 察，(4)“进入”人体器官内部观察内部构造，实时剪切效果渲染，(5)上下移动视图。
下面对本申请实施例涉及到的基于VRDS 4D医学影像的脾脏肿瘤识别方法进行详细介绍。
参见图2，图2是本申请实施例提供的一种基于VRDS 4D医学影像的脾脏肿瘤识别方法的流程示意图，应用于如图1所述的医学成像装置，如图2所示，本实施例提供的基于VRDS 4D医学影像的脾脏肿瘤识别方法包括：
201、获取目标用户的脾脏的扫描图像。
其中，扫描图像包括以下任意一种：CT图像、MRI图像、DTI图像、PET-CT图像。
在一个可能的示例中，扫描图像还包括打入对比剂后的增强扫描图像。
202、对所述脾脏的扫描图像进行处理，以得到目标医学影像数据，其中，所述目标医学影像数据包括所述脾脏的影像数据和所述脾脏周围的血管的影像数据。
在一种可能的示例中，对所述脾脏的扫描图像进行处理，以得到目标医学影像数据，包括：根据脾脏的扫描图像生成脾脏的图源；针对图源执行第一预设处理得到位图BMP数据源；将BMP数据源导入预设的VRDS医学网络模型，得到第一医学影像数据，其中，第一医学影像数据包括脾脏的影像数据和血管的第一影像数据；将第一医学影像数据导入预设的交叉血管网络模型，得到第二医学影像数据，其中，第二医学影像数据包括脾脏的影像数据和血管的第二影像数据，血管的第二影像数据包括血管的表面特征，血管的表面特征通过交叉血管网络模型筛选血管的第一影像数据中的平滑肌和弹性纤维图像数据得到；针对第二医学影像数据执行第二预设处理得到目标医学影像数据。
可选的，根据所述脾脏的扫描图像生成所述脾脏的图源，包括：医学成像装置获取通过医疗设备采集的反映目标用户的人体内部结构特征的多张扫描图像；从多张扫描图像中筛选出包含脾脏的至少一张扫描图像，将至少一张扫描图像作为目标用户的医学数字成像和通信DICOM数据；解析DICOM数据生成目标用户的图源，图源包括纹理Texture 2D/3D图像体数据。
可选的，所述第一预设处理包括以下至少一种操作：VRDS限制对比度自适应直方图均衡、混合偏微分去噪、VRDS Ai弹性变形处理。
其中，VRDS限制对比度自适应直方图均衡包括：区域噪音比度限幅、全局对比度限幅；将图源的局部直方图爱划分多个分区，针对每个分区，根据该分区的邻域的累积直方图的斜度确定变换函数的斜度，根据该变换函数的斜度确定该分区的像素值周边的对比度放大程度，然后根据该对比度放大程度进行限度裁剪处理，产生有效直方图的分布，同时也产生有效可用的邻域代销的取值，将这些裁剪掉的部分直方图均匀的分布到直方图的其他区域。
其中，混合偏微分去噪包括：不同于高斯低通滤波(对图像的高频成分不加区别的减弱，去噪的同时会产生图像边缘模糊化)，自然图像中的物体所形成的等照度线(包括边缘) 应该是足够光滑顺畅的曲线，即这些等照度线的曲率的绝对值应该足够小，当图像受到噪声污染后，图像的局部灰度值会发生随机起伏，导致等照度线的不规则震荡，形成局部曲率很大的等照度线，根据这一原理，设计通过VRDS Ai曲率驱动和VRDS Ai高阶混合去噪，实现即可保护图像边缘、又可以避免平滑过程中出现阶梯效应的混合偏微分去噪模型。
其中，VRDS Ai弹性变形处理包括：在原有点阵上，叠加正负向随机距离形成差值位置矩阵，然后在每个差值位置上的灰度，形成新的点阵，可以实现图像内部的扭曲变形，另外再对图像进行旋转、扭曲、平移等操作。
可见，本示例中，医学成像装置通过对原始扫描图像数据的处理，得到BMP数据源，提高了原始数据的信息量，且增加了深度信息，最终得到符合4D医学影像显示需求的数据。
可选的，血管的第一影像数据包括动脉和静脉的交叉位置的融合数据，血管的第二影像数据包括血管的表面特征，血管的表面特征通过交叉血管网络模型筛选血管的第一影像数据中的平滑肌和弹性纤维图像数据得到，血管的表面特征还包括动脉的数据集合以及静脉的数据集合，且动脉的数据集合中的第一数据和静脉的数据集合的第二数据相互独立，第一数据为与交叉位置关联的数据，第二数据为与交叉位置关联的数据。
其中，VRDS医学网络模型设置有脾脏的结构特性的传递函数和血管的结构特性的传递函数，BMP数据源通过传递函数的处理得到第一医学影像数据，所述交叉血管网络模型通过以下操作实现动脉和静脉的数据分离：(1)提取交叉位置的融合数据；(2)针对每个融合数据基于预设数据分离算法分离该融合数据，得到相互独立的动脉边界点数据和静脉边界点数据；(3)将处理后得到的多个动脉边界点数据整合为第一数据，将处理后得到的多个静脉边界点数据整合为第二数据。
可选的，所述第二预设处理包括以下至少一种操作：2D边界优化处理、3D边界优化处理、数据增强处理。
其中，2D边界优化处理包括：多次采样获取低分辨率信息和高分辨率信息，其中，低分辨率信息能够提供分割目标在整个图像中上下文语义信息，即反映目标与环境之间关系的特征，这些特征用于物体类别判断，高分辨率信息用于为分割目标提供更加精细的特征，如梯度等。
所述3D边界优化处理包括：3D卷积、3D最大池化和3D向上卷积层，输入数据的大小为a1、a2、a3，通道数为c，过滤器大小为f，即过滤器维度为f*f*f*c，过滤器数量为n，则3维卷积最终输出为：
(a1-f+1)*(a2-f+1)*(a3-f+1)*n
具有分析路径和合成路径。在分析路径中，每一层包含两个3*3*3的卷积核，每一个都跟随一个激活函数(Relu)，然后在每个维度上有2*2*2的最大池化合并两个步长。在合成路径中，每个层由2*2*2的向上卷积组成，每个维度上步长都为2，接着，两个3*3*3 的卷积，然后Relu。然后在分析路径中从相等分辨率层的shortcut连接提供了合成路径的基本高分辨特征。在最后一层中，1*1*1卷积减少了输出通道的数量。
其中，数据增强处理包括以下任意一种：基于任意角度旋转的数据增强、基于直方图均衡的数据增强、基于白平衡的数据增强、基于镜像操作的数据增强、基于随机剪切的数据增强和基于模拟不同光照变化的数据增强。
可见，本示例中，医学成像装置能够通过VRDS医学网络模型、交叉血管网络模型对BMP数据源进行处理，结合边界优化和数据增强处理得到目标影像数据，解决了传统的医学影像无法实现分割动脉和静脉的整体分离、以及无法提取血管的表面特征的医学领域的问题，提高医学影像显示的真实性、全面性和精细化程度。
203、根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤特性。
在一个可能的示例中，根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤特性，包括：根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤位置；从所述脾脏的影像数据中获取所述脾脏的特征，其中，所述脾脏的特征包括以下至少一条数据：所述脾脏的大小、所述脾脏中淋巴结的大小、所述脾脏中淋巴管的扩张程度；根据所述血管的影像数据确定所述血管的综合扩张程度。
可选的，血管的影像数据包括血管的管径和血管的表面特征，根据所述血管的影像数据确定所述血管的综合扩张程度包括：根据血管的表面特征确定血管的表面扩张程度；获取目标用户的身体参数，其中，目标用户的身体参数包括以下至少一项：目标用户的身高、体重、血压、血糖、心率；根据目标用户的身体参数确定血管的管径对应的第一权重和血管的表面扩张程度对应的第二权重；根据第一权重和第二权重分别对血管的管径和血管的表面扩张程度进行加权运算，得到血管的综合扩张程度。
其中，血管的管径可以通过以下方式得到：根据血管的第一影像数据对血管进行分段，得到M段血管，其中，M为正整数；确定M段血管中每一段血管的平均管径，其中，第i段血管的平均管径与第(i+1)段血管的平均管径的管径差不小于预设分段阈值，i为小于M的正整数；确定M段血管中每一段血管对应的权重；根据M段血管中每一段血管对应的权重对M段血管中每一段血管的平均管径进行加权运算，得到血管的管径。
可选的，所述根据所述血管的表面特征确定所述血管的表面扩张程度包括：根据血管的表面特征确定血管的目标表面区域，获取血管的目标表面区域的第一表面特征；获取正常血管的目标表面区域的第二表面特征；将第一表面特征与第二表面特征进行对比，以确定血管的表面扩张程度。
其中，所述根据所述血管的表面特征确定所述血管的目标表面区域包括：根据所述血管的表面特征分析所述血管的表面区域的特征点分布；将所述血管的表面区域按照N个不同圆心进行圆形图像截取，以得到N个圆形表面分区，N为大于3的整数；确定所述N个圆形表面分区中每个圆形表面分区所包含的特征点的数量；从所述N个圆形表面分区中选 出目标圆形表面分区，其中，所述目标圆形表面分区所包含的特征点的数量大于所述N个圆形表面分区中的其他圆形表面分区所包含的特征点的数量。
可见，本示例中，通过筛选血管的目标表面区域，将目标表面区域的表面特征与正常血管的表面特征进行对比，这样，可以减少特征对比的复杂度，缩短对比时间，提高对比效率。
在一种可能的示例中，血管的影像数据还包括血管的弯曲度，确定血管的弯曲度的方法可以是：
根据血管的第二影像数据建立坐标系，所述坐标系的原点为所述血管的任意位置，所述坐标系的X轴、Y轴和Z轴相互垂直并遵循右手螺旋法则；
从所述坐标系的原点出发，分别按照预设距离沿着所述坐标系的X轴的正方向和反方向、Y轴的正方向和反方向以及Z轴的正方向和反方向进行检测，每当检测到第一像素点对应的灰度值属于所述血管最外层的血管细胞数据对应的灰度值时，记录所述第一像素点对应的空间位置，每当检测到所述第二像素点对应的灰度值不属于所述血管最外层的血管细胞数据对应的灰度值且所述第二像素点相邻像素点对应的灰度值属于所述血管最外层的血管细胞数据对应的灰度值时，记录所述第二像素点对应的空间位置；
根据所有的所述第一像素点对应的空间位置以及所有的所述第二像素点对应的空间位置将所述血管的第二影像数据进行切分，以得到多个血管对应的多个最外层血管细胞数据集，每个最外层血管细胞数据集包括多个最外层血管细胞数据；
针对每个最外层血管细胞数据集，执行以下步骤：
获取当前处理的最外层血管细胞数据集投影在任意平面的特征曲线；选取在所述特征曲线的任意一点作为起始点；从所述起始点出发，沿着所述特征曲线的正方向和反方向不断标记像素点，当标记到目标像素点时停止标记，所述特征曲线的正方向为所述血管的第二影像数据的横向正方向，所述特征曲线的反方向为所述血管的第二影像数据的横向反方向，所述目标像素点为目标血管段曲率变化最大的像素点，所述目标血管段为目标血管在所述起始点至目标空间位置之间的血管，所述目标血管与当前处理的最外层血管细胞数据集对应，所述目标空间位置是所述目标像素点对应的位置；获取所述目标血管段对应的曲率；将所述目标血管段对应的曲率设置为所述目标血管对应弯曲度。
基于上述示例，血管的影像数据包括血管的管径、血管的表面特征和血管的弯曲度，根据所述血管的影像数据确定所述血管的综合扩张程度包括：根据血管的表面特征确定血管的表面扩张程度；获取目标用户的身体参数，其中，目标用户的身体参数包括以下至少一项：目标用户的身高、体重、血压、血糖、心率；根据目标用户的身体参数确定血管的管径对应的第三权重、血管的表面扩张程度对应的第四权重、血管的弯曲度对应的第五权重；根据第三权重、第四权重、第五权重分别对血管的管径、血管的表面扩张程度、血管的弯曲度进行加权运算，得到血管的综合扩张程度。
204、根据所述脾脏的肿瘤特性识别所述脾脏的异常类型。
脾脏是免疫器官，血运丰富，肿瘤的发生率低，脾脏良性肿瘤较多见，以血管瘤最常见，脾脏恶性肿瘤以恶性淋巴瘤最常见，脾脏良性肿瘤包括脾囊肿、血管瘤、淋巴管瘤和脾错构瘤等，脾脏恶性肿瘤包括血管肉瘤、淋巴瘤和转移瘤等，脾脏良性肿瘤一般形态较规则，边界较清晰，而且病灶外的脾脏轮廓较完整，脾脏恶性肿瘤一般脾脏肿大，而且淋巴结也肿大。
其中，脾囊肿是脾组织的瘤样囊性病变，临床上可分为寄生虫性囊肿和非寄生虫性囊肿，无血管，边界清晰；
血管瘤是脾脏良性肿瘤中最常见的一种，脾血管瘤以海绵状血管瘤最多见，根据血管扩张程度分为海绵状血管瘤和毛细血管瘤，也可混合存在，边界较为清晰；
淋巴管瘤，由淋巴管阻塞、扩张形成，病理为扩张的淋巴窦构成，分为毛细淋巴瘤、海绵状淋巴管瘤和囊性淋巴管瘤，边界较为清晰；
脾错构瘤，包括白髓型脾错构瘤、红髓型脾错构瘤和混合型脾错构瘤，白髓型脾错构瘤由异常淋巴组织构成，红髓型脾错构瘤由失调的脾窦构成，混合型脾错构瘤两者皆有，边界较为清晰；
淋巴瘤，恶性肿瘤，体现在脾门区、主动脉淋巴结肿大，重度脾大，边界不清晰，增强扫描后呈中低强化；
血管肉瘤，恶性肿瘤，脾内肿物及脾脏肿大，淋巴结肿大，边界不清晰，增强扫描后呈中高强化；
转移瘤，恶性肿瘤晚期表现，淋巴结肿大，肝内多同时发生转移灶，边界不清晰，增强扫描后呈牛眼征或者靶心征。
在一个可能的示例中，所述根据所述脾脏的肿瘤特性识别所述脾脏的异常类型包括：当脾脏的大小不超过第一预设阈值以及淋巴结的大小不超过第二预设阈值时，根据脾脏的肿瘤位置确定脾脏的肿瘤边界清晰度；判断脾脏的肿瘤边界清晰度是否超过预设清晰度阈值；若脾脏的肿瘤边界清晰度超过预设清晰度阈值，则识别脾脏的异常类型为脾囊肿；若脾脏的肿瘤边界清晰度不超过预设清晰度阈值，判断血管的综合扩张程度是否超过预设血管扩张程度阈值，若血管的综合扩张程度超过预设血管扩张程度阈值，则识别脾脏的异常类型为血管瘤，若血管的综合扩张程度不超过预设血管扩张程度阈值，则判断淋巴管的扩张程度是否超过预设淋巴管扩张程度阈值，若淋巴管的扩张程度超过预设淋巴管扩张程度阈值，则识别脾脏的异常类型为淋巴管瘤。
具体的，当脾脏的大小不超过正常脾脏大小阈值以及淋巴结的大小不超过正常淋巴结大小阈值时，可以初步判断为良性肿瘤，脾脏良性肿瘤包括脾囊肿、血管瘤、淋巴管瘤等，此时需要进行进一步判断，通过判断脾脏的肿瘤边界清晰度可以确定是否为脾囊肿，通过判断血管的综合扩张程度可以确定是否为血管瘤，通过判断淋巴管的扩张程度可以确定是 否为淋巴管瘤。
可选的，目标医学影像数据还包括对脾脏的增强扫描图像进行处理得到的增强医学影像数据，当脾脏的大小超过第一预设阈值或者淋巴结的大小超过第二预设阈值时，根据增强医学影像数据确定增强扫描后脾脏的肿瘤强化程度；判断脾脏的肿瘤强化程度是否超过预设强化程度阈值；若脾脏的肿瘤强化程度超过预设强化程度阈值，则识别脾脏的异常类型为血管肉瘤；若脾脏的肿瘤强化程度不超过预设强化程度阈值，则识别脾脏的异常类型为淋巴瘤。
具体的，当脾脏的大小超过正常脾脏大小阈值或者淋巴结的大小超过正常淋巴结大小阈值时，可以初步判断为恶性肿瘤，脾脏恶性肿瘤包括血管肉瘤、淋巴瘤和转移瘤等，通过判断增强扫描后脾脏的肿瘤强化程度，可以进一步确定脾脏的异常类型。若脾脏的肿瘤强化程度超过预设强化程度阈值，则识别脾脏的异常类型为血管肉瘤，若脾脏的肿瘤强化程度不超过预设强化程度阈值，则识别脾脏的异常类型为淋巴瘤。
可选的，还可以通过判断增强扫描后脾脏的肿瘤是否呈牛眼征或者靶心征来判断是否为转移瘤。
205、根据所述目标医学影像数据进行4D医学成像，并输出所述脾脏的异常类型。
其中，4D医学成像是指呈现4维医学影像。
在一个可能的示例中，根据所述目标医学影像数据进行4D医学成像，包括：医学成像装置从目标医学影像数据中筛选质量评分大于预设评分的增强数据作为VRDS 4D成像数据；根据VRDS 4D成像数据进行4D医学成像。
其中，质量评分可以从以下维度进行综合评价，平均梯度、信息熵、视觉信息保真度、峰值信噪比PSNR、结构相似性SSIM、均方误差MSE等，具体可以参考图像领域的常见图像质量评分算法，此处不再赘述。
可以看出，本申请实施例中，获取目标用户的脾脏的扫描图像，接着对脾脏的扫描图像进行处理，以得到目标医学影像数据，其中，目标医学影像数据包括脾脏的影像数据和脾脏周围的血管的影像数据，其次，根据脾脏的影像数据和血管的影像数据确定脾脏的肿瘤特性，其次，根据脾脏的肿瘤特性识别脾脏的异常类型，最后，根据目标医学影像数据进行4D医学成像，并输出脾脏的异常类型。可见，本申请中的医学成像装置能够通过处理脾脏的扫描图像，识别脾脏的异常类型，并输出该脾脏的异常类型，避免了基于人眼观察不够精准的情况，有利于提高医学成像装置进行脾脏肿瘤识别的准确度和效率。
参见图3，图3为本申请的一个实施例提供的一种医学成像装置300的示意图，医学成像装置300可以包括：
获取单元301，用于获取目标用户的脾脏的扫描图像；
处理单元302，用于对所述脾脏的扫描图像进行处理，以得到目标医学影像数据，其中，所述目标医学影像数据包括所述脾脏的影像数据和所述脾脏周围的血管的影像数据；
确定单元303，用于根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤特性；
识别单元304，用于根据所述脾脏的肿瘤特性识别所述脾脏的异常类型；
输出单元305，用于根据所述目标医学影像数据进行4D医学成像，并输出所述脾脏的异常类型。
在一个可能的示例中，所述处理单元302具体用于：根据所述脾脏的扫描图像生成所述脾脏的图源；针对所述图源执行第一预设处理得到位图BMP数据源；将所述BMP数据源导入预设的VRDS医学网络模型，得到第一医学影像数据，其中，所述第一医学影像数据包括所述脾脏的影像数据和所述血管的第一影像数据；将所述第一医学影像数据导入预设的交叉血管网络模型，得到第二医学影像数据，其中，所述第二医学影像数据包括所述脾脏的影像数据和所述血管的第二影像数据，所述血管的第二影像数据包括所述血管的表面特征，所述血管的表面特征通过所述交叉血管网络模型筛选所述血管的第一影像数据中的平滑肌和弹性纤维图像数据得到；针对所述第二医学影像数据执行第二预设处理得到所述目标医学影像数据。
在一个可能的示例中，所述确定单元303具体用于：根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤位置；从所述脾脏的影像数据中获取所述脾脏的特征，其中，所述脾脏的特征包括以下至少一条数据：所述脾脏的大小、所述脾脏中淋巴结的大小、所述脾脏中淋巴管的扩张程度；根据所述血管的影像数据确定所述血管的综合扩张程度。
在一个可能的示例中，所述血管的影像数据包括所述血管的管径和所述血管的表面特征，所述确定单元303具体用于：根据所述血管的表面特征确定所述血管的表面扩张程度；获取所述目标用户的身体参数，其中，所述目标用户的身体参数包括以下至少一项：所述目标用户的身高、体重、血压、血糖、心率；根据所述目标用户的身体参数确定所述血管的管径对应的第一权重和所述血管的表面扩张程度对应的第二权重；根据所述第一权重和所述第二权重分别对所述血管的管径和所述血管的表面扩张程度进行加权运算，得到所述血管的综合扩张程度。
在一个可能的示例中，所述医学成像装置300还包括管径获取单元306，所述管径获取单元306用于：根据所述血管的第一影像数据对所述血管进行分段，得到M段血管，其中，M为正整数；确定所述M段血管中每一段血管的平均管径，其中，第i段血管的平均管径与第(i+1)段血管的平均管径的管径差不小于预设分段阈值，i为小于M的正整数；确定所述M段血管中每一段血管对应的权重；根据所述M段血管中每一段血管对应的权重对所述M段血管中每一段血管的平均管径进行加权运算，得到所述血管的管径。
在一个可能的示例中，所述确定单元303具体用于：根据所述血管的表面特征确定所述血管的目标表面区域，获取所述血管的目标表面区域的第一表面特征；获取正常血管的 所述目标表面区域的第二表面特征；将所述第一表面特征与所述第二表面特征进行对比，以确定所述血管的表面扩张程度。
在一个可能的示例中，所述确定单元303具体用于：根据所述血管的表面特征分析所述血管的表面区域的特征点分布；将所述血管的表面区域按照N个不同圆心进行圆形图像截取，以得到N个圆形表面分区，N为大于3的整数；确定所述N个圆形表面分区中每个圆形表面分区所包含的特征点的数量；从所述N个圆形表面分区中选出目标圆形表面分区，其中，所述目标圆形表面分区所包含的特征点的数量大于所述N个圆形表面分区中的其他圆形表面分区所包含的特征点的数量。
在一个可能的示例中，所述识别单元304具体用于：当所述脾脏的大小不超过第一预设阈值以及所述淋巴结的大小不超过第二预设阈值时，根据所述脾脏的肿瘤位置确定所述脾脏的肿瘤边界清晰度；判断所述脾脏的肿瘤边界清晰度是否超过预设清晰度阈值；若所述脾脏的肿瘤边界清晰度超过所述预设清晰度阈值，则识别所述脾脏的异常类型为脾囊肿；若所述脾脏的肿瘤边界清晰度不超过所述预设清晰度阈值，判断所述血管的综合扩张程度是否超过预设血管扩张程度阈值，若所述血管的综合扩张程度超过所述预设血管扩张程度阈值，则识别所述脾脏的异常类型为血管瘤，若所述血管的综合扩张程度不超过所述预设血管扩张程度阈值，则判断所述淋巴管的扩张程度是否超过预设淋巴管扩张程度阈值，若所述淋巴管的扩张程度超过所述预设淋巴管扩张程度阈值，则识别所述脾脏的异常类型为淋巴管瘤。
在一个可能的示例中，所述目标医学影像数据还包括对所述脾脏的增强扫描图像进行处理得到的增强医学影像数据，所述识别单元304还用于：当所述脾脏的大小超过所述第一预设阈值或者所述淋巴结的大小超过所述第二预设阈值时，根据所述增强医学影像数据确定增强扫描后所述脾脏的肿瘤强化程度；判断所述脾脏的肿瘤强化程度是否超过预设强化程度阈值；若所述脾脏的肿瘤强化程度超过所述预设强化程度阈值，则识别所述脾脏的异常类型为血管肉瘤；若所述脾脏的肿瘤强化程度不超过所述预设强化程度阈值，则识别所述脾脏的异常类型为淋巴瘤。
本申请涉及的医学成像装置的具体实施可参见上述基于VRDS 4D医学影像的脾脏肿瘤识别方法的各实施例，在此不做赘述。
参见图4，图4为本申请的实施例涉及的硬件运行环境的医学成像装置结构示意图。其中，如图4所示，本申请的实施例涉及的硬件运行环境的医学成像装置可以包括：
处理器401，例如CPU。
存储器402，可选的，存储器可以为高速RAM存储器，也可以是稳定的存储器，例如磁盘存储器。
通信接口403，用于实现处理器401和存储器402之间的连接通信。
本领域技术人员可以理解，图4中示出的医学成像装置的结构并不构成对其的限定， 可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
如图4所示，存储器402中可以包括操作系统、网络通信模块以及脾脏肿瘤识别的程序。操作系统是管理和控制医学成像装置硬件和软件资源的程序，支持脾脏肿瘤识别的程序以及其他软件或程序的运行。网络通信模块用于实现存储器402内部各组件之间的通信，以及与医学成像装置内部其他硬件和软件之间通信。
在图4所示的医学成像装置中，处理器401用于执行存储器402中存储的脾脏肿瘤识别的程序，实现以下步骤：
获取目标用户的脾脏的扫描图像；
对所述脾脏的扫描图像进行处理，以得到目标医学影像数据，其中，所述目标医学影像数据包括所述脾脏的影像数据和所述脾脏周围的血管的影像数据；
根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤特性；
根据所述脾脏的肿瘤特性识别所述脾脏的异常类型；
根据所述目标医学影像数据进行4D医学成像，并输出所述脾脏的异常类型。
本申请涉及的医学成像装置的具体实施可参见上述基于VRDS 4D医学影像的脾脏肿瘤识别方法的各实施例，在此不做赘述。
本申请还提供了一种计算机可读存储介质，所述计算机可读存储介质用于存储计算机程序，所述存储计算机程序被所述处理器执行，以实现以下步骤：
获取目标用户的脾脏的扫描图像；
对所述脾脏的扫描图像进行处理，以得到目标医学影像数据，其中，所述目标医学影像数据包括所述脾脏的影像数据和所述脾脏周围的血管的影像数据；
根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤特性；
根据所述脾脏的肿瘤特性识别所述脾脏的异常类型；
根据所述目标医学影像数据进行4D医学成像，并输出所述脾脏的异常类型。
本申请涉及的计算机可读存储介质的具体实施可参见上述基于VRDS 4D医学影像的脾脏肿瘤识别方法的各实施例，在此不做赘述。
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员应所述知悉，本申请并不受所描述的动作顺序的限制，因为依据本申请，某些步骤可以采用其他顺序或者同时进行。其次，本领域技术人员也应所述知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作和模块并不一定是本申请所必须的。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。
在本申请所提供的几个实施例中，应所述理解到，所揭露的装置，可通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如所述模块的划分，仅仅为一 种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个模块或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或模块的间接耦合或通信连接，可以是电性或者其它的形式。
所述作为分离部件说明的模块可以是或者也可以不是物理上分开的，作为模块显示的部件可以是或者也可以不是物理模块，即可以位于一个地方，或者也可以分布到多个网络模块上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能模块可以集成在一个处理模块中，也可以是各个模块单独物理存在，也可以两个或两个以上模块集成在一个模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。
所述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者所述技术方案的全部或部分可以以软件产品的形式体现出来，所述计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可为个人计算机、医学成像装置或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、移动硬盘、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的范围。

Claims (20)

  1. 一种基于VRDS 4D医学影像的脾脏肿瘤识别方法，其特征在于，应用于医学成像装置，所述方法包括：
    获取目标用户的脾脏的扫描图像；
    对所述脾脏的扫描图像进行处理，以得到目标医学影像数据，其中，所述目标医学影像数据包括所述脾脏的影像数据和所述脾脏周围的血管的影像数据；
    根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤特性；
    根据所述脾脏的肿瘤特性识别所述脾脏的异常类型；
    根据所述目标医学影像数据进行4D医学成像，并输出所述脾脏的异常类型。
  2. 根据权利要求1所述的方法，其特征在于，所述对所述脾脏的扫描图像进行处理，以得到目标医学影像数据，包括：
    根据所述脾脏的扫描图像生成所述脾脏的图源；
    针对所述图源执行第一预设处理得到位图BMP数据源；
    将所述BMP数据源导入预设的VRDS医学网络模型，得到第一医学影像数据，其中，所述第一医学影像数据包括所述脾脏的影像数据和所述血管的第一影像数据；
    将所述第一医学影像数据导入预设的交叉血管网络模型，得到第二医学影像数据，其中，所述第二医学影像数据包括所述脾脏的影像数据和所述血管的第二影像数据，所述血管的第二影像数据包括所述血管的表面特征，所述血管的表面特征通过所述交叉血管网络模型筛选所述血管的第一影像数据中的平滑肌和弹性纤维图像数据得到；
    针对所述第二医学影像数据执行第二预设处理得到所述目标医学影像数据。
  3. 根据权利要求2所述的方法，其特征在于，所述根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤特性包括：
    根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤位置；
    从所述脾脏的影像数据中获取所述脾脏的特征，其中，所述脾脏的特征包括以下至少一条数据：所述脾脏的大小、所述脾脏中淋巴结的大小、所述脾脏中淋巴管的扩张程度；
    根据所述血管的影像数据确定所述血管的综合扩张程度。
  4. 根据权利要求3所述的方法，其特征在于，所述血管的影像数据包括所述血管的管径和所述血管的表面特征，所述根据所述血管的影像数据确定所述血管的综合扩张程度包括：
    根据所述血管的表面特征确定所述血管的表面扩张程度；
    获取所述目标用户的身体参数，其中，所述目标用户的身体参数包括以下至少一项：所述目标用户的身高、体重、血压、血糖、心率；
    根据所述目标用户的身体参数确定所述血管的管径对应的第一权重和所述血管的表面扩张程度对应的第二权重；
    根据所述第一权重和所述第二权重分别对所述血管的管径和所述血管的表面扩张程度进行加权运算，得到所述血管的综合扩张程度。
  5. 根据权利要求4所述的方法，其特征在于，所述将所述BMP数据源导入预设的VRDS医学网络模型，得到第一医学影像数据之后，所述方法还包括：
    根据所述血管的第一影像数据对所述血管进行分段，得到M段血管，其中，M为正整数；
    确定所述M段血管中每一段血管的平均管径，其中，第i段血管的平均管径与第(i+1)段血管的平均管径的管径差不小于预设分段阈值，i为小于M的正整数；
    确定所述M段血管中每一段血管对应的权重；
    根据所述M段血管中每一段血管对应的权重对所述M段血管中每一段血管的平均管径进行加权运算，得到所述血管的管径。
  6. 根据权利要求4或5所述的方法，其特征在于，所述根据所述血管的表面特征确定所述血管的表面扩张程度包括：
    根据所述血管的表面特征确定所述血管的目标表面区域，获取所述血管的目标表面区域的第一表面特征；
    获取正常血管的所述目标表面区域的第二表面特征；
    将所述第一表面特征与所述第二表面特征进行对比，以确定所述血管的表面扩张程度。
  7. 根据权利要求6所述的方法，其特征在于，所述根据所述血管的表面特征确定所述血管的目标表面区域包括：
    根据所述血管的表面特征分析所述血管的表面区域的特征点分布；
    将所述血管的表面区域按照N个不同圆心进行圆形图像截取，以得到N个圆形表面分区，N为大于3的整数；
    确定所述N个圆形表面分区中每个圆形表面分区所包含的特征点的数量；
    从所述N个圆形表面分区中选出目标圆形表面分区，其中，所述目标圆形表面分区所包含的特征点的数量大于所述N个圆形表面分区中的其他圆形表面分区所包含的特征点的数量。
  8. 根据权利要求3所述的方法，其特征在于，所述根据所述脾脏的肿瘤特性识别所述脾脏的异常类型包括：
    当所述脾脏的大小不超过第一预设阈值以及所述淋巴结的大小不超过第二预设阈值时，根据所述脾脏的肿瘤位置确定所述脾脏的肿瘤边界清晰度；
    判断所述脾脏的肿瘤边界清晰度是否超过预设清晰度阈值；
    若所述脾脏的肿瘤边界清晰度超过所述预设清晰度阈值，则识别所述脾脏的异常类型为脾囊肿；
    若所述脾脏的肿瘤边界清晰度不超过所述预设清晰度阈值，判断所述血管的综合扩张 程度是否超过预设血管扩张程度阈值，若所述血管的综合扩张程度超过所述预设血管扩张程度阈值，则识别所述脾脏的异常类型为血管瘤，若所述血管的综合扩张程度不超过所述预设血管扩张程度阈值，则判断所述淋巴管的扩张程度是否超过预设淋巴管扩张程度阈值，若所述淋巴管的扩张程度超过所述预设淋巴管扩张程度阈值，则识别所述脾脏的异常类型为淋巴管瘤。
  9. 根据权利要求8所述的方法，其特征在于，所述目标医学影像数据还包括对所述脾脏的增强扫描图像进行处理得到的增强医学影像数据，所述根据所述脾脏的肿瘤特性识别所述脾脏的异常类型还包括：
    当所述脾脏的大小超过所述第一预设阈值或者所述淋巴结的大小超过所述第二预设阈值时，根据所述增强医学影像数据确定增强扫描后所述脾脏的肿瘤强化程度；
    判断所述脾脏的肿瘤强化程度是否超过预设强化程度阈值；
    若所述脾脏的肿瘤强化程度超过所述预设强化程度阈值，则识别所述脾脏的异常类型为血管肉瘤；
    若所述脾脏的肿瘤强化程度不超过所述预设强化程度阈值，则识别所述脾脏的异常类型为淋巴瘤。
  10. 一种医学成像装置，其特征在于，所述装置包括：
    获取单元，用于获取目标用户的脾脏的扫描图像；
    处理单元，用于对所述脾脏的扫描图像进行处理，以得到目标医学影像数据，其中，所述目标医学影像数据包括所述脾脏的影像数据和所述脾脏周围的血管的影像数据；
    确定单元，用于根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤特性；
    识别单元，用于根据所述脾脏的肿瘤特性识别所述脾脏的异常类型；
    输出单元，用于根据所述目标医学影像数据进行4D医学成像，并输出所述脾脏的异常类型。
  11. 根据权利要求10所述的装置，其特征在于，所述处理单元具体用于：
    根据所述脾脏的扫描图像生成所述脾脏的图源；
    针对所述图源执行第一预设处理得到位图BMP数据源；
    将所述BMP数据源导入预设的VRDS医学网络模型，得到第一医学影像数据，其中，所述第一医学影像数据包括所述脾脏的影像数据和所述血管的第一影像数据；
    将所述第一医学影像数据导入预设的交叉血管网络模型，得到第二医学影像数据，其中，所述第二医学影像数据包括所述脾脏的影像数据和所述血管的第二影像数据，所述血管的第二影像数据包括所述血管的表面特征，所述血管的表面特征通过所述交叉血管网络模型筛选所述血管的第一影像数据中的平滑肌和弹性纤维图像数据得到；
    针对所述第二医学影像数据执行第二预设处理得到所述目标医学影像数据。
  12. 根据权利要求11所述的装置，其特征在于，所述确定单元具体用于：
    根据所述脾脏的影像数据和所述血管的影像数据确定所述脾脏的肿瘤位置；
    从所述脾脏的影像数据中获取所述脾脏的特征，其中，所述脾脏的特征包括以下至少一条数据：所述脾脏的大小、所述脾脏中淋巴结的大小、所述脾脏中淋巴管的扩张程度；
    根据所述血管的影像数据确定所述血管的综合扩张程度。
  13. 根据权利要求12所述的装置，其特征在于，所述血管的影像数据包括所述血管的管径和所述血管的表面特征，所述确定单元具体用于：
    根据所述血管的表面特征确定所述血管的表面扩张程度；
    获取所述目标用户的身体参数，其中，所述目标用户的身体参数包括以下至少一项：所述目标用户的身高、体重、血压、血糖、心率；
    根据所述目标用户的身体参数确定所述血管的管径对应的第一权重和所述血管的表面扩张程度对应的第二权重；
    根据所述第一权重和所述第二权重分别对所述血管的管径和所述血管的表面扩张程度进行加权运算，得到所述血管的综合扩张程度。
  14. 根据权利要求13所述的装置，其特征在于，所述装置还包括管径获取单元，所述管径获取单元用于：
    根据所述血管的第一影像数据对所述血管进行分段，得到M段血管，其中，M为正整数；
    确定所述M段血管中每一段血管的平均管径，其中，第i段血管的平均管径与第(i+1)段血管的平均管径的管径差不小于预设分段阈值，i为小于M的正整数；
    确定所述M段血管中每一段血管对应的权重；
    根据所述M段血管中每一段血管对应的权重对所述M段血管中每一段血管的平均管径进行加权运算，得到所述血管的管径。
  15. 根据权利要求13或14所述的装置，其特征在于，所述确定单元具体用于：
    根据所述血管的表面特征确定所述血管的目标表面区域，获取所述血管的目标表面区域的第一表面特征；
    获取正常血管的所述目标表面区域的第二表面特征；
    将所述第一表面特征与所述第二表面特征进行对比，以确定所述血管的表面扩张程度。
  16. 根据权利要求15所述的装置，其特征在于，所述确定单元具体用于：
    根据所述血管的表面特征分析所述血管的表面区域的特征点分布；
    将所述血管的表面区域按照N个不同圆心进行圆形图像截取，以得到N个圆形表面分区，N为大于3的整数；
    确定所述N个圆形表面分区中每个圆形表面分区所包含的特征点的数量；
    从所述N个圆形表面分区中选出目标圆形表面分区，其中，所述目标圆形表面分区所 包含的特征点的数量大于所述N个圆形表面分区中的其他圆形表面分区所包含的特征点的数量。
  17. 根据权利要求12所述的装置，其特征在于，所述识别单元具体用于：
    当所述脾脏的大小不超过第一预设阈值以及所述淋巴结的大小不超过第二预设阈值时，根据所述脾脏的肿瘤位置确定所述脾脏的肿瘤边界清晰度；
    判断所述脾脏的肿瘤边界清晰度是否超过预设清晰度阈值；
    若所述脾脏的肿瘤边界清晰度超过所述预设清晰度阈值，则识别所述脾脏的异常类型为脾囊肿；
    若所述脾脏的肿瘤边界清晰度不超过所述预设清晰度阈值，判断所述血管的综合扩张程度是否超过预设血管扩张程度阈值，若所述血管的综合扩张程度超过所述预设血管扩张程度阈值，则识别所述脾脏的异常类型为血管瘤，若所述血管的综合扩张程度不超过所述预设血管扩张程度阈值，则判断所述淋巴管的扩张程度是否超过预设淋巴管扩张程度阈值，若所述淋巴管的扩张程度超过所述预设淋巴管扩张程度阈值，则识别所述脾脏的异常类型为淋巴管瘤。
  18. 根据权利要求17所述的装置，其特征在于，所述目标医学影像数据还包括对所述脾脏的增强扫描图像进行处理得到的增强医学影像数据，所述识别单元还用于：
    当所述脾脏的大小超过所述第一预设阈值或者所述淋巴结的大小超过所述第二预设阈值时，根据所述增强医学影像数据确定增强扫描后所述脾脏的肿瘤强化程度；
    判断所述脾脏的肿瘤强化程度是否超过预设强化程度阈值；
    若所述脾脏的肿瘤强化程度超过所述预设强化程度阈值，则识别所述脾脏的异常类型为血管肉瘤；
    若所述脾脏的肿瘤强化程度不超过所述预设强化程度阈值，则识别所述脾脏的异常类型为淋巴瘤。
  19. 一种医学成像装置，其特征在于，包括处理器、存储器、通信接口以及一个或多个程序，其中，所述一个或多个程序被存储在所述存储器中，并且被生成由所述处理器执行，以执行权利要求1-9任一项方法中的步骤的指令。
  20. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质用于存储计算机程序，所述存储计算机程序被所述处理器执行，以实现权利要求1-9任一项所述的方法。
PCT/CN2019/114481 2019-10-30 2019-10-30 基于vrds 4d医学影像的脾脏肿瘤识别方法及相关装置 WO2021081841A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2019/114481 WO2021081841A1 (zh) 2019-10-30 2019-10-30 基于vrds 4d医学影像的脾脏肿瘤识别方法及相关装置
CN201980099987.8A CN114365190A (zh) 2019-10-30 2019-10-30 基于vrds 4d医学影像的脾脏肿瘤识别方法及相关装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/114481 WO2021081841A1 (zh) 2019-10-30 2019-10-30 基于vrds 4d医学影像的脾脏肿瘤识别方法及相关装置

Publications (1)

Publication Number Publication Date
WO2021081841A1 true WO2021081841A1 (zh) 2021-05-06

Family

ID=75715674

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/114481 WO2021081841A1 (zh) 2019-10-30 2019-10-30 基于vrds 4d医学影像的脾脏肿瘤识别方法及相关装置

Country Status (2)

Country Link
CN (1) CN114365190A (zh)
WO (1) WO2021081841A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113674254A (zh) * 2021-08-25 2021-11-19 上海联影医疗科技股份有限公司 医学图像异常点识别方法、设备、电子装置和存储介质
CN115274119A (zh) * 2022-09-30 2022-11-01 中国医学科学院北京协和医院 一种融合多影像组学特征的免疫治疗预测模型的构建方法
CN115919464A (zh) * 2023-03-02 2023-04-07 四川爱麓智能科技有限公司 肿瘤定位方法、系统、装置及肿瘤发展预测方法
CN116703784A (zh) * 2023-08-02 2023-09-05 济南宝林信息技术有限公司 一种心脏超声图像视觉增强方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104008386A (zh) * 2014-05-13 2014-08-27 中国科学院深圳先进技术研究院 肿瘤类型识别方法和系统
CN108399354A (zh) * 2017-02-08 2018-08-14 上海点医计算机科技有限公司 计算机视觉识别肿瘤的方法和装置
CN109949899A (zh) * 2019-02-28 2019-06-28 未艾医疗技术(深圳)有限公司 图像三维测量方法、电子设备、存储介质及程序产品

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104008386A (zh) * 2014-05-13 2014-08-27 中国科学院深圳先进技术研究院 肿瘤类型识别方法和系统
CN108399354A (zh) * 2017-02-08 2018-08-14 上海点医计算机科技有限公司 计算机视觉识别肿瘤的方法和装置
CN109949899A (zh) * 2019-02-28 2019-06-28 未艾医疗技术(深圳)有限公司 图像三维测量方法、电子设备、存储介质及程序产品

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113674254A (zh) * 2021-08-25 2021-11-19 上海联影医疗科技股份有限公司 医学图像异常点识别方法、设备、电子装置和存储介质
CN113674254B (zh) * 2021-08-25 2024-05-14 上海联影医疗科技股份有限公司 医学图像异常点识别方法、设备、电子装置和存储介质
CN115274119A (zh) * 2022-09-30 2022-11-01 中国医学科学院北京协和医院 一种融合多影像组学特征的免疫治疗预测模型的构建方法
CN115919464A (zh) * 2023-03-02 2023-04-07 四川爱麓智能科技有限公司 肿瘤定位方法、系统、装置及肿瘤发展预测方法
CN115919464B (zh) * 2023-03-02 2023-06-23 四川爱麓智能科技有限公司 肿瘤定位方法、系统、装置及肿瘤发展预测方法
CN116703784A (zh) * 2023-08-02 2023-09-05 济南宝林信息技术有限公司 一种心脏超声图像视觉增强方法
CN116703784B (zh) * 2023-08-02 2023-10-20 济南宝林信息技术有限公司 一种心脏超声图像视觉增强方法

Also Published As

Publication number Publication date
CN114365190A (zh) 2022-04-15

Similar Documents

Publication Publication Date Title
WO2021081841A1 (zh) 基于vrds 4d医学影像的脾脏肿瘤识别方法及相关装置
AU2019430369B2 (en) VRDS 4D medical image-based vein Ai endoscopic analysis method and product
CN114365188A (zh) 基于vrds ai下腔静脉影像的分析方法及产品
WO2021081771A1 (zh) 基于vrds ai医学影像的心脏冠脉的分析方法和相关装置
AU2019431568B2 (en) Method and product for processing of vrds 4d medical images
WO2021081850A1 (zh) 基于vrds 4d医学影像的脊椎疾病识别方法及相关装置
WO2020168695A1 (zh) 基于VRDS 4D医学影像的肿瘤与血管Ai处理方法及产品
WO2020168697A1 (zh) 基于VRDS 4D医学影像的栓塞的Ai识别方法及产品
WO2021081839A1 (zh) 基于vrds 4d的病情分析方法及相关产品
WO2020168694A1 (zh) 基于VRDS 4D医学影像的肿瘤Ai处理方法及产品
WO2020168696A1 (zh) 基于VRDS 4D医学影像的动脉与静脉Ai处理方法及产品
WO2021081842A1 (zh) 基于vrds ai医学影像的肠肿瘤与血管分析方法和相关装置
WO2021081772A1 (zh) 基于vrds ai脑部影像的分析方法和相关装置
WO2021081836A1 (zh) 基于vrds 4d医学影像的胃肿瘤识别方法及相关产品
WO2021081845A1 (zh) 一种基于vrds ai的肝脏肿瘤和血管分析方法及相关产品
WO2021081835A1 (zh) 基于VRDS 4D医学影像的动脉瘤Ai处理方法及产品

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19950339

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19950339

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 19950339

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 24.10.2022)

122 Ep: pct application non-entry in european phase

Ref document number: 19950339

Country of ref document: EP

Kind code of ref document: A1