PAE guiding method for MIP post-processing reconstruction and intervention reconstruction of magnetic resonance

Abstract

The invention provides a PAE guiding method for magnetic resonance post-processing reconstruction of an MIP map and intervention reconstruction. The method comprises the following steps: constructing an MIP image for a prostatic artery blood vessel through an MIP algorithm; meanwhile, constructing an interventional artery image of a prostatic artery blood vessel; whether the prostatic artery blood vessel characteristics on the axial position, the coronal position and the sagittal position can be automatically compared successfully or not is judged. The MIP image and the interventional artery image can be automatically identified and are matched with each other, so that dissection of each layer can be performed on thin layers of an axial position, a coronal position and a sagittal position of multi-parameter scanning of magnetic resonance scanning during a clinical interventional operation, and detailed dissection knowledge can be provided for the operation. Therefore, the problems of fuzziness and confusion caused by blood vessel variation, overlapping, crossing and the like can be avoided, a specific prostatic artery target blood vessel can be accurately guided for a doctor, time consumed by the doctor is greatly shortened, and preoperative preparation and patient operation time efficiency are prevented from being affected.

Description

Technical Field

[0001] The present invention relates to the technical field, and in particular to a PAE guidance method for reconstructing MIP images through post-processing magnetic resonance imaging and interventional reconstruction, a prostate artery embolization guidance device and electronic equipment. Background Art

[0002] Prostate embolization, also known as prostate artery embolization (PAE), is an emerging urological surgery, mainly used to treat benign prostatic hyperplasia (BPH). This surgical method has the advantages of less trauma, good treatment effect, simple operation, and fast postoperative recovery. The following is a detailed introduction to prostate embolization:

[0003] 1. Principle of surgery

[0004] Prostate embolization is an interventional technique that injects embolic agents into the arteries that control the prostate, thereby embolizing these arteries and causing the prostate to lose blood supply and gradually shrink. This process can avoid prostate hyperplasia and compression of the urethral opening, thereby improving the patient's lower urinary tract symptoms.

[0005] 2. Indications

[0006] Prostate embolization is mainly used in the following situations:

[0007] Patients with severe lower urinary tract symptoms caused by benign prostatic hyperplasia who are ineffective in drug treatment or cannot tolerate long-term drug treatment.

[0008] Patients with poor physical condition who cannot tolerate anesthesia or surgical risks of prostate surgery.

[0009] Patients who are unwilling to undergo cystostomy or long-term indwelling urinary catheterization.

[0010] 3. Surgical Procedure

[0011] Prostate embolization is usually performed under local or general anesthesia and under the guidance of digital subtraction angiography (DSA). During the operation, the doctor will insert a catheter into the prostate blood supply artery through a vascular puncture, and then inject an embolic agent to block blood flow. The entire operation is relatively simple, with little trauma and little bleeding.

[0012] 4. Efficacy and risks

[0013] Prostate embolization has a significant effect, which can reduce the size of the prostate and improve the patient's lower urinary tract symptoms to a certain extent. However, this operation also has certain risks, such as infection, fever, hematuria and other symptoms during the embolization process. In addition, due to the rich blood supply of the prostate, it may not be possible to completely embolize it, and the prostate may proliferate again after the operation.

[0014] However, when performing prostate embolization surgery in hospitals, it is necessary to observe the anatomical position of the prostate artery and whether there are any variations. For this reason, all patients must undergo an MRI prostate artery MIP scan.

[0015] As attached Figure 1 The figure shows a nuclear magnetic resonance image of a prostate artery MIP scan. Although the prostate artery image acquired under magnetic resonance scanning can show the doctor the direction of the prostate artery, the prostate artery MRI image obtained by this method still has fuzziness and confusion caused by vascular variation, overlap, and intersection, and cannot accurately guide the doctor to the specific prostate artery target vessel, requiring the doctor to spend a lot of time, affecting the preoperative preparation and the patient's surgical timeliness, and is inefficient.

[0016] In addition, the above problems also exist when observing the prostate artery based on interventional radiology images. Summary of the invention

[0017] In order to solve the technical problems existing in the prior art, the present invention provides the following technical solutions:

[0018] On the one hand, a PAE-guided method for reconstructing an MIP image by post-processing magnetic resonance imaging and an interventional reconstruction is provided, the method being implemented by an electronic device and comprising:

[0019] S1, obtaining the original vascular image of the prostatic artery by magnetic resonance scanning and performing anatomical preprocessing to obtain a pelvic MRI post-processing image of the prostatic artery;

[0020] S2, performing image MIP reconstruction on the pelvic MRI post-processed image based on the MIP algorithm, and obtaining a MIP image of the prostate artery after reconstruction, including:

[0021] Analyzing the MRI post-processed image of the prostate artery and blood vessels to obtain MRI scanning data in different scanning directions;

[0022] Traversing the MRI scan data, reading the MRI scan data of the prostate artery in different scanning directions, and calculating the maximum three-dimensional pixel value, and saving the value to obtain the MRI scan data set of the prostate artery;

[0023] Based on the MIP algorithm, performing MIP projection imaging on the MRI scan data set in axial, coronal and sagittal scanning directions to generate a MIP projection data set of the prostate artery;

[0024] According to the same scanning direction, each MIP projection image in the MIP projection data set is fused to reconstruct a MIP image of the prostate artery;

[0025] S3. Obtain interventional radiological scanning images of the prostate artery through magnetic resonance scanning, and construct interventional arterial images of the prostate artery based on three-dimensional reconstruction and vascular segmentation technology;

[0026] S4. Identify and determine whether the prostate artery vascular features of the MIP image and the interventional artery image in the axial, coronary and sagittal planes can be automatically compared successfully:

[0027] If yes, go to S5;

[0028] If not, go to S6;

[0029] S5, outputting the MIP image and displaying it on a front-end page of the operating room for guiding prostate artery embolization;

[0030] S6. Using a preset prostate artery recognition AI model, the prostate artery image features in the interventional artery image are identified and located and fused on the MIP image, and the located and fused MIP image is output and displayed on the front page of the operating room for guiding prostate artery embolization.

[0031] As a preferred embodiment of the present invention, preferably, S6, using a preset prostate artery recognition AI model, identifying the prostate artery image features in the interventional artery image and merging them on the MIP image, and outputting and displaying the positioned fused MIP image on the front page of the operating room for guiding the prostate artery embolization, comprises:

[0032] S600, performing image enhancement processing on the interventional arterial image to obtain an arterial vascular interventional enhanced image, and importing the arterial vascular interventional enhanced image into a preset prostate artery recognition AI model, wherein the prostate artery recognition AI model recognizes and marks the prostate artery image features in the axial, coronary and sagittal positions of the arterial vascular interventional enhanced image;

[0033] S601, based on image registration and fusion, positioning and fusing the prostate artery image features on the MIP image, forming contrast features on the axial, coronal and sagittal planes with the original prostate artery blood vessels on the MIP image, and generating vascular tomography contrast features of the positioned and fused prostate artery blood vessels on the MIP image;

[0034] S602: Output the interventional arterial image of the located and fused prostate artery and display it on the front page of the operating room for guiding the prostate artery embolization.

[0035] As a preferred embodiment of the present invention, preferably, S3, obtaining an interventional radiological scanning image of the prostatic artery through magnetic resonance scanning, and constructing an interventional arterial image of the prostatic artery based on three-dimensional reconstruction and vascular segmentation technology, includes:

[0036] Performing MRI scanning on the prostate artery and blood vessels to obtain MRI scanning data corresponding to the prostate artery and blood vessels;

[0037] Annotating the MRI scan data of the prostate artery with vascular contour data, and converting the data to generate a two-dimensional original scan image of the prostate artery obtained under MRI scanning;

[0038] Combining a volume rendering three-dimensional reconstruction algorithm, the two-dimensional original scanned images under each scanning sequence are converted into three dimensions, the prostate artery and blood vessels are reconstructed, and a corresponding three-dimensional model of the prostate artery and blood vessels is generated;

[0039] Using the preset prostate artery segmentation model, the image segmentation processing of the prostate artery three-dimensional model in the axial, coronary and sagittal planes is performed respectively, and the prostate artery images of the prostate artery three-dimensional model in the axial, coronary and sagittal planes are obtained to obtain the magnetic resonance artery MIP image of the prostate artery.

[0040] As a preferred embodiment of the present invention, preferably, the method for generating the prostate artery segmentation model comprises:

[0041] The MRI image sets of the prostate artery vessels in the axial, coronary and sagittal planes were collected;

[0042] The axial, coronal and sagittal MRI images of the prostate artery were preprocessed respectively;

[0043] Performing labeling processing on the prostate artery vascular MRI images in the axial, coronary and sagittal planes in the prostate artery vascular MRI image set, and determining the boundary positions of the prostate artery vascular in the axial, coronary and sagittal planes;

[0044] After labeling is completed, a training set is obtained, and the training set is imported into a preset convolutional neural network model to perform labeled data training and learning, thereby obtaining the prostate artery segmentation model;

[0045] Using the validation sets of prostate artery vascular images in the axial, coronary and sagittal planes, respectively, the prostate artery segmentation model is segmented and validated;

[0046] After verification, the prostate artery segmentation model is deployed on the background server.

[0047] As a preferred embodiment of the present invention, preferably, S4, identifying and judging whether the prostate artery vascular features of the MIP image and the interventional artery image in the axial, coronary and sagittal planes can be automatically compared successfully, comprises:

[0048] Extracting the first prostatic artery vascular features of the MIP image in the axial, coronary and sagittal planes respectively;

[0049] Respectively extracting second prostate artery vascular features of the prostate artery vascular image in the axial, coronary and sagittal planes of the prostate artery vascular three-dimensional model;

[0050] Automatic comparison and judgment:

[0051] (1) performing feature comparison between the first prostate artery vascular feature and the second prostate artery vascular feature in the axial position;

[0052] (2) performing feature comparison between the first prostate artery vascular feature and the second prostate artery vascular feature in the coronal plane;

[0053] (3) performing feature comparison between the first prostate artery vascular feature and the second prostate artery vascular feature in the sagittal plane;

[0054] If (1)-(3) can all be successfully compared, it is considered that the prostate artery vascular features of the MIP image and the interventional artery image in the axial, coronary and sagittal planes can be automatically compared successfully;

[0055] Otherwise it fails.

[0056] On the other hand, a prostate artery embolization guiding device is provided, the prostate artery embolization guiding device is used to implement a PAE guiding method for reconstructing MIP images and interventional reconstruction by magnetic resonance post-processing, the device comprising:

[0057] (1) An MRI image processing system, used to obtain an original vascular image of the prostatic artery through magnetic resonance scanning and perform anatomical preprocessing to obtain a pelvic MRI post-processing image of the prostatic artery;

[0058] (2) A MIP image processing system, for performing MIP reconstruction on the pelvic MRI post-processed image based on a MIP algorithm, and obtaining a MIP image of the prostate artery after reconstruction, including:

[0059] Analyzing the MRI post-processed image of the prostate artery and blood vessels to obtain MRI scanning data in different scanning directions;

[0060] Traversing the MRI scan data, reading the MRI scan data of the prostate artery in different scanning directions, and calculating the maximum three-dimensional pixel value, and saving the value to obtain the MRI scan data set of the prostate artery;

[0061] Based on the MIP algorithm, performing MIP projection imaging on the MRI scan data set in axial, coronal and sagittal scanning directions to generate a MIP projection data set of the prostate artery;

[0062] According to the same scanning direction, each MIP projection image in the MIP projection data set is fused to reconstruct a MIP image of the prostate artery;

[0063] (3) An interventional arterial image processing system, which is used to obtain interventional radiological scan images of the prostate artery through magnetic resonance scanning, and to construct interventional arterial images of the prostate artery based on three-dimensional reconstruction and vascular segmentation technology;

[0064] (4) An arterial feature recognition system, for recognizing and determining whether the prostate artery vascular features of the MIP image and the interventional arterial image in the axial, coronary and sagittal planes can be automatically compared successfully:

[0065] If yes, the MIP image is output and displayed on the front page of the operating room for guiding the prostate artery embolization;

[0066] If not, the preset prostate artery recognition AI model is used to identify the prostate artery image features in the interventional artery image and position and fuse them on the MIP image, and the positioned and fused MIP image is output and displayed on the front page of the operating room to guide the prostate artery embolization.

[0067] (5) a front end of an operating room, for displaying the MIP image;

[0068] The MRI image processing system is communicatively connected with the MIP image processing system;

[0069] The MIP image processing system and the interventional artery image processing system are respectively connected to the artery feature recognition system for communication;

[0070] The artery feature recognition system is communicatively connected to the front end of the operating room.

[0071] On the other hand, an electronic device is provided, comprising: a processor; a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, any one of the above-mentioned PAE-guided methods for magnetic resonance post-processing reconstruction of MIP images and interventional reconstruction is implemented.

[0072] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, and the at least one instruction is loaded and executed by a processor to implement any one of the above-mentioned PAE-guided methods for magnetic resonance post-processing reconstruction of MIP images and interventional reconstruction.

[0073] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0074] Based on the implementation of the present invention, the present invention reconstructs the MRI image of the prostate artery based on the MIP algorithm to construct the MIP image of the prostate artery; at the same time, the interventional radiological scanning image of the prostate artery is obtained through magnetic resonance scanning, and the interventional arterial image of the prostate artery is constructed based on three-dimensional reconstruction and vascular segmentation technology; then the prostate artery vascular features in the axial, coronary and sagittal planes are respectively checked to see whether they can be automatically compared successfully: if the automatic comparison is successful, the MIP image is output and displayed on the front page of the operating room for guiding prostate artery embolization; if it is unsuccessful, the preset prostate artery recognition AI model is used to identify the prostate artery image features in the interventional arterial image and locate and fuse them on the MIP image, and the positioned and fused MIP image is output and displayed on the front page of the operating room for guiding prostate artery embolization.

[0075] Therefore, the present invention automatically compares all the original data of magnetic resonance scanning and the original data of interventional radiology when doing intervention after reconstruction, so as to meet the automatic comparison between the two. If the comparison between the two is successful, the artery can be directly embolized during the operation to meet the diagnosis of clinical surgery. By using the post-processing software to perform maximum density projection on the scanned original data, it is possible to clearly see the origin of the artery and the course between the prostate on the post-processing image, and whether there is a variation. Using the above scheme, if the MIP image and the interventional artery image can be automatically identified and matched, the axial, coronal, and sagittal thin layers of the multi-parameter scan of the magnetic resonance scan can be compared during the clinical interventional surgery to perform dissection at each level, which can provide detailed anatomical knowledge for the surgery. Therefore, it can avoid the fuzziness and confusion caused by vascular variation, overlap, and intersection, accurately guide the doctor to the specific prostate artery target blood vessel, greatly reduce the doctor's time consumption, and avoid affecting the preoperative preparation and the patient's surgical timeliness. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0077] Figure 1 This is a MRI image of a conventional prostate artery MIP scan;

[0078] Figure 2 It is a flow chart of a PAE-guided method for reconstructing MIP images and interventional reconstruction after magnetic resonance imaging provided by an embodiment of the present invention;

[0079] Figure 3 It is a block diagram of a prostate artery embolization guidance device provided by an embodiment of the present invention;

[0080] Figure 4 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0081] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0082] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0083] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0084] In the embodiments of the present invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are consistent.

[0085] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0086] The embodiment of the present invention provides a PAE guidance method for reconstructing MIP images and interventional reconstruction after magnetic resonance imaging. The method can be implemented by an electronic device, which can be a terminal or a server. Figure 2 The flowchart of the PAE-guided method for reconstructing MIP images after magnetic resonance imaging processing and interventional reconstruction is shown. The processing flow of the method may include the following steps:

[0087] On the one hand, a PAE-guided method for reconstructing an MIP image by post-processing magnetic resonance imaging and an interventional reconstruction is provided, the method being implemented by an electronic device and comprising:

[0088] S1, obtaining the original vascular image of the prostatic artery by magnetic resonance scanning and performing anatomical preprocessing to obtain a pelvic MRI post-processing image of the prostatic artery;

[0089] S2, performing image MIP reconstruction on the pelvic MRI post-processed image based on the MIP algorithm, and obtaining a MIP image of the prostate artery after reconstruction, including:

[0090] Analyzing the MRI post-processed image of the prostate artery and blood vessels to obtain MRI scanning data in different scanning directions;

[0091] Traversing the MRI scan data, reading the MRI scan data of the prostate artery in different scanning directions, and calculating the maximum three-dimensional pixel value, and saving the value to obtain the MRI scan data set of the prostate artery;

[0092] Based on the MIP algorithm, performing MIP projection imaging on the MRI scan data set in axial, coronal and sagittal scanning directions to generate a MIP projection data set of the prostate artery;

[0093] According to the same scanning direction, each MIP projection image in the MIP projection data set is fused to reconstruct a MIP image of the prostate artery;

[0094] S3. Obtain interventional radiological scanning images of the prostate artery through magnetic resonance scanning, and construct interventional arterial images of the prostate artery based on three-dimensional reconstruction and vascular segmentation technology;

[0095] S4. Identify and determine whether the prostate artery vascular features of the MIP image and the interventional artery image in the axial, coronary and sagittal planes can be automatically compared successfully:

[0096] If yes, go to S5;

[0097] If not, go to S6;

[0098] S5, outputting the MIP image and displaying it on a front-end page of the operating room for guiding prostate artery embolization;

[0099] S6. Using a preset prostate artery recognition AI model, the prostate artery image features in the interventional artery image are identified and located and fused on the MIP image, and the located and fused MIP image is output and displayed on the front page of the operating room for guiding prostate artery embolization.

[0100] When performing prostate embolization surgery in traditional hospitals, it is necessary to observe the anatomical position of the prostate artery and whether there are any variations. For this reason, all patients must undergo an MRI scan of the prostate artery. However, relying solely on MRI, it is impossible to clearly distinguish the veins, direction, origin, and other information of the prostate artery.

[0101] The advantages of interventional scanning MIP (maximum intensity projection) technology in arterial vascular imaging include:

[0102] 1). Display of vascular structure: MIP can clearly display the three-dimensional structure of blood vessels, helping doctors better understand the direction and distribution of blood vessels.

[0103] 2). Improve contrast: By enhancing the contrast agent signal in the blood vessels, MIP technology can increase the contrast between blood vessels and surrounding tissues, making the vascular image more prominent.

[0104] 3). Reduce image noise: MIP technology reduces noise in the image and improves image quality by selectively projecting the strongest signal in the image.

[0105] 4). Rapid imaging: MIP technology can complete the imaging of a wide range of blood vessels in a shorter time, thus improving the inspection efficiency.

[0106] 5). Diagnostic assistance: MIP technology can provide important auxiliary information for the diagnosis of vascular diseases such as atherosclerosis, vascular stenosis, and hemangioma.

[0107] 6). Treatment planning: Before interventional treatment, MIP imaging can help doctors plan the surgical path, evaluate the condition of diseased blood vessels, and provide accurate anatomical information for surgery.

[0108] 7). Non-invasiveness: Compared with traditional angiography, MIP technology is a non-invasive imaging method that reduces patient discomfort and potential risks.

[0109] Therefore, the present invention automatically compares all the original data of magnetic resonance scanning and the original data of interventional radiology after reconstruction to meet the automatic comparison between the two. If the comparison between the two is successful, the artery can be directly embolized during the operation to meet the diagnosis of clinical surgery.

[0110] First, it is necessary to: S1. obtain the original vascular image of the prostatic artery through magnetic resonance scanning and perform anatomical preprocessing to obtain a pelvic MRI post-processing image of the prostatic artery.

[0111] The acquisition of the original MRI image and its preprocessing can be handled by the administrator to obtain the corresponding Figure 1 The original MRI image is shown.

[0112] Next, we will reconstruct the MIP image of the prostate artery:

[0113] S2, performing image MIP reconstruction on the pelvic MRI post-processed image based on the MIP algorithm, and obtaining a MIP image of the prostate artery after reconstruction, including:

[0114] Analyzing the MRI post-processed image of the prostate artery and blood vessels to obtain MRI scanning data in different scanning directions;

[0115] Traversing the MRI scan data, reading the MRI scan data of the prostate artery in different scanning directions, and calculating the maximum three-dimensional pixel value, and saving the value to obtain the MRI scan data set of the prostate artery;

[0116] Based on the MIP algorithm, performing MIP projection imaging on the MRI scan data set in axial, coronal and sagittal scanning directions to generate a MIP projection data set of the prostate artery;

[0117] In the same scanning direction, each MIP projection image in the MIP projection data set is fused to reconstruct a MIP image of the prostate artery.

[0118] The MIP algorithm, namely Maximum Intensity Projection, is a 3D reconstruction technology commonly used in medical image processing. Its steps are as follows:

[0119] 1. Data acquisition: First, obtain two-dimensional tomographic image data of blood vessels through imaging equipment such as MRI or MRI.

[0120] 2. Image preprocessing: Perform preprocessing operations such as denoising and contrast enhancement on the collected images to improve the accuracy of subsequent processing.

[0121] 3. Segment blood vessels: Use image segmentation techniques such as threshold segmentation and edge detection to extract blood vessel areas from the preprocessed image.

[0122] 4. Voxel intensity calculation: Calculate the intensity value of each voxel (three-dimensional pixel), usually taking the maximum intensity value of the area where the voxel is located.

[0123] 5. Projection calculation: Starting from the user-specified perspective, the 3D data is projected, and only the voxel values ​​with the highest intensity along the line of sight are retained, and other voxel values ​​are ignored.

[0124] 6. Image reconstruction: Combine the projected two-dimensional images to form the final three-dimensional vascular image.

[0125] The advantages of the MIP algorithm in vascular image reconstruction include:

[0126] 1. Simple and efficient: The MIP algorithm is relatively simple and computationally efficient, making it suitable for rapid reconstruction of vascular images.

[0127] 2. Display of blood vessel structure: It can clearly display the three-dimensional structure of blood vessels, making it easier for doctors to observe the direction and shape of blood vessels.

[0128] 3. Contrast enhancement: Since only the maximum intensity value is displayed, the MIP algorithm can enhance the contrast between blood vessels and surrounding tissues, making the blood vessels more prominent.

[0129] 4. Diagnostic assistance: It helps doctors diagnose vascular diseases, such as vascular stenosis and aneurysm.

[0130] 5. Non-invasive: As a non-invasive imaging technique, the MIP algorithm does not cause additional harm to patients.

[0131] Therefore, the present invention utilizes the MIP algorithm to construct a MIP image (three-dimensional vascular image) of the prostate artery, thereby obtaining a high-definition vascular map of the prostate artery, which is used for the next step of image interventional contrast research.

[0132] First, it is necessary to read the MRI scan data of the prostate artery in different scanning directions, calculate the maximum three-dimensional pixel value, and save it to obtain the MRI scan data set of the prostate artery; then based on the MIP algorithm, perform MIP projection imaging on the MRI scan data set in the axial, coronal and sagittal scanning directions to generate a MIP projection data set of the prostate artery; finally, according to the same scanning direction, fuse the various MIP projection images in the MIP projection data set to reconstruct the MIP image of the prostate artery. Specifically: the MRI vascular scan data needs to be preprocessed, including denoising, correction and standardization steps to ensure data quality. Next, the maximum intensity projection (MIP) algorithm is applied to process the preprocessed data. The specific steps are as follows:

[0133] 1. Choose an appropriate threshold to distinguish blood vessels from other tissues.

[0134] 2. Traverse the 3D data set and calculate the maximum intensity value of each voxel along the preset projection direction (such as coronal, sagittal or axial).

[0135] 3. Based on the calculation results, a two-dimensional MIP image is generated, which shows the maximum intensity value of all voxels along the projection direction.

[0136] 4. Post-process the generated MIP image, including contrast adjustment, edge enhancement, etc., to improve image quality.

[0137] 5. Finally, the MIP images are analyzed and interpreted to reconstruct the three-dimensional structure of the blood vessels.

[0138] In order to clearly project the vascular images in each scanning direction, the present invention adopts MIP projection imaging in the axial, coronal and sagittal scanning directions, and finally fuses the two-dimensional and generates a three-dimensional arterial vascular image:

[0139] MIP images in a single direction may not be sufficient to fully display the vascular structure. Therefore, MIP projections can be performed along multiple directions (such as X, Y, Z axes and their combinations) to generate a series of MIP images. These images can provide detailed information about the blood vessels from different angles, which helps doctors or researchers to more fully understand the structure and direction of the blood vessels. Fusion of MIP images in multiple directions can create a more comprehensive three-dimensional view. This usually involves the use of professional medical image processing software or visualization tools. The fused images can be interactively viewed through operations such as rotation, zooming, and translation, allowing the observer to observe the vascular structure from different angles and distances. In addition to visual evaluation, MIP images can also be quantitatively analyzed. For example, parameters such as the diameter, length, and volume of the blood vessels can be measured. These can be completed by the administrator to facilitate subsequent prostate vascular diagnostic evaluation and the formulation of corresponding diagnosis and treatment plans. These parameters are of great significance for evaluating the health of the blood vessels, the degree of lesions, and formulating treatment plans.

[0140] After completing the reconstruction and analysis of the MIP image, a detailed report needs to be written. The report should contain the patient's basic information, scanning parameters, image processing process, MIP image results, and analysis conclusions. The report should convey information clearly and accurately so that doctors or researchers can quickly understand the patient's vascular status and make corresponding decisions. The processed MIP images and their related data are stored in a secure and reliable system for future review or sharing with other medical institutions.

[0141] As a preferred embodiment of the present invention, preferably, S3, obtaining an interventional radiological scanning image of the prostatic artery through magnetic resonance scanning, and constructing an interventional arterial image of the prostatic artery based on three-dimensional reconstruction and vascular segmentation technology, includes:

[0142] Performing MRI scanning on the prostate artery and blood vessels to obtain MRI scanning data corresponding to the prostate artery and blood vessels;

[0143] Annotating the MRI scan data of the prostate artery with vascular contour data, and converting the data to generate a two-dimensional original scan image of the prostate artery obtained under MRI scanning;

[0144] Combining a volume rendering three-dimensional reconstruction algorithm, the two-dimensional original scanned images under each scanning sequence are converted into three dimensions, the prostate artery and blood vessels are reconstructed, and a corresponding three-dimensional model of the prostate artery and blood vessels is generated;

[0145] Using the preset prostate artery segmentation model, the image segmentation processing of the prostate artery three-dimensional model in the axial, coronary and sagittal planes is performed respectively, and the prostate artery images of the prostate artery three-dimensional model in the axial, coronary and sagittal planes are obtained to obtain the magnetic resonance artery MIP image of the prostate artery.

[0146] We use interventional scanning technology to perform angiographic interventional scanning of the prostate artery and perform intelligent image processing on the interventional scanning images to identify and extract two-dimensional scanning images of the prostate artery.

[0147] Next, the three-dimensional reconstruction technology is used to perform three-dimensional synthesis of the two-dimensional images in different scanning directions to construct a three-dimensional model of the prostatic artery. The prostatic artery segmentation model is then used to perform image segmentation processing on the three-dimensional model of the prostatic artery in the axial, coronary and sagittal planes to obtain an interventional arterial image of the prostatic artery. Therefore, the interventional arterial image of the prostatic artery is obtained from the three-dimensional image in three scanning directions, which is used for image contrast calibration to avoid the defects of unclear location contrast and inaccurate position caused by single image contrast.

[0148] After obtaining the interventional radiology scan image, image preprocessing is first required, including steps such as denoising and contrast enhancement to improve image quality. Next, 3D reconstruction technology is applied to convert the 2D scan image into a 3D model. Then, the structure of the arterial blood vessels is accurately extracted from the 3D model using vascular segmentation technology. This usually involves threshold segmentation, edge detection, or model-based segmentation methods. Ultimately, through these steps, a clear 3D model of the arterial blood vessels can be constructed for further diagnosis or surgical planning.

[0149] In the 3D reconstruction stage, a variety of algorithms can be used, such as volume rendering or surface rendering. Volume rendering methods can display details in the entire volume data, while surface rendering focuses more on extracting and displaying the outer surface of the data. For the 3D reconstruction of arterial blood vessels, the surface rendering method may be more suitable because it can outline the contours of the blood vessels more clearly.

[0150] In the blood vessel segmentation stage, since arterial blood vessels may appear as complex network structures in images, and their shapes, sizes, and positions may vary significantly between individuals, it is crucial to choose an appropriate segmentation method. A commonly used method is a region-based segmentation method, such as region growing or level set methods. These methods can start from one or more seed points and iteratively expand the segmentation region according to the similarity of pixels or voxels (such as grayscale value, texture, etc.) until the entire artery is covered.

[0151] In addition, deep learning technology can also be used to improve the accuracy and efficiency of blood vessel segmentation. For example, a convolutional neural network (CNN) can be used to automatically identify and segment arterial blood vessels in an image. By training a large amount of labeled data, CNN can learn the characteristic representation of arterial blood vessels and accurately identify arterial blood vessels in new images.

[0152] Finally, after constructing the 3D model of the artery, it can be further analyzed and processed using visualization tools. For example, the diameter, length, branch angle and other parameters of the blood vessel can be calculated to evaluate its morphology and functional status. In addition, the 3D model can be combined with clinical information to provide doctors with more comprehensive diagnostic basis and surgical guidance.

[0153] Compared with the traditional method of PAE guidance that relies solely on prostate artery MRI images, the present invention uses vascular segmentation technology to accurately extract the structure of arterial vessels from the three-dimensional model, and can construct a clear three-dimensional model of arterial vessels for further diagnosis or surgical planning. When implementing PAE, it can allow doctors to clearly observe the three-dimensional direction and structure of the target blood vessels of the prostate artery, avoiding ambiguity and confusion caused by vascular variation, overlap, and intersection, accurately guiding doctors to specific prostate artery target vessels, and saving time for doctors to implement PAE.

[0154] As a preferred embodiment of the present invention, preferably, the method for generating the prostate artery segmentation model comprises:

[0155] The MRI image sets of the prostate artery vessels in the axial, coronary and sagittal planes were collected;

[0156] The axial, coronal and sagittal MRI images of the prostate artery were preprocessed respectively;

[0157] Performing labeling processing on the prostate artery vascular MRI images in the axial, coronary and sagittal planes in the prostate artery vascular MRI image set, and determining the boundary positions of the prostate artery vascular in the axial, coronary and sagittal planes;

[0158] After labeling is completed, a training set is obtained, and the training set is imported into a preset convolutional neural network model to perform labeled data training and learning, thereby obtaining the prostate artery segmentation model;

[0159] Using the validation sets of prostate artery vascular images in the axial, coronary and sagittal planes, respectively, the prostate artery segmentation model is segmented and validated;

[0160] After verification, the prostate artery segmentation model is deployed on the background server.

[0161] The MRI image sets of the prostate arteries and vessels in the axial, coronal and sagittal planes can be obtained by the administrator from the background database of the diagnosis and treatment system (the diagnostic data of a large number of prostate patients on a daily basis include these data).

[0162] The specific method for constructing the prostate artery segmentation model includes the following steps:

[0163] 1. Data preparation: Collect medical imaging data of prostate arteries, such as MRI or MRI scan images, and ensure that the data is of sufficient quality and diversity.

[0164] 2. Preprocessing: Preprocess the collected image data, including denoising, standardization, contrast enhancement, etc., to improve the accuracy and efficiency of subsequent processing.

[0165] 3. Labeling: Professional radiologists or medical imaging experts manually label the image data to determine the exact location of the prostatic artery.

[0166] 4. Model training: Use deep learning algorithms, such as convolutional neural networks (CNNs), to train a prostate artery segmentation model based on the labeled data. The U-Net architecture can be used because it performs well in medical image segmentation tasks.

[0167] 5. Model validation: Use the validation set to test the trained model, evaluate its segmentation performance, and adjust and optimize the model if necessary.

[0168] 6. Model deployment: Deploy the trained model to actual applications for automatic segmentation of prostate arteries.

[0169] After the model is deployed, the steps that can be used to perform image segmentation processing on the prostate artery vascular 3D model include:

[0170] 1. Input the 3D model: Input the 3D model of the prostate artery into the trained segmentation model.

[0171] 2. Segmentation: The model automatically identifies and segments the prostate arteries based on the learned features.

[0172] 3. Result output: The model outputs segmented images, which correspond to the prostate artery images in the axial, coronal and sagittal planes.

[0173] 4. Post-processing: Post-process the segmentation results, such as smoothing edges, filling holes, etc., to improve image quality.

[0174] Specifically understand it in combination with the above image segmentation method.

[0175] Traditional vascular image segmentation techniques mainly rely on manual or semi-automatic methods, including threshold segmentation, edge detection, region growing, level set method, active contour model, etc. When processing vascular images, these techniques usually require image preprocessing, such as denoising, contrast enhancement, etc., to improve segmentation accuracy. Although these methods can provide good segmentation results in some cases, they often have high requirements on image quality and may not be robust enough when dealing with complex vascular structures.

[0176] During model training, the present invention uses a deep learning method to achieve automatic segmentation of vascular images. By marking the boundary positions of the prostate artery in the axial, coronary and sagittal planes, the prostate artery segmentation model generated by training can perform image segmentation processing on the prostate artery three-dimensional model in the axial, coronary and sagittal planes, and obtain the prostate artery images of the prostate artery three-dimensional model in the axial, coronary and sagittal planes (prostate artery images constructed by vascular boundaries).

[0177] Therefore, by adopting the above-mentioned model segmentation, it is possible to quickly locate and identify the boundaries of the prostate artery vascular images in the axial, coronary and sagittal planes, quickly segment the images, and improve the processing efficiency.

[0178] As a preferred embodiment of the present invention, preferably, S4, identifying and judging whether the prostate artery vascular features of the MIP image and the interventional artery image in the axial, coronary and sagittal planes can be automatically compared successfully, comprises:

[0179] Extracting the first prostatic artery vascular features of the MIP image in the axial, coronary and sagittal planes respectively;

[0180] Respectively extracting second prostate artery vascular features of the prostate artery vascular image in the axial, coronary and sagittal planes of the prostate artery vascular three-dimensional model;

[0181] Automatic comparison and judgment:

[0182] (1) performing feature comparison between the first prostate artery vascular feature and the second prostate artery vascular feature in the axial position;

[0183] (2) performing feature comparison between the first prostate artery vascular feature and the second prostate artery vascular feature in the coronal plane;

[0184] (3) performing feature comparison between the first prostate artery vascular feature and the second prostate artery vascular feature in the sagittal plane;

[0185] If (1)-(3) can all be successfully compared, it is considered that the prostate artery vascular features of the MIP image and the interventional artery image in the axial, coronary and sagittal planes can be automatically compared successfully;

[0186] Otherwise it fails.

[0187] The present invention automatically compares all the original data of magnetic resonance scanning and the original data of interventional radiology after reconstruction to meet the automatic comparison between the two. If the comparison between the two is successful, the artery can be directly embolized during the operation to meet the diagnosis of clinical surgery.

[0188] Therefore, it is necessary to compare the MIP image with the interventional arterial image, and the specific comparison basis is: comparison of the vascular characteristics of the prostate artery in the axial, coronary and sagittal planes of the MIP image and the interventional arterial image respectively (the above (1)-(3)).

[0189] Extracting the prostate artery vascular features on the prostate artery vascular image may be performed by the following steps:

[0190] 1. Image preprocessing: First, the image is preprocessed by denoising, contrast enhancement, etc. to improve the visibility of vascular features.

[0191] 2. Vessel segmentation: Use image segmentation techniques such as threshold segmentation, region growing, edge detection, or model-based methods to identify and separate the prostate arteries.

[0192] 3. Feature extraction: Apply feature extraction algorithms to the segmented vascular images, such as morphological features (vascular width, length, number of branches, etc.), texture features (texture pattern on the vascular surface), and geometric features (vascular curvature, angle, etc.).

[0193] 4. Analysis and identification: The extracted features are analyzed using machine learning or deep learning methods to identify and classify different vascular characteristics, such as normal and abnormal vessels.

[0194] Image feature comparison method: You can use the following methods to perform feature comparison:

[0195] 1. Feature extraction: First, extract key feature points or regions from the two images. This can be achieved through edge detection, corner detection, SIFT (Scale Invariant Feature Transform), SURF (Speeded Up Robust Features) and other algorithms.

[0196] 2. Feature description: Describe the extracted feature points for easy comparison. The descriptor can be in vector form, such as SIFT feature descriptor.

[0197] 3. Feature matching: Match feature points in one image with feature points in another image. This is usually done by calculating the similarity between feature descriptors, such as using Euclidean distance or Hamming distance (feature image similarity, or cosine similarity).

[0198] 4. Confirm the match: Remove false matches and retain the correct matching point pairs by setting a threshold or using algorithms such as RANSAC (Random Sampling Consensus).

[0199] 5. Geometric transformation estimation: Use matching point pairs to calculate the geometric transformation relationship between two images, such as affine transformation or homography matrix.

[0200] 6. Image alignment: Transform one image to align it with another image based on the estimated geometric transformation.

[0201] 7. Result evaluation: The contrast effect is evaluated by comparing the similarity between the transformed image and the target image. Indicators such as mean square error (MSE) and structural similarity index (SSIM) can be used.

[0202] Specifically, the administrator can randomly select the above-mentioned methods for feature comparison. If the first prostate artery vascular feature and the second prostate artery vascular feature in the axial, coronal and sagittal planes of the MIP image match (for example, the similarity meets 95% or more), the comparison between the two is considered successful (the judgment conditions (1)-(3) above are all met). Otherwise, it is unsuccessful.

[0203] If the authentication is successful, the MIP image (the MIP algorithm is used to construct the MIP image (three-dimensional vascular image) of the prostate artery to obtain a high-definition vascular path diagram of the prostate artery, which will be used for the next step of image interventional comparative study) will be output and displayed on the front page of the operating room to guide prostate artery embolization. The MIP image displayed here will be a three-dimensional vascular image of the prostate artery reconstructed based on the MIP algorithm, so it will be displayed on the front page of the operating room (generally the display page of the PC terminal in the operating room) to guide doctors in performing prostate surgery and to facilitate the patient's prostate vascularity.

[0204] As a preferred embodiment of the present invention, preferably, S6, using a preset prostate artery recognition AI model, identifying the prostate artery image features in the interventional artery image and merging them on the MIP image, and outputting and displaying the positioned fused MIP image on the front page of the operating room for guiding the prostate artery embolization, comprises:

[0205] S600, performing image enhancement processing on the interventional arterial image to obtain an arterial vascular interventional enhanced image, and importing the arterial vascular interventional enhanced image into a preset prostate artery recognition AI model, wherein the prostate artery recognition AI model recognizes and marks the prostate artery image features in the axial, coronary and sagittal positions of the arterial vascular interventional enhanced image;

[0206] S601, based on image registration and fusion, positioning and fusing the prostate artery image features on the MIP image, forming contrast features on the axial, coronal and sagittal planes with the original prostate artery blood vessels on the MIP image, and generating vascular tomography contrast features of the positioned and fused prostate artery blood vessels on the MIP image;

[0207] S602: Output the interventional arterial image of the located and fused prostate artery and display it on the front page of the operating room for guiding the prostate artery embolization.

[0208] If it is found that the feature comparison between the interventional image and the MIP image fails, the prostate artery image features in the interventional arterial image (same scanning direction, such as axial position) are used to enhance and supplement the prostate artery vascular image at the corresponding position in the MIP image. In this way, the image advantages of the interventional image are combined to form image complementary features for the MIP image. The original data of all magnetic resonance scans and the original data of the interventional radiology department during intervention are automatically compared after reconstruction to meet the automatic comparison between the two, so as to make the prostate veins clearer, facilitate doctors to understand the patient's vascular direction and other conditions, and facilitate surgery.

[0209] Image enhancement processing usually includes:

[0210] 1. Edge enhancement: Use edge detection algorithms (such as Sobel, Canny, Prewitt, etc.) to highlight the edges of blood vessels in order to better identify the contours of blood vessels.

[0211] 2. Multi-scale and multi-modal fusion: If there are multi-modal images (such as MRI, MRI, etc.), image fusion can be performed to obtain more comprehensive vascular information.

[0212] 3. Deep learning method: Use deep learning models such as convolutional neural networks (CNN) to automatically learn the features of image enhancement (the administrator prepares the corresponding interventional arterial images after artificial enhancement in advance) and enhance the image.

[0213] The construction method of the prostate artery recognition AI model here can be understood in combination with the generation method of the above-mentioned "prostate artery segmentation model", except that the corresponding training data set prepared is: interventional enhanced images of arterial vessels in the axial, coronary and sagittal planes. The model training steps can be combined with the application principle of CNN (convolutional neural network model) for training and learning, and this embodiment will not be repeated.

[0214] Both MIP images and interventional images can be benchmarked and positioned in the same coordinate system, and the extracted prostate artery image features in the axial, coronary and sagittal planes of the arterial interventional enhanced image can be fused to the corresponding positions on the corresponding MIP image according to the preset positioning benchmark, so as to cover and replace the original prostate artery feature images in the axial, coronary and sagittal planes on the MIP image, thereby achieving interventional replacement and coverage of the corresponding vascular features and enhancing its venous visibility.

[0215] Through the above processing, a vascular tomography contrast feature of the prostate artery blood vessels after positioning and fusion is generated on the MIP image, which allows the doctor to clearly see the patient's prostate artery veins on the MIP image. Therefore, the MIP image of the prostate artery blood vessels after positioning and fusion can be output and displayed on the front page of the operating room for guiding prostate artery embolization.

[0216] In actual operation, image enhancement processing is usually performed by professional medical imaging software or image processing software, such as MATLAB, ImageJ, Fiji, etc. In some cases, dedicated medical imaging workstations can also be used, which usually integrate a variety of image processing and analysis tools.

[0217] In another aspect, a prostate artery embolization guiding device is provided.

[0218] Figure 3 The block diagram of a prostate artery embolization guidance device according to an exemplary embodiment is shown, and the device is used for a PAE guidance method for reconstructing MIP images after magnetic resonance imaging processing and interventional reconstruction. Figure 3 The prostate artery embolization guidance device is used to implement a PAE guidance method for reconstructing MIP images after magnetic resonance imaging processing and interventional reconstruction, and the device comprises:

[0219] (1) An MRI image processing system 310, for acquiring an original vascular image of the prostatic artery and blood vessels through magnetic resonance scanning and performing anatomical preprocessing to obtain a pelvic MRI post-processing image of the prostatic artery and blood vessels;

[0220] (2) A MIP image processing system 320, for performing MIP reconstruction on the pelvic MRI post-processed image based on a MIP algorithm, and obtaining a MIP image of the prostate artery after reconstruction, including:

[0221] Analyzing the MRI post-processed image of the prostate artery and blood vessels to obtain MRI scanning data in different scanning directions;

[0222] Traversing the MRI scan data, reading the MRI scan data of the prostate artery in different scanning directions, and calculating the maximum three-dimensional pixel value, and saving the value to obtain the MRI scan data set of the prostate artery;

[0223] Based on the MIP algorithm, performing MIP projection imaging on the MRI scan data set in axial, coronal and sagittal scanning directions to generate a MIP projection data set of the prostate artery;

[0224] According to the same scanning direction, each MIP projection image in the MIP projection data set is fused to reconstruct a MIP image of the prostate artery;

[0225] (3) an interventional arterial image processing system 330, for obtaining interventional radiological scan images of prostate arterial vessels through magnetic resonance scanning, and constructing interventional arterial images of prostate arterial vessels based on three-dimensional reconstruction and vessel segmentation technology;

[0226] (4) An artery feature recognition system 340 is used to recognize and determine whether the prostate artery vascular features of the MIP image and the interventional artery image in the axial, coronary and sagittal planes can be automatically compared successfully:

[0227] If yes, the MIP image is output and displayed on the front page of the operating room for guiding the prostate artery embolization;

[0228] If not, the preset prostate artery recognition AI model is used to identify the prostate artery image features in the interventional artery image and position and fuse them on the MIP image, and the positioned and fused MIP image is output and displayed on the front page of the operating room to guide the prostate artery embolization.

[0229] (5) an operating room front end 350, for displaying the MIP image;

[0230] The MRI image processing system 310 is communicatively connected with the MIP image processing system 320;

[0231] The MIP image processing system 320 and the interventional artery image processing system 330 are respectively connected to the artery feature recognition system 340 for communication;

[0232] The artery feature recognition system 340 is in communication with the operating room front end 350 .

[0233] The functions and interactions of the above-mentioned systems or front ends should be understood in conjunction with the corresponding steps in the above-mentioned method, and will not be described in detail in this embodiment.

[0234] Figure 4 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 4 As shown, the electronic device may include the above Figure 3 The prostate artery embolization guiding device is shown. Optionally, the electronic device 410 may include a first processor 2001 .

[0235] Optionally, the electronic device 410 may further include a memory 2002 and a transceiver 2003 .

[0236] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.

[0237] Combine the following Figure 4 The components of the electronic device 410 are described in detail:

[0238] The first processor 2001 is the control center of the electronic device 410, and may be a processor or a general term for multiple processing elements. For example, the first processor 2001 is one or more central processing units (CPUs), or may be application specific integrated circuits (ASICs), or may be one or more integrated circuits configured to implement the embodiments of the present invention, such as one or more microprocessors (digital signal processors, DSPs), or one or more field programmable gate arrays (field programmable gate arrays, FPGAs).

[0239] Optionally, the first processor 2001 can execute various functions of the electronic device 410 by running or executing a software program stored in the memory 2002 and calling data stored in the memory 2002 .

[0240] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 4 CPU0 and CPU1 are shown in FIG.

[0241] In a specific implementation, as an embodiment, the electronic device 410 may also include multiple processors, such as Figure 4 The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors may be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0242] The memory 2002 is used to store the software program for executing the solution of the present invention, and is controlled to be executed by the first processor 2001. The specific implementation method can refer to the above method embodiment, which will not be repeated here.

[0243] Optionally, the memory 2002 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 2002 may be integrated with the first processor 2001, or may exist independently and access the first processor 2001 through the interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0244] The transceiver 2003 is used to communicate with a network device or a terminal device.

[0245] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 4 The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0246] Optionally, the transceiver 2003 may be integrated with the first processor 2001, or may exist independently and communicate with the first processor 2001 through the interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0247] It should be noted that Figure 4 The structure of the electronic device 410 shown in the figure does not constitute a limitation on the router, and the actual knowledge structure recognition device may include more or fewer components than those shown in the figure, or combine certain components, or arrange the components differently.

[0248] In addition, the technical effects of the electronic device 410 can refer to the technical effects of the magnetic resonance post-processing reconstruction MIP map and the PAE guidance method for interventional reconstruction described in the above method embodiments, which will not be repeated here.

[0249] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0250] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct memory bus random access memory (DIR RAM).

[0251] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0252] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0253] In the present invention, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0254] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0255] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0256] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0257] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0258] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0259] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0260] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0261] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A PAE-guided method for reconstructing MIP images after magnetic resonance imaging post-processing and interventional reconstruction, characterized in that: The method comprises: S1, obtaining the original vascular image of the prostatic artery by magnetic resonance scanning and performing anatomical preprocessing to obtain a pelvic MRI post-processing image of the prostatic artery; S2, performing image MIP reconstruction on the pelvic MRI post-processed image based on the MIP algorithm, and obtaining a MIP image of the prostate artery after reconstruction, including: Analyzing the MRI post-processed image of the prostate artery and blood vessels to obtain MRI scanning data in different scanning directions; Traversing the MRI scan data, reading the MRI scan data of the prostate artery in different scanning directions, and calculating the maximum three-dimensional pixel value, and saving the value to obtain the MRI scan data set of the prostate artery; Based on the MIP algorithm, performing MIP projection imaging on the MRI scan data set in axial, coronal and sagittal scanning directions to generate a MIP projection data set of the prostate artery; According to the same scanning direction, each MIP projection image in the MIP projection data set is fused to reconstruct a MIP image of the prostate artery; S3. Obtain interventional radiological scanning images of the prostate artery through magnetic resonance scanning, and construct interventional arterial images of the prostate artery based on three-dimensional reconstruction and vascular segmentation technology; S4. Identify and determine whether the prostate artery vascular features of the MIP image and the interventional artery image in the axial, coronary and sagittal planes can be automatically compared successfully: If yes, go to S5; If not, go to S6; S5, outputting the MIP image and displaying it on a front-end page of the operating room for guiding prostate artery embolization; S6. Using a preset prostate artery recognition AI model, the prostate artery image features in the interventional artery image are identified and located and fused on the MIP image, and the located and fused MIP image is output and displayed on the front page of the operating room for guiding prostate artery embolization.

2. The PAE-guided method for reconstructing MIP images by post-processing magnetic resonance imaging and interventional reconstruction according to claim 1, characterized in that: S6. Using a preset prostate artery recognition AI model, identifying the prostate artery image features in the interventional artery image and merging them on the MIP image, and outputting and displaying the positioned fused MIP image on the front page of the operating room for guiding prostate artery embolization, including: S600, performing image enhancement processing on the interventional arterial image to obtain an arterial vascular interventional enhanced image, and importing the arterial vascular interventional enhanced image into a preset prostate artery recognition AI model, wherein the prostate artery recognition AI model recognizes and marks the prostate artery image features in the axial, coronary and sagittal positions of the arterial vascular interventional enhanced image; S601, based on image registration and fusion, positioning and fusing the prostate artery image features on the MIP image, forming contrast features on the axial, coronal and sagittal planes with the original prostate artery blood vessels on the MIP image, and generating vascular tomography contrast features of the positioned and fused prostate artery blood vessels on the MIP image; S602: Output the interventional arterial image of the located and fused prostate artery and display it on the front page of the operating room for guiding the prostate artery embolization.

3. The PAE-guided method for magnetic resonance post-processing reconstruction of MIP images and interventional reconstruction according to claim 1, characterized in that: S3. Obtain interventional radiological scan images of the prostate artery through magnetic resonance scanning, and construct interventional arterial images of the prostate artery based on three-dimensional reconstruction and vascular segmentation technology, including: Performing MRI scanning on the prostate artery and blood vessels to obtain MRI scanning data corresponding to the prostate artery and blood vessels; Annotating the MRI scan data of the prostate artery with vascular contour data, and converting the data to generate a two-dimensional original scan image of the prostate artery obtained under MRI scanning; Combining a volume rendering three-dimensional reconstruction algorithm, the two-dimensional original scanned images under each scanning sequence are converted into three dimensions, the prostate artery and blood vessels are reconstructed, and a corresponding three-dimensional model of the prostate artery and blood vessels is generated; Using the preset prostate artery segmentation model, the image segmentation processing of the prostate artery three-dimensional model in the axial, coronary and sagittal planes is performed respectively, and the prostate artery images of the prostate artery three-dimensional model in the axial, coronary and sagittal planes are obtained to obtain the magnetic resonance artery MIP image of the prostate artery.

4. The PAE-guided method for magnetic resonance post-processing reconstruction of MIP images and interventional reconstruction according to claim 3, characterized in that: The method for generating the prostate artery segmentation model comprises: The MRI image sets of the prostate artery vessels in the axial, coronary and sagittal planes were collected; The prostate artery MRI image sets in the axial, coronal and sagittal planes were preprocessed respectively; Performing labeling processing on the prostate artery vascular MRI images in the axial, coronary and sagittal planes in the prostate artery vascular MRI image set, and determining the boundary positions of the prostate artery vascular in the axial, coronary and sagittal planes; After labeling is completed, a training set is obtained, and the training set is imported into a preset convolutional neural network model to perform labeled data training and learning, thereby obtaining the prostate artery segmentation model; Using the validation sets of prostate artery vascular images in axial, coronary and sagittal planes, respectively, the prostate artery segmentation model is segmented and validated; After verification, the prostate artery segmentation model is deployed on the background server.

5. The PAE-guided method for magnetic resonance post-processing reconstruction of MIP images and interventional reconstruction according to claim 3, characterized in that: S4, identifying and judging whether the vascular features of the prostate artery in the axial, coronary and sagittal planes of the MIP image and the interventional artery image can be automatically compared successfully, including: Extracting the first prostatic artery vascular features of the MIP image in the axial, coronary and sagittal planes respectively; Respectively extracting second prostate artery vascular features of the prostate artery vascular image in the axial, coronary and sagittal planes of the prostate artery vascular three-dimensional model; Automatic comparison and judgment: (1) performing feature comparison between the first prostate artery vascular feature and the second prostate artery vascular feature in the axial position; (2) performing feature comparison between the first prostate artery vascular feature and the second prostate artery vascular feature in the coronal plane; (3) performing feature comparison between the first prostate artery vascular feature and the second prostate artery vascular feature in the sagittal plane; If (1)-(3) can all be successfully compared, it is considered that the prostate artery vascular features of the MIP image and the interventional artery image in the axial, coronary and sagittal planes can be automatically compared successfully; Otherwise it fails.

6. A prostate artery embolization guiding device, the prostate artery embolization guiding device is used to implement the PAE guiding method of magnetic resonance post-processing reconstruction of MIP image and interventional reconstruction as described in any one of claims 1 to 5, characterized in that: The device comprises: (1) An MRI image processing system, used to obtain an original vascular image of the prostatic artery through magnetic resonance scanning and perform anatomical preprocessing to obtain a pelvic MRI post-processing image of the prostatic artery; (2) A MIP image processing system, for performing MIP reconstruction on the pelvic MRI post-processed image based on a MIP algorithm, and obtaining a MIP image of the prostate artery after reconstruction, including: Analyzing the MRI post-processed image of the prostate artery and blood vessels to obtain MRI scanning data in different scanning directions; Traversing the MRI scan data, reading the MRI scan data of the prostate artery in different scanning directions, and calculating the maximum three-dimensional pixel value, and saving the value to obtain the MRI scan data set of the prostate artery; Based on the MIP algorithm, performing MIP projection imaging on the MRI scan data set in axial, coronal and sagittal scanning directions to generate a MIP projection data set of the prostate artery; According to the same scanning direction, each MIP projection image in the MIP projection data set is fused to reconstruct a MIP image of the prostate artery; (3) An interventional arterial image processing system, which is used to obtain interventional radiological scan images of the prostate artery through magnetic resonance scanning, and to construct interventional arterial images of the prostate artery based on three-dimensional reconstruction and vascular segmentation technology; (4) An arterial feature recognition system, for recognizing and determining whether the prostate artery vascular features of the MIP image and the interventional arterial image in the axial, coronary and sagittal planes can be automatically compared successfully: If yes, the MIP image is output and displayed on the front page of the operating room for guiding the prostate artery embolization; If not, the preset prostate artery recognition AI model is used to identify the prostate artery image features in the interventional artery image and position and fuse them on the MIP image, and the positioned and fused MIP image is output and displayed on the front page of the operating room to guide the prostate artery embolization. (5) a front end of an operating room, for displaying the MIP image; The MRI image processing system is communicatively connected with the MIP image processing system; The MIP image processing system and the interventional artery image processing system are respectively connected to the artery feature recognition system for communication; The artery feature recognition system is communicatively connected to the front end of the operating room.

7. An electronic device, characterized in that: The electronic device comprises: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, which can be called by a processor to execute the method according to any one of claims 1 to 5.

Images