Focus segmentation method and device and electronic equipment

Abstract

The invention relates to a focus segmentation method and device and electronic equipment. The method includes determining at least one preset configuration corresponding to a target object and at least one preset magnetic resonance image based on each of the at least one preset object under at least one preset magnetic resonance signal and at least one preset magnetic resonance image correspondence of each preset object under at least one preset magnetic resonance signal injecting at least one target magnetic resonance signal determined by the overlapping degree of the images; performing configuration recognition on the at least one target magnetic resonance image, and determining at least one target configuration; determining at least one target segmentation model based on the at least one target magnetic resonance signal and the at least one target configuration; and inputting the at least one target magnetic resonance image into at least one target segmentation model for segmentation, and performing weighting processing on the obtained at least one lesion segmentation image to obtain a target lesion segmentation image. According to the embodiment of the invention, the lesion segmentation efficiency and accuracy can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to a lesion segmentation method, device and electronic device. Background Art

[0002] With the development of artificial intelligence, the application scenarios of artificial intelligence technology are becoming more and more extensive. For example, artificial intelligence can be combined for lesion segmentation. In related technologies, MRI image lesion segmentation is often achieved through traditional full convolution segmentation networks that mainly rely on supervised learning and manual annotation of training data sets. However, the images corresponding to different MRI magnetic resonance signals display lesions differently, and the training data set lacks data on all standard morphologies of lesions, resulting in inaccurate lesion identification and low efficiency and accuracy of lesion segmentation. Summary of the invention

[0003] The present disclosure provides a lesion segmentation method, device, electronic device and storage medium to at least solve the problems in the related art that images corresponding to different MRI magnetic resonance signals display lesions differently, and the training data set lacks data of all standard morphologies of lesions, resulting in inaccurate lesion identification and low efficiency and accuracy of lesion segmentation. The technical solution of the present disclosure is as follows:

[0004] According to a first aspect of an embodiment of the present disclosure, a lesion segmentation method is provided, comprising:

[0005] Determining at least one target magnetic resonance signal and at least one preset configuration corresponding to the target object, wherein the at least one target magnetic resonance signal is determined based on an overlap between at least one preset magnetic resonance image of each preset object in at least one preset magnetic resonance signal and a corresponding labeled image of at least one preset magnetic resonance image of each preset object in at least one preset magnetic resonance signal;

[0006] Acquiring at least one target magnetic resonance image of the target object under the at least one target magnetic resonance signal;

[0007] Based on the at least one preset configuration, performing configuration recognition on the at least one target magnetic resonance image to determine at least one target configuration corresponding to the target object;

[0008] Based on the at least one target magnetic resonance signal and the at least one target configuration, determining at least one target segmentation model from at least one preset segmentation model; the at least one preset segmentation model is a segmentation model corresponding to the at least one preset configuration under the at least one preset magnetic resonance signal;

[0009] Inputting the at least one target magnetic resonance image into the at least one target segmentation model for segmentation to obtain at least one lesion segmentation image;

[0010] Weighted processing is performed on the at least one lesion segmentation image to obtain a target lesion segmentation image corresponding to the target object.

[0011] In an optional embodiment, performing configuration recognition on the at least one target magnetic resonance image based on the at least one preset configuration to determine at least one target configuration corresponding to the target object includes:

[0012] Performing feature extraction processing on the at least one target magnetic resonance image to obtain at least one target feature information;

[0013] Acquire at least one preset sample data; the at least one preset sample data is obtained by fusing the at least one target magnetic resonance image and the at least one preset configuration;

[0014] Performing feature extraction processing on the at least one preset sample data to obtain at least one preset feature information;

[0015] Performing clustering processing on the at least one preset feature information and the at least one target feature information to obtain cluster feature information sets corresponding to each of the at least one target magnetic resonance images;

[0016] Based on the first association information between each preset feature information in each cluster feature information set and each target feature information in each cluster feature information set, configuration recognition is performed on the at least one target magnetic resonance image to obtain at least one target configuration corresponding to the target object.

[0017] In an optional embodiment, clustering the at least one preset feature information and the at least one target feature information to obtain a cluster feature information set corresponding to each of the at least one target magnetic resonance images comprises:

[0018] Determine second association information between each target feature information in the at least one target feature information and the at least one preset feature information;

[0019] Based on the second association information, clustering processing is performed on the at least one preset feature information and the at least one target feature information to obtain cluster feature information sets corresponding to each of the at least one target magnetic resonance images.

[0020] In an optional embodiment, the method further includes:

[0021] Acquiring a target annotated image corresponding to the at least one target magnetic resonance image;

[0022] determining, based on the at least one target magnetic resonance image and the target annotated image, a degree of overlap between the at least one target magnetic resonance image and the target annotated image;

[0023] The performing weighted processing on the at least one lesion segmentation image to obtain a target lesion segmentation image corresponding to the target object comprises:

[0024] Performing weighted merging processing on the lesion segmentation images under the same magnetic resonance signal in the at least one lesion segmentation image to obtain at least one lesion segmentation image under the at least one target magnetic resonance signal;

[0025] Normalizing the overlap between the at least one target magnetic resonance image and the target annotated image to obtain weights corresponding to the at least one lesion segmentation image under the at least one target magnetic resonance signal;

[0026] Based on the weights corresponding to the at least one lesion segmentation image under the at least one magnetic resonance signal, weighted processing is performed on the at least one lesion segmentation image under the at least one target magnetic resonance signal to obtain a target lesion segmentation image corresponding to the target object.

[0027] In an optional embodiment, determining the degree of overlap between the at least one target magnetic resonance image and the target annotated image based on the at least one target magnetic resonance image and the target annotated image comprises:

[0028] performing normalization processing on the at least one target magnetic resonance image to obtain at least one normalized target magnetic resonance image;

[0029] Performing filtering processing on the at least one normalized target magnetic resonance image to obtain at least one characteristic image;

[0030] Based on the overlap between the at least one feature image and the target annotated image, the overlap between the at least one target magnetic resonance image and the target annotated image is determined.

[0031] In an optional embodiment, the method further includes:

[0032] Determining at least one magnetic resonance signal to be selected corresponding to each preset object based on an overlap between at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal and a corresponding labeled image of at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal;

[0033] At least one target magnetic resonance signal corresponding to the target object is determined based on the at least one to-be-selected magnetic resonance signal corresponding to each preset object.

[0034] In an optional embodiment, the method further includes:

[0035] Acquire a training image set and a preset training model; the training image set includes at least one to-be-trained magnetic resonance image of each preset object in the at least one preset object under the at least one target magnetic resonance signal and an annotated image corresponding to the at least one to-be-trained magnetic resonance image of each preset object under the at least one target magnetic resonance signal;

[0036] Based on the training image set, the preset training model is trained to obtain the at least one preset segmentation model.

[0037] In an optional embodiment, the method further includes:

[0038] Acquire at least one to-be-trained magnetic resonance image and a labeled image corresponding to the target training magnetic resonance image of each preset object under the at least one target magnetic resonance signal, wherein the target training magnetic resonance image is any one of the at least one to-be-trained magnetic resonance images;

[0039] After spatially aligning at least one to-be-trained magnetic resonance image of each preset object under the at least one target magnetic resonance signal, the annotated image corresponding to the target training magnetic resonance image is used as the annotated image corresponding to the at least one to-be-trained magnetic resonance image of each preset object under the at least one preset magnetic resonance signal.

[0040] According to a second aspect of an embodiment of the present disclosure, there is provided a lesion segmentation device, comprising:

[0041] A first determination module is configured to determine at least one target magnetic resonance signal and at least one preset configuration corresponding to the target object, wherein the at least one target magnetic resonance signal is determined based on an overlap between at least one preset magnetic resonance image of each preset object in at least one preset magnetic resonance signal and a corresponding labeled image of at least one preset magnetic resonance image of each preset object in at least one preset magnetic resonance signal;

[0042] a target magnetic resonance image acquisition module, configured to acquire at least one target magnetic resonance image of the target object under the at least one target magnetic resonance signal;

[0043] A second determination module is configured to perform configuration recognition on the at least one target magnetic resonance image based on the at least one preset configuration, and determine at least one target configuration corresponding to the target object;

[0044] a third determination module, configured to determine at least one target segmentation model from at least one preset segmentation model based on the at least one target magnetic resonance signal and the at least one target configuration; the at least one preset segmentation model is a segmentation model corresponding to the at least one preset configuration under the at least one preset magnetic resonance signal;

[0045] A lesion segmentation image acquisition module, used for inputting the at least one target magnetic resonance image into the at least one target segmentation model for segmentation to obtain at least one lesion segmentation image;

[0046] The target lesion segmentation image acquisition module is used to perform weighted processing on the at least one lesion segmentation image to obtain a target lesion segmentation image corresponding to the target object.

[0047] In an optional embodiment, the second determining module includes:

[0048] a target feature information acquisition unit, configured to perform feature extraction processing on the at least one target magnetic resonance image to obtain at least one target feature information;

[0049] A preset sample data acquisition unit, used to acquire at least one preset sample data; the at least one preset sample data is obtained by fusing the at least one target magnetic resonance image and the at least one preset configuration;

[0050] A preset feature information acquisition unit, used to perform feature extraction processing on the at least one preset sample data to obtain at least one preset feature information;

[0051] A cluster feature information set acquisition unit, configured to perform clustering processing on the at least one preset feature information and the at least one target feature information to obtain cluster feature information sets corresponding to the at least one target magnetic resonance image;

[0052] The target configuration acquisition unit is used to perform configuration recognition on the at least one target magnetic resonance image based on the first association information between each preset feature information in each cluster feature information set and each target feature information in each cluster feature information set to obtain at least one target configuration corresponding to the target object.

[0053] In an optional embodiment, the cluster feature information set acquisition unit includes:

[0054] A second association information determination subunit, configured to determine second association information between each target feature information in the at least one target feature information and the at least one preset feature information;

[0055] The cluster feature information set acquisition subunit is used to perform clustering processing on the at least one preset feature information and the at least one target feature information based on the second association information to obtain the cluster feature information sets corresponding to the at least one target magnetic resonance image.

[0056] In an optional embodiment, the device further comprises:

[0057] A target annotated image acquisition module, used to acquire a target annotated image corresponding to the at least one target magnetic resonance image;

[0058] an overlap determination module, configured to determine an overlap between the at least one target magnetic resonance image and the target annotated image based on the at least one target magnetic resonance image and the target annotated image;

[0059] The target lesion segmentation image acquisition module comprises:

[0060] A lesion segmentation image unit, configured to perform weighted merging processing on the lesion segmentation images under the same magnetic resonance signal in the at least one lesion segmentation image, so as to obtain at least one lesion segmentation image under the at least one target magnetic resonance signal;

[0061] A weight acquisition unit, configured to normalize the overlap between the at least one target magnetic resonance image and the target annotated image, and obtain a weight corresponding to each of the at least one lesion segmentation images under the at least one target magnetic resonance signal;

[0062] The target lesion segmentation image acquisition unit is used to perform weighted processing on the at least one lesion segmentation image under the at least one target magnetic resonance signal based on the weight corresponding to each of the at least one lesion segmentation images under the at least one magnetic resonance signal, so as to obtain the target lesion segmentation image corresponding to the target object.

[0063] In an optional embodiment, the overlap determination module includes:

[0064] a normalized target magnetic resonance image determining unit, configured to perform normalization processing on the at least one target magnetic resonance image to obtain at least one normalized target magnetic resonance image;

[0065] At least one characteristic image acquisition unit, used for performing filtering processing on the at least one normalized target magnetic resonance image to obtain at least one characteristic image;

[0066] The overlap determination unit is used to determine the overlap between the at least one target magnetic resonance image and the target annotated image based on the overlap between the at least one feature image and the target annotated image.

[0067] In an optional embodiment, the device further comprises:

[0068] a fourth determination module, configured to determine at least one magnetic resonance signal to be selected corresponding to each preset object based on an overlap between at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal and a corresponding labeled image of at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal;

[0069] The fifth determination module is used to determine at least one target magnetic resonance signal corresponding to the target object based on the at least one to-be-selected magnetic resonance signal corresponding to each preset object.

[0070] In an optional embodiment, the device further comprises:

[0071] A first acquisition module is used to acquire a training image set and a preset training model; the training image set includes at least one to-be-trained magnetic resonance image of each preset object in the at least one preset object under the at least one target magnetic resonance signal and a labeled image corresponding to the at least one to-be-trained magnetic resonance image of each preset object under the at least one target magnetic resonance signal;

[0072] The preset segmentation model acquisition module is used to train the preset training model based on the training image set to obtain the at least one preset segmentation model.

[0073] In an optional embodiment, the device further comprises:

[0074] A second acquisition module is used to acquire at least one to-be-trained magnetic resonance image and a labeled image corresponding to the target training magnetic resonance image of each preset object under the at least one target magnetic resonance signal, wherein the target training magnetic resonance image is any one of the at least one to-be-trained magnetic resonance image;

[0075] The annotated image acquisition module is used to, after spatially aligning at least one to-be-trained magnetic resonance image of each preset object under the at least one target magnetic resonance signal, use the annotated image corresponding to the target training magnetic resonance image as the annotated image corresponding to the at least one to-be-trained magnetic resonance image of each preset object under the at least one preset magnetic resonance signal.

[0076] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement a method as described in any one of the above-mentioned first aspects.

[0077] The technical solution provided by the embodiments of the present disclosure brings at least the following beneficial effects:

[0078] Determine at least one preset configuration corresponding to the target object and at least one target magnetic resonance signal determined based on the overlap of at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal and the corresponding labeled image of at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal; obtain at least one target magnetic resonance image of the target object under at least one target magnetic resonance signal; perform configuration recognition on at least one target magnetic resonance image based on at least one preset configuration to determine at least one target configuration corresponding to the target object; determine at least one target segmentation model corresponding to at least one preset configuration under at least one preset magnetic resonance signal from at least one preset segmentation model based on at least one target magnetic resonance signal and at least one target configuration; improve the accuracy and pertinence of segmentation based on achieving a high degree of matching between the target object and at least one target segmentation model; input at least one target magnetic resonance image into at least one target segmentation model for segmentation to obtain at least one lesion segmentation image; perform weighted processing on at least one lesion segmentation image to obtain a target lesion segmentation image corresponding to the target object, which can further optimize the segmentation effect, greatly improve the efficiency and accuracy of lesion segmentation, and thus improve the reliability of diagnosis.

[0079] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] The drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute improper limitations on the present disclosure.

[0081] Figure 1 is a schematic diagram of an application environment according to an exemplary embodiment;

[0082] Figure 2 is a flow chart of a lesion segmentation method according to an exemplary embodiment;

[0083] Figure 3 is a schematic diagram of a preset training model according to an exemplary embodiment;

[0084] Figure 4 is a block diagram of a lesion segmentation device according to an exemplary embodiment;

[0085] Figure 5 It is a block diagram of an electronic device for lesion segmentation according to an exemplary embodiment. DETAILED DESCRIPTION

[0086] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings.

[0087] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar different contents, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the attached claims.

[0088] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties.

[0089] See also Figure 1 , Figure 1 is a schematic diagram of an application environment according to an exemplary embodiment. Figure 1 As shown, the application environment may include a terminal 100 and a server 200 .

[0090] In an optional embodiment, the terminal 100 can be used to provide lesion segmentation services to users. Specifically, the terminal 100 may include but is not limited to electronic devices such as smart phones, desktop computers, tablet computers, laptops, smart speakers, digital assistants, augmented reality (AR) / virtual reality (VR) devices, smart wearable devices, etc., and may also be software running on the above electronic devices, such as applications. Optionally, the operating system running on the electronic device may include but is not limited to Android, IOS, Linux, Windows, etc.

[0091] In an optional embodiment, the server 200 can be used to pre-train at least one preset segmentation model, and provide lesion segmentation support for the terminal 100 based on at least one preset segmentation model. Specifically, the server 200 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0092] In addition, it should be noted that Figure 1 What is shown is only one application environment provided by the present disclosure. In actual application, other application environments may also be included.

[0093] In the embodiments of this specification, the terminal 100 and the server 200 may be directly or indirectly connected via wired or wireless communication, which is not limited in this disclosure.

[0094] Figure 2 is a flow chart of a lesion segmentation method according to an exemplary embodiment. Figure 2 As shown, the lesion segmentation method is used in a server and includes the following steps.

[0095] In step S201 , at least one target magnetic resonance signal and at least one preset configuration corresponding to a target object are determined.

[0096] In a specific embodiment, the target object may be a provider of an image that requires lesion segmentation. Exemplarily, each preset object in the target object may be a patient.

[0097] In a specific embodiment, at least one target magnetic resonance signal corresponding to the target object can be determined based on the overlap between at least one preset magnetic resonance image of each preset object in at least one preset magnetic resonance signal and the corresponding labeled image of at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal.

[0098] In an optional embodiment, the above method may further include:

[0099] Determine at least one magnetic resonance signal to be selected corresponding to each preset object based on the overlap between at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal and the corresponding labeled image of at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal;

[0100] Determine at least one target magnetic resonance signal corresponding to a target object based on at least one magnetic resonance signal to be selected corresponding to each preset object.

[0101] In a specific embodiment, the overlap degree between at least one preset magnetic resonance image of each preset object among at least one preset object under at least one preset magnetic resonance signal and at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal corresponding to the annotation image can be the overlap degree between at least one preset feature expression image corresponding to at least one preset magnetic resonance image of each preset object among at least one preset object under at least one preset magnetic resonance signal and the high-value part of the annotation image corresponding to at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal. Specifically, at least one preset feature expression image can be obtained by performing normalization processing and filtering processing on at least one preset magnetic resonance image of each preset object among at least one preset object under at least one preset magnetic resonance signal.

[0102] In a specific embodiment, based on a preset threshold, magnetic resonance signals with an overlap degree greater than the preset threshold can be selected from at least one preset magnetic resonance signal as at least one magnetic resonance signal to be selected corresponding to each preset object. Overall statistics are performed on at least one magnetic resonance signal to be selected corresponding to at least one preset object to obtain the number of objects corresponding to each magnetic resonance signal to be selected. Based on the number of objects corresponding to each magnetic resonance signal to be selected, at least one target magnetic resonance signal is determined. Correspondingly, the target magnetic resonance signal can also be determined in combination with a preset object number threshold. When the number of objects corresponding to the magnetic resonance signal to be selected is greater than the object number threshold, determine this magnetic resonance signal to be selected as the target magnetic resonance signal.

[0103] In the above embodiments, determining at least one target magnetic resonance signal corresponding to a target object based on the magnetic resonance signal to be selected corresponding to each preset object can screen out the most targeted magnetic resonance signals for the target object, avoiding the use of irrelevant or ineffective signals for analysis, and improving the analysis efficiency and accuracy. Determining at least one target magnetic resonance signal corresponding to a target object based on at least one magnetic resonance signal to be selected corresponding to each preset object among at least one preset object comprehensively considers the situations of multiple objects, provides more comprehensive information for the target object, and improves the accuracy and comprehensiveness of lesion segmentation.

[0104] In a specific embodiment, at least one preset object can be at least one object for providing at least one preset magnetic resonance image of at least one preset object under at least one preset magnetic resonance signal and at least one annotation image corresponding to at least one preset magnetic resonance image under at least one preset magnetic resonance signal. At least one preset object can be at least one first preset object.

[0105] In a specific embodiment, at least one preset magnetic resonance signal can be set in combination with actual needs. Specifically, the at least one preset magnetic resonance signal may include: at least one imaging sequence commonly used in magnetic resonance imaging, such as T1WI sequence (T1 weighted imaging), T2WI sequence (T2 weighted imaging), T2FS sequence, etc.

[0106] In a specific embodiment, at least one preset magnetic resonance image of each preset object in at least one preset object under at least one preset magnetic resonance signal can be at least one preset magnetic resonance image obtained by detecting the lesion to be segmented of each preset object through at least one preset magnetic resonance signal. The lesion to be segmented can be an area where lesion segmentation is required, and the lesion to be segmented can correspond to different body lesion sites in combination with actual needs.

[0107] In a specific embodiment, the annotated image corresponding to at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal can be an image obtained by annotating lesions in any preset magnetic resonance image of at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal. Specifically, the annotated image corresponding to at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal can be obtained by the annotator using a binary annotation form to mark the exact position of the lesion area on the magnetic resonance image, and the voxel value of the lesion area in the annotated image can be 1.

[0108] In a specific embodiment, the degree of overlap between at least one preset magnetic resonance image of each preset object in at least one preset object under at least one preset magnetic resonance signal and the corresponding annotated image of at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal may be the degree of overlap of the high-value part. Specifically, it may be obtained by first normalizing and filtering at least one preset magnetic resonance image of each preset object in at least one preset object under at least one preset magnetic resonance signal to obtain at least one preset feature expression image, and then calculating the intersection area of ​​the at least one preset feature expression image and the high-value part of the annotated image corresponding to the at least one preset magnetic resonance image under at least one preset magnetic resonance signal.

[0109] In a specific embodiment, at least one preset configuration may be at least one form of the lesion to be segmented. Specifically, when the lesion to be segmented is anal fistula, at least one preset configuration may include four configurations of anal fistula: low simple type, low complex type, high simple type and high complex type.

[0110] In step S203, at least one target magnetic resonance image of the target object under at least one target magnetic resonance signal is acquired.

[0111] In a specific embodiment, the at least one target magnetic resonance image may be at least one target magnetic resonance image obtained by detecting the lesion to be segmented of the target object through at least one target magnetic resonance signal.

[0112] In step S205, based on at least one preset configuration, configuration recognition is performed on at least one target magnetic resonance image to determine at least one target configuration corresponding to the target object.

[0113] In a specific embodiment, at least one target configuration may be at least one form corresponding to the lesion to be segmented of the target object.

[0114] In an optional embodiment, the above-mentioned performing configuration recognition on at least one target magnetic resonance image based on at least one preset configuration to determine at least one target configuration corresponding to the target object includes:

[0115] Performing feature extraction processing on at least one target magnetic resonance image to obtain at least one target feature information;

[0116] Obtain at least one preset sample data;

[0117] Performing feature extraction processing on at least one preset sample data to obtain at least one preset feature information;

[0118] Performing clustering processing on at least one preset feature information and at least one target feature information to obtain cluster feature information sets corresponding to at least one target magnetic resonance image;

[0119] Based on the first association information between each preset feature information in each cluster feature information set and each target feature information in each cluster feature information set, configuration recognition is performed on at least one target magnetic resonance image to obtain at least one target configuration corresponding to the target object.

[0120] In a specific embodiment, the at least one preset feature information may be at least one feature vector corresponding to at least one target magnetic resonance image.

[0121] In a specific embodiment, at least one preset sample data can be used to perform configuration recognition on at least one target magnetic resonance image. The at least one preset sample data is obtained by fusing at least one target magnetic resonance image and at least one preset configuration. The at least one preset feature information can be at least one feature vector corresponding to the at least one preset sample data. Specifically, at least one standard three-dimensional model corresponding to the at least one preset configuration can be drawn, and the at least one standard three-dimensional model is fused with the at least one target magnetic resonance image through a data generation method (through a Cycle-GAN model, a Diffusion model, etc.) to obtain at least one preset sample data.

[0122] In a specific embodiment, the clustering feature information set corresponding to each of the at least one target magnetic resonance images may be a set of at least one feature vector. Specifically, when the feature information is in vector form, the clustering feature information set corresponding to each of the at least one target magnetic resonance images may be obtained by performing hierarchical clustering on at least one preset feature vector and at least one target feature vector to obtain a feature vector set corresponding to each of the at least one target magnetic resonance images.

[0123] In an optional embodiment, the clustering process is performed on at least one preset feature information and at least one target feature information to obtain a cluster feature information set corresponding to each of the at least one target magnetic resonance images, including:

[0124] Determine second association information between each target feature information in at least one target feature information and at least one preset feature information;

[0125] Based on the second association information, clustering processing is performed on at least one preset feature information and at least one target feature information to obtain cluster feature information sets corresponding to at least one target magnetic resonance image.

[0126] In a specific embodiment, the second association information can be the distance information between each target feature information in at least one target feature information and at least one preset feature information. Specifically, when the feature information is in vector form, it can be the distance between each target feature vector in at least one target feature vector and at least one preset feature vector. Optionally, the distance between feature vectors can be measured using Chebyshev distance.

[0127] In a specific embodiment, when the feature information is in vector form, the second association information can be the distance between each target feature vector and at least one preset feature vector in at least one target feature vector. Based on the second association information, hierarchical clustering can be performed on at least one preset feature vector and at least one target feature vector to obtain a feature vector set corresponding to at least one target magnetic resonance image.

[0128] In the above embodiment, based on the second correlation information between each target feature information and at least one preset feature information in at least one target feature information, at least one preset feature information and at least one target feature information are clustered to obtain a cluster feature information set corresponding to at least one target magnetic resonance image. By classifying similar target feature information and preset feature information into one category, the similarity between the lesion to be segmented of the target object and at least one preset sample data can be accurately understood, thereby providing a diagnostic basis for doctors and improving the reliability of diagnosis.

[0129] In a specific embodiment, the first association relationship may be distance information between each preset feature information in each cluster feature information set and each target feature information in each cluster feature information set. Specifically, when the feature information is in vector form, the preset feature information in each cluster feature information set may be first divided according to the configuration, and then the Chebyshev distance between each preset feature information in each cluster feature information set and each target feature information in each cluster feature information set under the same configuration is calculated, and the Chebyshev distances obtained under the same configuration are added to obtain the distance information d between at least one target magnetic resonance image and each configuration. 1 , d 2 , d 3 , d 4 , the similarity between at least one target magnetic resonance image and at least one preset configuration can be calculated based on the distance information Thereby, configuration recognition is performed on at least one target magnetic resonance image, and based on a preset threshold, a preset configuration with a similarity greater than the preset threshold is determined as at least one target configuration corresponding to the target object.

[0130] In the above embodiment, by clustering at least one preset feature information and at least one target feature information to obtain a cluster feature information set corresponding to at least one target magnetic resonance image, similar feature information can be classified into one category, which is helpful to find the connection between at least one target feature information and each preset feature information. By analyzing the association between the feature information, the configuration information of at least one target magnetic resonance image can be more accurately judged. Based on the first association information between each preset feature information in each cluster feature information set and each target feature information in each cluster feature information set, configuration recognition is performed on at least one target magnetic resonance image to obtain at least one target configuration corresponding to the target object. By integrating a large amount of feature information into different clusters, the complexity of the data is reduced, the efficiency of data processing is improved, and the analysis results are made more intuitive and easy to understand.

[0131] In step S207 , at least one target segmentation model is determined from at least one preset segmentation model based on at least one target magnetic resonance signal and at least one target configuration.

[0132] In a specific embodiment, at least one preset segmentation model may be a segmentation model corresponding to at least one preset configuration under at least one preset magnetic resonance signal. Correspondingly, at least one target segmentation model may be a segmentation model corresponding to at least one target configuration under at least one target magnetic resonance signal.

[0133] In an alternative embodiment, the above method may further include:

[0134] Obtain a training image set and a preset training model;

[0135] Based on the training image set, train the preset training model to obtain at least one preset segmentation model.

[0136] In a specific embodiment, the training image set may be a data set for training the preset training model. The training image set may include at least one to-be-trained magnetic resonance image of each preset object in at least one target magnetic resonance signal among at least one preset object and a labeled image corresponding to at least one to-be-trained magnetic resonance image of each preset object in at least one target magnetic resonance signal. Specifically, at least one preset object may be at least one second preset object. At least one to-be-trained magnetic resonance image of each preset object in at least one target magnetic resonance signal may be at least one magnetic resonance image to be segmented of each preset object in at least one target magnetic resonance signal. The labeled image corresponding to at least one to-be-trained magnetic resonance image of each preset object in at least one target magnetic resonance signal may be at least one magnetic resonance image including lesion annotation information corresponding to each preset object in at least one target magnetic resonance signal.

[0137] In a specific embodiment, gray value normalization processing may be performed on at least one to-be-trained magnetic resonance image of each preset object in at least one target magnetic resonance signal, and the image may be filtered respectively by using a density-related filter, a contour-related filter, an image entropy-related filter, etc., to obtain at least one feature expression image corresponding to at least one to-be-trained magnetic resonance image of each preset object in at least one target magnetic resonance signal.

[0138] In a specific embodiment, the preset training model may be a deep learning model to be trained, and the specific model structure may be set according to actual requirements. Optionally, the preset training model may include an SC Block module and a DecodeBlock module, and the branch formed by connecting the SC Block module and the Decode Block module may be used to enhance the feature extraction ability.

[0139] In a specific embodiment, such as Figure 3 As shown, Figure 3 is a schematic diagram of a preset training model according to an exemplary embodiment. Specifically, Figure 3 It can be seen that at least one feature expression image is used as the guide data and the training image set is input into the preset training model. Here, SC Block mainly includes the visual Transformer layer and the convolution layer, DecodeBlock mainly includes the deep supervision layer and the SoftMax function layer, including two branches connected by SC Block and Decode Block, wherein the branch connected by SC Block and Decode Block is used to enhance the feature extraction capability and perform feature fusion on different magnetic resonance images under different magnetic resonance signals so that the model can learn the lesion characteristics.

[0140] In a specific embodiment, the above-mentioned training of a preset training model based on a training image set to obtain at least one preset segmentation model includes: determining a current preset object from at least one preset object, using at least one current feature expression image corresponding to at least one current magnetic resonance image to be trained of the current preset object under at least one target magnetic resonance signal as guidance data, inputting at least one magnetic resonance image to be trained of the current preset object under at least one target magnetic resonance signal into the preset training model for lesion segmentation, obtaining at least one currently segmented magnetic resonance image of the current preset object under at least one target magnetic resonance signal, determining the lesion segmentation loss according to at least one currently segmented magnetic resonance image of the current preset object under at least one target magnetic resonance signal and a current annotated image corresponding to at least one current magnetic resonance image to be trained of the current preset object under at least one target magnetic resonance signal, and training the preset training model based on the lesion segmentation loss to obtain the preset segmentation model.

[0141] In a specific embodiment, the current preset object may be a training object of the current training cycle. Specifically, the current preset object may be randomly determined from at least one preset object, or some preset objects that do not participate in model training may be randomly selected from at least one preset object as the current preset object.

[0142] In a specific embodiment, the above-mentioned determination of the lesion segmentation loss based on the current annotated image corresponding to at least one current segmented magnetic resonance image of the current preset object under at least one target magnetic resonance signal and at least one current magnetic resonance image to be trained of the current preset object under at least one target magnetic resonance signal can be combined with a preset loss function. Specifically, the lesion segmentation loss can characterize the lesion segmentation performance of the current preset training model. Specifically, the preset loss function can be set in combination with actual applications.

[0143] In a specific embodiment, satisfying the preset convergence conditions can be set in combination with actual applications, for example, the number of executions of the loop iteration operation reaches a preset number, the change detection loss is less than a specified threshold, etc., which can be set specifically in combination with the training speed and model accuracy requirements.

[0144] In the above embodiment, based on at least one to-be-trained magnetic resonance image of each preset object in at least one preset object under at least one target magnetic resonance signal and an annotated image corresponding to at least one to-be-trained magnetic resonance image of each preset object under at least one target magnetic resonance signal, a preset training model is trained to obtain at least one preset segmentation model. By inputting at least one to-be-trained magnetic resonance image of at least one preset object under at least one target magnetic resonance signal during the training process, the model can learn diversified data, reduce overfitting of the model to specific data, and thereby improve the accuracy of lesion segmentation.

[0145] In an optional embodiment, the above method may further include:

[0146] Acquire at least one to-be-trained magnetic resonance image and a labeled image corresponding to the target training magnetic resonance image of each preset object under at least one target magnetic resonance signal;

[0147] After spatially aligning at least one to-be-trained magnetic resonance image of each preset object under at least one target magnetic resonance signal, the annotated image corresponding to the target training magnetic resonance image is used as the annotated image corresponding to at least one to-be-trained magnetic resonance image of each preset object under at least one preset magnetic resonance signal.

[0148] In a specific embodiment, the target training magnetic resonance image can be any one of the at least one magnetic resonance image to be trained, and accordingly, the target training magnetic resonance image can correspond to any one of the at least one target magnetic resonance signal, which can be selected based on actual needs.

[0149] In a specific embodiment, at least one to-be-trained magnetic resonance image of each preset object under at least one target magnetic resonance signal is spatially aligned. Specifically, based on the spatial posture relationship in the DICOM (Digital Imaging and Communications in Medicine) header information, the magnetic resonance images other than the target training magnetic resonance in at least one to-be-trained magnetic resonance image under at least one target magnetic resonance signal can be roughly aligned to the target training magnetic resonance image, and then the skin contour is used as a feature point, and ICP registration (Iterative Closest Point) is used to perform refined alignment to achieve consistency in the image posture of at least one to-be-trained magnetic resonance image and the target training magnetic resonance image.

[0150] In the above embodiment, at least one to-be-trained magnetic resonance image of each preset object under at least one target magnetic resonance signal is spatially aligned, so that the same anatomical structure and lesion in at least one to-be-trained magnetic resonance image of each preset object under at least one target magnetic resonance signal correspond spatially, which helps the model to learn the characteristics of the lesion more accurately, and uses the annotated image corresponding to the target training magnetic resonance image as the annotated image corresponding to at least one to-be-trained magnetic resonance image of each preset object under at least one preset magnetic resonance signal, which greatly simplifies the annotation work and improves the annotation efficiency.

[0151] In step S209, at least one target magnetic resonance image is input into at least one target segmentation model for segmentation to obtain at least one lesion segmentation image.

[0152] In a specific embodiment, at least one target lesion segmentation image may be at least one segmentation image output by at least one target segmentation model. Specifically, at least one target lesion segmentation image corresponds to at least one image obtained by inputting at least one target magnetic resonance image into a segmentation model corresponding to at least one target configuration under at least one target magnetic resonance signal.

[0153] In step S2011, weighted processing is performed on at least one lesion segmentation image to obtain a target lesion segmentation image corresponding to the target object.

[0154] In a specific embodiment, the target lesion segmentation image may be a lesion segmentation image corresponding to the target object.

[0155] In an optional embodiment, the above method may further include:

[0156] Acquire a target annotated image corresponding to at least one target magnetic resonance image;

[0157] Determining, based on the at least one target magnetic resonance image and the target annotated image, a degree of overlap between the at least one target magnetic resonance image and the target annotated image;

[0158] In an optional embodiment, the weighted processing of at least one lesion segmentation image to obtain a target lesion segmentation image corresponding to the target object includes:

[0159] Performing weighted merging processing on the lesion segmentation images under the same magnetic resonance signal in at least one lesion segmentation image to obtain at least one lesion segmentation image under at least one target magnetic resonance signal;

[0160] Normalizing the overlap between at least one target magnetic resonance image and the target annotated image to obtain weights corresponding to at least one lesion segmentation image under at least one target magnetic resonance signal;

[0161] Based on the weights corresponding to the at least one lesion segmentation image under the at least one magnetic resonance signal, weighted processing is performed on the at least one lesion segmentation image under the at least one target magnetic resonance signal to obtain a target lesion segmentation image corresponding to the target object.

[0162] In a specific embodiment, the target annotated image may be an annotated image corresponding to at least one target magnetic resonance image, and may be an image obtained by annotating lesions on any one of the at least one target magnetic resonance images. Specifically, the annotating method of the target annotated image may refer to the above content.

[0163] In a specific embodiment, the degree of overlap between at least one target magnetic resonance image and the target annotated image may be the degree of overlap between at least one feature image and a high-value portion of the target annotated image.

[0164] In an optional embodiment, the determining of the degree of overlap between the at least one target magnetic resonance image and the target annotated image based on the at least one target magnetic resonance image and the target annotated image comprises:

[0165] performing normalization processing on at least one target magnetic resonance image to obtain at least one normalized target magnetic resonance image;

[0166] Performing filtering processing on the normalized at least one target magnetic resonance image to obtain at least one characteristic image;

[0167] Based on the overlap between at least one feature image and the target annotation image, an overlap between at least one target magnetic resonance image and the target annotation image is determined.

[0168] In a specific embodiment, at least one normalized target magnetic resonance image may be at least one magnetic resonance image corresponding to the normalization processing of at least one target magnetic resonance image. Specifically, the normalization processing may be to normalize the gray value of the magnetic resonance image so that its value range is between 0 and 1.

[0169] In a specific embodiment, at least one feature image may be an image obtained by filtering at least one normalized target magnetic resonance image. Specifically, the filtering method may refer to the above content.

[0170] In a specific embodiment, the overlap degree between at least one feature image and the target annotation image may be used as the overlap degree between at least one target magnetic resonance image and the target annotation image.

[0171] In the above embodiment, normalizing at least one target magnetic resonance image can unify the gray values of different images to the same range, eliminate the image feature differences, improve the stability and efficiency of model training, and thus more accurately learn the feature information of the lesion. Filtering the normalized target magnetic resonance image can enhance the features of the image and reduce the influence of noise on the overlap degree calculation and subsequent analysis. After normalization and filtering, at least one feature image highlights the key features of the lesion and improves the accuracy and reliability of the target lesion segmentation image.

[0172] In a specific embodiment, at least one lesion segmentation image may correspond to at least one magnetic resonance signal. The at least one lesion segmentation image is divided according to the type of magnetic resonance signal to obtain at least one lesion segmentation image corresponding to each of different types of magnetic resonance signals. The lesion segmentation images under the same magnetic resonance signal are weighted and combined to obtain at least one lesion segmentation image corresponding to each magnetic resonance signal.

[0173] In a specific embodiment, the weights corresponding to at least one lesion segmentation image may be used to weight at least one lesion segmentation image. Specifically, the weights corresponding to at least one lesion segmentation image may be obtained by normalizing the overlap degree between at least one target magnetic resonance image and the target annotation image.

[0174] In the above embodiment, the lesion segmentation images under the same magnetic resonance signal are first weighted and merged, which can comprehensively utilize the information of at least one magnetic resonance image under the same magnetic resonance signal, reduce the error caused by a single segmentation image, and improve the stability and reliability of the segmentation result. The overlap is normalized to obtain the weight corresponding to each lesion segmentation image. A weight can be assigned to each segmented image based on the similarity between at least one target magnetic resonance image and a target annotated image. A higher weight will be assigned to an image with a higher overlap. This weight allocation method based on the actual segmentation effect is more scientific and reasonable, and greatly improves the accuracy of target lesion segmentation.

[0175] It can be seen from the technical solutions provided in the above embodiments of this specification that this specification determines at least one preset configuration corresponding to the target object and at least one target magnetic resonance signal determined based on the overlap of at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal and the corresponding labeled image of at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal; obtains at least one target magnetic resonance image of the target object under at least one target magnetic resonance signal; based on at least one preset configuration, performs configuration recognition on at least one target magnetic resonance image to determine at least one target configuration corresponding to the target object; based on at least one target magnetic resonance signal and at least one target configuration, determines at least one target segmentation model corresponding to at least one preset configuration under at least one preset magnetic resonance signal from at least one preset segmentation model; can improve the accuracy and pertinence of segmentation based on achieving a high degree of matching between the target object and at least one target segmentation model; inputs at least one target magnetic resonance image into at least one target segmentation model for segmentation to obtain at least one lesion segmentation image; performs weighted processing on at least one lesion segmentation image to obtain a target lesion segmentation image corresponding to the target object, which can further optimize the segmentation effect, greatly improve the efficiency and accuracy of lesion segmentation, and thus improve the reliability of diagnosis.

[0176] Figure 4 FIG. 1 is a block diagram of a lesion segmentation device according to an exemplary embodiment. Figure 4 , the device comprises:

[0177] A first determination module 410 is configured to determine at least one target magnetic resonance signal and at least one preset configuration corresponding to the target object, wherein the at least one target magnetic resonance signal is determined based on an overlap between at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal and an annotation image corresponding to at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal;

[0178] A target magnetic resonance image acquisition module 430 is used to acquire at least one target magnetic resonance image of a target object under at least one target magnetic resonance signal;

[0179] A second determination module 450 is used to perform configuration recognition on at least one target magnetic resonance image based on at least one preset configuration to determine at least one target configuration corresponding to the target object;

[0180] A third determination module 470 is configured to determine at least one target segmentation model from at least one preset segmentation model based on at least one target magnetic resonance signal and at least one target configuration; the at least one preset segmentation model is a segmentation model corresponding to at least one preset configuration under at least one preset magnetic resonance signal;

[0181] The lesion segmentation image acquisition module 490 is used to input at least one target magnetic resonance image into at least one target segmentation model for segmentation to obtain at least one lesion segmentation image;

[0182] The target lesion segmentation image acquisition module 4110 is used to perform weighted processing on at least one lesion segmentation image to obtain a target lesion segmentation image corresponding to the target object.

[0183] In an optional embodiment, the second determining module 450 includes:

[0184] A target feature information acquisition unit, used for performing feature extraction processing on at least one target magnetic resonance image to obtain at least one target feature information;

[0185] A preset sample data acquisition unit, used to acquire at least one preset sample data; the at least one preset sample data is obtained by fusing at least one target magnetic resonance image and at least one preset configuration;

[0186] A preset feature information acquisition unit, used to perform feature extraction processing on at least one preset sample data to obtain at least one preset feature information;

[0187] A cluster feature information set acquisition unit, used to perform clustering processing on at least one preset feature information and at least one target feature information to obtain a cluster feature information set corresponding to at least one target magnetic resonance image;

[0188] The target configuration acquisition unit is used to perform configuration recognition on at least one target magnetic resonance image based on the first association information between each preset feature information in each cluster feature information set and each target feature information in each cluster feature information set to obtain at least one target configuration corresponding to the target object.

[0189] In an optional embodiment, the clustering feature information set acquisition unit includes:

[0190] A second association information determination subunit, used to determine second association information between each target feature information in at least one target feature information and at least one preset feature information;

[0191] The cluster feature information set acquisition subunit is used to perform clustering processing on at least one preset feature information and at least one target feature information based on the second association information to obtain cluster feature information sets corresponding to at least one target magnetic resonance image.

[0192] In an optional embodiment, the above device further includes:

[0193] A target annotated image acquisition module, used to acquire a target annotated image corresponding to at least one target magnetic resonance image;

[0194] an overlap determination module, configured to determine an overlap between at least one target magnetic resonance image and a target annotated image based on at least one target magnetic resonance image and a target annotated image;

[0195] The target lesion segmentation image acquisition module 4110 includes:

[0196] A lesion segmentation image unit, used for performing weighted merging processing on the lesion segmentation images under the same magnetic resonance signal in at least one lesion segmentation image to obtain at least one lesion segmentation image under at least one target magnetic resonance signal;

[0197] A weight acquisition unit, used for normalizing the overlap between at least one target magnetic resonance image and a target annotated image to obtain a weight corresponding to each of at least one lesion segmentation image under at least one target magnetic resonance signal;

[0198] The target lesion segmentation image acquisition unit is used to perform weighted processing on at least one lesion segmentation image under at least one target magnetic resonance signal based on the weights corresponding to the at least one lesion segmentation image under at least one magnetic resonance signal, so as to obtain a target lesion segmentation image corresponding to the target object.

[0199] In an optional embodiment, the overlap determination module includes:

[0200] a normalized target magnetic resonance image determining unit, configured to perform normalization processing on at least one target magnetic resonance image to obtain at least one normalized target magnetic resonance image;

[0201] At least one characteristic image acquisition unit, used for filtering the at least one normalized target magnetic resonance image to obtain at least one characteristic image;

[0202] The overlap determination unit is used to determine the overlap between at least one target magnetic resonance image and the target annotated image based on the overlap between at least one feature image and the target annotated image.

[0203] In an optional embodiment, the above device further includes:

[0204] a fourth determination module, configured to determine at least one magnetic resonance signal to be selected corresponding to each preset object based on an overlap between at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal and a corresponding labeled image of at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal;

[0205] The fifth determination module is used to determine at least one target magnetic resonance signal corresponding to the target object based on the at least one to-be-selected magnetic resonance signal corresponding to each preset object.

[0206] In an optional embodiment, the above device further includes:

[0207] A first acquisition module is used to acquire a training image set and a preset training model; the training image set includes at least one to-be-trained magnetic resonance image of each preset object under at least one target magnetic resonance signal and a labeled image corresponding to at least one to-be-trained magnetic resonance image of each preset object under at least one target magnetic resonance signal;

[0208] The preset segmentation model acquisition module is used to train the preset training model based on the training image set to obtain at least one preset segmentation model.

[0209] In an optional embodiment, the above device further includes:

[0210] A second acquisition module is used to acquire at least one to-be-trained magnetic resonance image and a labeled image corresponding to the target training magnetic resonance image of each preset object under at least one target magnetic resonance signal, wherein the target training magnetic resonance image is any one of the at least one to-be-trained magnetic resonance images;

[0211] The annotated image acquisition module is used to use the annotated image corresponding to the target training magnetic resonance image as the annotated image corresponding to the at least one to-be-trained magnetic resonance image of each preset object under at least one preset magnetic resonance signal after spatially aligning the at least one to-be-trained magnetic resonance image of each preset object under at least one target magnetic resonance signal.

[0212] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0213] Figure 5 is a block diagram of an electronic device for lesion segmentation according to an exemplary embodiment. The electronic device may be a server, and its internal structure diagram may be as shown in FIG. Figure 5 As shown. The electronic device includes a processor, a memory and a network interface connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a lesion segmentation method is implemented.

[0214] Those skilled in the art will understand that Figure 5 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present disclosure, and does not constitute a limitation on the electronic device to which the scheme of the present disclosure is applied. The specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0215] In an exemplary embodiment, an electronic device is also provided, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement the lesion segmentation method as in the embodiment of the present disclosure.

[0216] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0217] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A lesion segmentation method, characterized in that: include: Determining at least one target magnetic resonance signal and at least one preset configuration corresponding to the target object, wherein the at least one target magnetic resonance signal is determined based on an overlap between at least one preset magnetic resonance image of each preset object in at least one preset magnetic resonance signal and a corresponding labeled image of at least one preset magnetic resonance image of each preset object in at least one preset magnetic resonance signal; Acquiring at least one target magnetic resonance image of the target object under the at least one target magnetic resonance signal; Based on the at least one preset configuration, performing configuration recognition on the at least one target magnetic resonance image to determine at least one target configuration corresponding to the target object; Based on the at least one target magnetic resonance signal and the at least one target configuration, determining at least one target segmentation model from at least one preset segmentation model; the at least one preset segmentation model is a segmentation model corresponding to the at least one preset configuration under the at least one preset magnetic resonance signal; Inputting the at least one target magnetic resonance image into the at least one target segmentation model for segmentation to obtain at least one lesion segmentation image; Weighted processing is performed on the at least one lesion segmentation image to obtain a target lesion segmentation image corresponding to the target object.

2. A lesion segmentation method according to claim 1, characterized in that: The performing configuration recognition on the at least one target magnetic resonance image based on the at least one preset configuration to determine at least one target configuration corresponding to the target object comprises: Performing feature extraction processing on the at least one target magnetic resonance image to obtain at least one target feature information; Acquire at least one preset sample data; the at least one preset sample data is obtained by fusing the at least one target magnetic resonance image and the at least one preset configuration; Performing feature extraction processing on the at least one preset sample data to obtain at least one preset feature information; Performing clustering processing on the at least one preset feature information and the at least one target feature information to obtain cluster feature information sets corresponding to each of the at least one target magnetic resonance images; Based on the first association information between each preset feature information in each cluster feature information set and each target feature information in each cluster feature information set, configuration recognition is performed on the at least one target magnetic resonance image to obtain at least one target configuration corresponding to the target object.

3. A lesion segmentation method according to claim 2, characterized in that: The clustering process is performed on the at least one preset feature information and the at least one target feature information to obtain a cluster feature information set corresponding to each of the at least one target magnetic resonance images, comprising: Determine second association information between each target feature information in the at least one target feature information and the at least one preset feature information; Based on the second association information, clustering processing is performed on the at least one preset feature information and the at least one target feature information to obtain cluster feature information sets corresponding to each of the at least one target magnetic resonance images.

4. A lesion segmentation method according to claim 1, characterized in that: The method further comprises: Acquiring a target annotated image corresponding to the at least one target magnetic resonance image; determining, based on the at least one target magnetic resonance image and the target annotated image, a degree of overlap between the at least one target magnetic resonance image and the target annotated image; The performing weighted processing on the at least one lesion segmentation image to obtain a target lesion segmentation image corresponding to the target object comprises: Performing weighted merging processing on the lesion segmentation images under the same magnetic resonance signal in the at least one lesion segmentation image to obtain at least one lesion segmentation image under the at least one target magnetic resonance signal; Normalizing the overlap between the at least one target magnetic resonance image and the target annotated image to obtain weights corresponding to the at least one lesion segmentation image under the at least one target magnetic resonance signal; Based on the weights corresponding to the at least one lesion segmentation image under the at least one magnetic resonance signal, weighted processing is performed on the at least one lesion segmentation image under the at least one target magnetic resonance signal to obtain a target lesion segmentation image corresponding to the target object.

5. A lesion segmentation method according to claim 4, characterized in that: The determining, based on the at least one target magnetic resonance image and the target annotated image, a degree of overlap between the at least one target magnetic resonance image and the target annotated image comprises: performing normalization processing on the at least one target magnetic resonance image to obtain at least one normalized target magnetic resonance image; Performing filtering processing on the at least one normalized target magnetic resonance image to obtain at least one characteristic image; Based on the overlap between the at least one feature image and the target annotated image, the overlap between the at least one target magnetic resonance image and the target annotated image is determined.

6. A lesion segmentation method according to claim 1, characterized in that: The method further comprises: Determining at least one magnetic resonance signal to be selected corresponding to each preset object based on an overlap between at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal and a corresponding labeled image of at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal; At least one target magnetic resonance signal corresponding to the target object is determined based on the at least one to-be-selected magnetic resonance signal corresponding to each preset object.

7. A lesion segmentation method according to any one of claims 1 to 6, characterized in that: The method further comprises: Acquire a training image set and a preset training model; the training image set includes at least one to-be-trained magnetic resonance image of each preset object in the at least one preset object under the at least one target magnetic resonance signal and an annotated image corresponding to the at least one to-be-trained magnetic resonance image of each preset object under the at least one target magnetic resonance signal; Based on the training image set, the preset training model is trained to obtain the at least one preset segmentation model.

8. A lesion segmentation method according to claim 7, characterized in that: The method further comprises: Acquire at least one to-be-trained magnetic resonance image and a labeled image corresponding to the target training magnetic resonance image of each preset object under the at least one target magnetic resonance signal, wherein the target training magnetic resonance image is any one of the at least one to-be-trained magnetic resonance images; After spatially aligning at least one to-be-trained magnetic resonance image of each preset object under the at least one target magnetic resonance signal, the annotated image corresponding to the target training magnetic resonance image is used as the annotated image corresponding to the at least one to-be-trained magnetic resonance image of each preset object under the at least one preset magnetic resonance signal.

9. A lesion segmentation device, characterized in that: include: A first determination module is configured to determine at least one target magnetic resonance signal and at least one preset configuration corresponding to the target object, wherein the at least one target magnetic resonance signal is determined based on an overlap between at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal and a corresponding labeled image of at least one preset magnetic resonance image of each preset object under at least one preset magnetic resonance signal; a target magnetic resonance image acquisition module, configured to acquire at least one target magnetic resonance image of the target object under the at least one target magnetic resonance signal; A second determination module is configured to perform configuration recognition on the at least one target magnetic resonance image based on the at least one preset configuration, and determine at least one target configuration corresponding to the target object; a third determination module, configured to determine at least one target segmentation model from at least one preset segmentation model based on the at least one target magnetic resonance signal and the at least one target configuration; the at least one preset segmentation model is a segmentation model corresponding to the at least one preset configuration under the at least one preset magnetic resonance signal; A lesion segmentation image acquisition module, used for inputting the at least one target magnetic resonance image into the at least one target segmentation model for segmentation to obtain at least one lesion segmentation image; The target lesion segmentation image acquisition module is used to perform weighted processing on the at least one lesion segmentation image to obtain a target lesion segmentation image corresponding to the target object.

10. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the instructions to implement the lesion segmentation method as described in any one of claims 1 to 8.

Images