Tissue segmentation method and device, computer equipment and storage medium

Abstract

The invention relates to a tissue segmentation method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring a plurality of medical images of a to-be-detected object, and acquiring at least one to-be-segmented tissue of the to-be-detected object; wherein the modalities of at least part of the medical images in the plurality of medical images are different; for each to-be-segmented tissue, determining a mode corresponding to the to-be-segmented tissue according to a preset first corresponding relationship; the first corresponding relation comprises a corresponding relation between different tissues and different modes, and the segmentation effect of the tissues in the medical image of the corresponding mode meets a segmentation condition; determining a target medical image corresponding to the to-be-segmented tissue from the plurality of medical images according to the modality corresponding to the to-be-segmented tissue; and performing segmentation processing on the to-be-segmented tissue in the target medical image to obtain a tissue segmentation result of the target medical image. The to-be-segmented tissue and the modal image are automatically matched, the modal image with a good segmentation effect is determined for the to-be-segmented tissue, and the tissue segmentation effect is improved.

Description

Technical Field

[0001] The present application relates to the field of medical equipment technology, and in particular to a tissue segmentation method, apparatus, computer equipment and storage medium. Background Art

[0002] With the development of intelligent medical equipment, the auxiliary role of medical equipment in surgical operations has become increasingly prominent. For example, in neurosurgery, whether medical equipment can accurately and efficiently determine the location of various tissues in the patient's head based on the patient's medical images plays a vital role in the efficiency and safety of surgical operations.

[0003] Since different tissues have different display effects on different modality images, it is usually necessary to extract different tissues from multiple modality images and fuse the extracted different tissues to assist the normal progress of the operation.

[0004] However, when users use traditional medical equipment to segment and fuse different tissues in multiple modality images, there is a problem of poor tissue segmentation effect. Summary of the invention

[0005] Based on this, it is necessary to provide a tissue segmentation method, apparatus, computer equipment, computer-readable storage medium and computer program product that can improve the segmentation effect of different tissues in order to solve the above technical problems.

[0006] In a first aspect, the present application provides a tissue segmentation method, comprising:

[0007] Acquire multiple medical images of the object to be tested, and acquire at least one tissue to be segmented of the object to be tested; wherein at least some of the multiple medical images have different modalities;

[0008] For each tissue to be segmented, the modality corresponding to the tissue to be segmented is determined according to a preset first correspondence relationship; the first correspondence relationship includes a correspondence relationship between different tissues and different modalities, and the segmentation effect of the tissue in the medical image of the corresponding modality meets the segmentation condition;

[0009] Determining a target medical image corresponding to the tissue to be segmented from a plurality of medical images according to a modality corresponding to the tissue to be segmented;

[0010] The tissue to be segmented in the target medical image is segmented to obtain a tissue segmentation result of the target medical image.

[0011] In one embodiment, obtaining a plurality of medical images of the object to be tested includes:

[0012] Obtaining a target formula of the object to be tested selected by a user from at least one candidate formula;

[0013] According to a preset second correspondence relationship, multiple candidate modalities corresponding to the target procedure are determined; the second correspondence relationship includes a correspondence relationship between different procedures and different modalities;

[0014] Based on multiple candidate modalities, medical images corresponding to each candidate modality of the object to be tested are obtained.

[0015] In one embodiment, based on multiple candidate modalities, obtaining medical images corresponding to each candidate modality of the object to be tested includes:

[0016] Based on multiple candidate modalities, generate and display image input components corresponding to each candidate modality;

[0017] For each image input component, in response to the user's image input operation on the image input component, a medical image of a candidate modality corresponding to the image input component of the object to be tested is obtained.

[0018] In one embodiment, in response to a user's image input operation on an image input component, obtaining a medical image of a candidate modality corresponding to the image input component of the object to be tested includes:

[0019] In response to an image input operation of the user on the image input component, acquiring a candidate medical image input by the user;

[0020] When the modality of the candidate medical image is consistent with the candidate modality corresponding to the image input component, the candidate medical image is imported as the medical image of the candidate modality corresponding to the image input component of the object to be tested.

[0021] In one of the embodiments, the image input component corresponding to the candidate modality includes at least one image input subcomponent of the candidate modality, and the image input subcomponent is used to import the medical image of the candidate modality.

[0022] In one of the embodiments, the image input component corresponding to the candidate modality also includes an adding subcomponent, and the adding subcomponent is used to increase the image input subcomponent of the candidate modality.

[0023] In one embodiment, obtaining at least one tissue to be segmented of the object to be detected includes:

[0024] According to a preset third correspondence, at least one to-be-segmented tissue corresponding to the target procedure is determined; the third correspondence includes a correspondence between different procedures and different tissues.

[0025] In one embodiment, a method for obtaining a first corresponding relationship includes:

[0026] Performing tissue recognition on multiple medical images respectively to obtain tissue recognition results of each medical image; the tissue recognition results include tissues recognized from the corresponding medical images;

[0027] Determine the tissue corresponding to each medical image based on the identification results of each tissue and the segmentation conditions;

[0028] A correspondence relationship between the modality of the medical image and the corresponding tissue is established to obtain a first correspondence relationship.

[0029] In one embodiment, before segmenting the tissue to be segmented in the target medical image and obtaining the tissue segmentation result of the target medical image, the method further includes:

[0030] displaying at least one group of segmentation options; each group of segmentation options includes a tissue to be segmented and a corresponding target medical image;

[0031] Obtaining a target segmentation option selected by a user from at least one set of segmentation options;

[0032] Accordingly, segmenting the tissue to be segmented in the target medical image to obtain a tissue segmentation result of the target medical image includes:

[0033] The tissue to be segmented in the target medical image corresponding to the target segmentation option is segmented to obtain a tissue segmentation result of the target medical image.

[0034] In one embodiment, if the tissue to be segmented displayed by the target segmentation option corresponds to multiple target medical images, segmenting the tissue to be segmented in the target medical image corresponding to the target segmentation option to obtain the tissue segmentation result of the target medical image includes:

[0035] Acquire at least one selected target medical image selected by a user from a plurality of target medical images;

[0036] The tissue to be segmented in each selected target medical image corresponding to the target segmentation option is segmented to obtain a tissue segmentation result of each selected target medical image.

[0037] In one embodiment, segmenting the tissue to be segmented in the target medical image corresponding to the target segmentation option to obtain the tissue segmentation result of the target medical image further includes:

[0038] If no user selection operation on multiple target medical images corresponding to the target segmentation option is detected, a default target medical image is determined from the multiple target medical images, and the tissue to be segmented in the default target medical image is segmented to obtain a tissue segmentation result of the default target medical image.

[0039] In one embodiment, the plurality of target medical images are sorted in descending order according to the segmentation effect, and a default target medical image is determined from the plurality of target medical images, including:

[0040] The first target medical image in the descending order is used as a default target medical image.

[0041] In one embodiment, the first target medical image is displayed in the target segmentation option, and other target medical images except the first target medical image among the multiple target medical images are hidden; the target segmentation option also includes a drop-down component, and the method further includes:

[0042] In response to a user triggering operation on the pull-down component, other hidden target medical images are displayed.

[0043] In one embodiment, the method further comprises:

[0044] A segmentation response corresponding to the target segmentation option is displayed at a preset position on the target segmentation option; the segmentation response includes segmentation success or segmentation failure.

[0045] In one embodiment, the method further comprises:

[0046] The tissue segmentation results of each target medical image are fused and displayed.

[0047] In one embodiment, before segmenting the tissue to be segmented in the target medical image and obtaining the tissue segmentation result of the target medical image, the method further includes:

[0048] Determine the target initialization operation corresponding to the target technique according to a preset fourth corresponding relationship; the fourth corresponding relationship includes a corresponding relationship between different techniques and different initialization operations;

[0049] Perform target initialization operations.

[0050] In a second aspect, the present application further provides a tissue segmentation device, comprising:

[0051] A first acquisition module is used to acquire multiple medical images of the object to be tested and to acquire at least one tissue to be segmented of the object to be tested; wherein at least some of the multiple medical images have different modalities;

[0052] A first determination module is used to determine, for each tissue to be segmented, a modality corresponding to the tissue to be segmented according to a preset first correspondence relationship; the first correspondence relationship includes a correspondence relationship between different tissues and different modalities, and the segmentation effect of the tissue in the medical image of the corresponding modality meets the segmentation condition;

[0053] A second determination module is used to determine a target medical image corresponding to the tissue to be segmented from a plurality of medical images according to a modality corresponding to the tissue to be segmented;

[0054] The segmentation module is used to perform segmentation processing on the tissue to be segmented in the target medical image to obtain the tissue segmentation result of the target medical image.

[0055] In a third aspect, the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the tissue segmentation method in the first aspect when executing the computer program.

[0056] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the tissue segmentation method in the first aspect.

[0057] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the tissue segmentation method in the first aspect.

[0058] The above-mentioned tissue segmentation method, apparatus, computer equipment, storage medium and computer program product, the computer equipment obtains multiple medical images of the object to be tested, and obtains at least one tissue to be segmented of the object to be tested; wherein, the modalities of at least some of the multiple medical images are different; then, for each tissue to be segmented, according to a preset first correspondence, the modality corresponding to the tissue to be segmented is determined; and according to the modality corresponding to the tissue to be segmented, a target medical image corresponding to the tissue to be segmented is determined from the multiple medical images; and then the tissue to be segmented in the target medical image is segmented to obtain a tissue segmentation result of the target medical image. Wherein, the first correspondence includes the correspondence between different tissues and different modalities, and the segmentation effect of the tissue in the medical image of the corresponding modality meets the segmentation condition. That is to say, in the embodiment of the present application, for the situation where multiple modal images are used to segment different tissues, the computer device can automatically match the tissue to be segmented with the modal image to determine the target medical image corresponding to the modality with the best segmentation effect for the tissue to be segmented; using this method, there is no need for the user to match the tissue to be segmented with the corresponding modal image based on subjective judgment or experience judgment, which can not only reduce the user's human participation in the tissue segmentation process, but also improve the matching degree between the tissue and the modal image, and avoid the problem of poor tissue segmentation effect when the user selects a modal image with poor segmentation effect for tissue segmentation due to lack of experience; that is, using the method of the present application, not only can the segmentation effect of different tissues in multi-modal images be improved, but also the efficiency of tissue segmentation can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 A diagram showing an application environment of a tissue segmentation method in an embodiment;

[0061] Figure 2 A schematic diagram of a flow chart of a tissue segmentation method in one embodiment;

[0062] Figure 3 is a flow chart of a tissue segmentation method in another embodiment;

[0063] Figure 4 is a flow chart of a tissue segmentation method in another embodiment;

[0064] Figure 5 is a flow chart of a tissue segmentation method in another embodiment;

[0065] Figure 6 is a flow chart of a tissue segmentation method in another embodiment;

[0066] Figure 7 A schematic diagram of the workflow of a tissue segmentation software system in one embodiment;

[0067] FIG8( a ) is a schematic diagram of a display interface of a split option in one embodiment;

[0068] FIG8( b ) is a schematic diagram of a display interface of segmentation progress in one embodiment;

[0069] FIG8( c ) is a schematic diagram of a display interface after segmentation is completed in one embodiment;

[0070] Fig. 9 is a structural block diagram of a tissue segmentation device in one embodiment;

[0071] Fig.10 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0072] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0073] With the development of intelligent medical equipment, the auxiliary role of medical equipment in surgical operations has become increasingly prominent. For example, in neurosurgery, whether medical equipment can accurately and efficiently determine the location of various tissues in the patient's head based on the patient's medical images plays a vital role in the efficiency and safety of surgical operations.

[0074] There are many common neurosurgery procedures, such as stereotactic electroencephalography (SEEG) electrode implantation, deep brain stimulation (DBS) (also known as brain pacemaker), puncture, etc. The auxiliary functions required by different procedures are also quite different.

[0075] Traditional neurosurgery navigation software can usually provide a full set of functions for neurosurgery to meet the functional requirements of different surgical procedures. When assisting a certain surgical procedure, users need to filter the corresponding functional subset according to the surgical procedure, which greatly affects the efficiency and convenience of users in using the software system.

[0076] In addition, traditional neurosurgery navigation software usually does not perform strong verification and matching on the surgical procedure selected by the user and the medical images of the patient input by the user in the initial stage, nor does it remind the user when the modality of the medical image does not match the surgical procedure; the user needs to match different surgical procedures with modality data based on experience. Once the imported medical image is wrong or missing, it will often lead to the inability to execute subsequent functions, which in turn causes the operation to fail.

[0077] Furthermore, when using traditional neurosurgery navigation software, users are required to judge the interdependence between various functions, or in other words, the order in which the various functions are executed. For example, when performing image segmentation processing, nuclear group segmentation usually depends on the anterior commissure (AC) and posterior commissure (PC), referred to as ACPC points. If the ACPC points are not determined first, the software system will feedback that the extraction failed when extracting the nuclear group.

[0078] In addition, traditional neurosurgery navigation software extracts different tissues from multimodal images for fusion display to assist the normal progress of surgery. For multimodal data, when performing image segmentation, users often need to consider which modality of tissue segmentation will achieve the best segmentation effect. Different modality data are often more suitable for different tissue segmentation, such as nuclear tissue segmentation effects are better on T1 and T2 data. Therefore, it is particularly important to provide users with a segmentation solution that can efficiently extract the best tissue under multimodal images.

[0079] It can be seen that traditional neurosurgery navigation software has the following problems when processing preoperative images:

[0080] 1. Missing of procedure and procedure modality: It is impossible to determine the procedure and the modality data required for the procedure when entering the navigation software. This missing part may cause the procedure and modality data to not match, and also limit the related operations of automatic initialization of the system.

[0081] 2. Dependency before and after tissue extraction: Users need to consider the order of some functions. For example, if the user does not perform ACPC correction and cortical landmark extraction, it is impossible to extract nuclei and brain map related data.

[0082] 3. Prior knowledge is required: Data of different modalities support the segmentation of different tissues. Segmentation on modality data that is not suitable for the segmentation of the tissue may result in poor extraction results or even extraction failure.

[0083] 4. Poor feedback effect: In order to avoid interfering with user usage, the segmentation results are often prompted to users in the form of bubbles. When the extraction fails, the image area is not displayed. If the user does not pay close attention to the bubble prompt, after the bubble disappears, it is easy for the user to feel that the function is invalid.

[0084] 5. Non-standardized automated process: Standardized process is an important means to ensure the image extraction effect, and automation can help users save a lot of time in finding function entrances (as the complexity of the system increases, the number of different function entrances also increases). The existing software system only mechanically executes the function operation selected by the user after the user selects the function, and cannot guide and automatically help the user complete the necessary function initialization operation.

[0085] Based on the various drawbacks of the above-mentioned traditional neurosurgery navigation software, the embodiment of the present application proposes a system interaction solution for automatically guiding the extraction of the best tissue in multimodal data based on different surgical procedures in the context of neurosurgery.

[0086] The tissue segmentation method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, the computer device 102 may include but is not limited to various personal computers, laptops, smart phones, tablet computers, and portable wearable devices, etc.; wherein the portable wearable device may be a smart watch, a smart bracelet, a head-mounted device, etc.

[0087] Exemplarily, neurosurgery navigation software may be installed in the computer device 102 , and the neurosurgery navigation software can provide users with various functions required for different surgical procedures and provide the required data support for neurosurgery to assist in the smooth execution of neurosurgery.

[0088] In an exemplary embodiment, Figure 2 As shown, a tissue segmentation method is provided, which is applied to Figure 1 The computer device in is taken as an example to illustrate, including the following steps 202 to 208.

[0089] in:

[0090] Step 202 : acquiring a plurality of medical images of the object to be detected, and acquiring at least one tissue to be segmented of the object to be detected.

[0091] Among them, the modalities of at least some of the multiple medical images are different, including that the modalities of at least some of the medical images are different from each other; in other words, the modalities of the multiple medical images can be different from each other, or they can be partially the same; for example: among five medical images, there are three medical images whose modalities are different from each other, and the modalities of the other two medical images can be the same or different, but it should be noted that the modalities of the other two medical images should be included in the three modalities corresponding to the three medical images. The modalities of medical images can include but are not limited to computed tomography (CT), magnetic resonance (MR), positron emission tomography (PET), X-ray angiography (XA), nuclear medicine (NM) and other image modalities.

[0092] Exemplarily, a user can import / upload multiple medical images of the object to be tested under different modalities, and the computer device can also obtain multiple medical images of the object to be tested under different modalities from a preset database; wherein the preset database may include a local database of the computer device, a cloud database connected to the computer device in communication, such as a Picture Archiving and Communication System (PACS), etc.

[0093] Exemplarily, the computer device can obtain medical images of the object to be tested in different modalities from a preset database according to the identification of the object to be tested. In one implementation, assuming that images of multiple modalities of the object to be tested are stored in the preset database, and there are multiple images corresponding to at least one modality, then the computer device can obtain the latest image of the medical scan time among the multiple images under the modality with multiple images, and import it. For example: the object to be tested includes CT images, MR images, and PET images, wherein the CT images of the object to be tested include CT images obtained by scanning at different stages; then, when the computer device imports the CT images of the object to be tested, it obtains the CT image obtained by scanning at the most recent stage and imports it.

[0094] Exemplarily, when a computer device obtains images of different modalities of an object to be measured from a preset database, it can obtain images corresponding to each modality of the object to be measured from the preset database, or it can obtain images corresponding to some modalities of the object to be measured; for example: the computer device can determine the required modality according to the target procedure of the object to be measured, and then obtain the image corresponding to the modality required for the target procedure from the preset database; for another example: the computer device can also obtain images corresponding to each modality input by the user from the preset database based on multiple modalities input by the user.

[0095] In addition, it should be noted that the modalities of the above-mentioned multiple medical images can be different; of course, the modalities of some medical images can also be the same, that is, one modality can also correspond to multiple medical images. For example: importing CT images of the object to be tested scanned at different stages, such as a CT image scanned when the patient was just admitted to the hospital and a CT image scanned before the patient's operation.

[0096] Furthermore, when performing tissue segmentation, it is necessary not only to obtain multiple medical images of the object to be tested, but also to obtain the tissue to be segmented of the object to be tested; illustratively, the computer device can obtain at least one tissue to be segmented of the object to be tested input by the user, and can also determine at least one tissue to be segmented required for the target procedure according to the target procedure of the object to be tested. For example: the computer device can determine at least one tissue to be segmented corresponding to the target procedure according to a preset third correspondence; the third correspondence may include a correspondence between different procedures and different tissues, wherein one procedure may correspond to one or more tissues; illustratively, the third correspondence may be pre-set by the user according to the tissues required by different procedures and stored in the computer device; for example: SEEG procedures may correspond to tissues such as blood vessels, brain parenchyma, and skull; DBS procedures may correspond to tissues such as nuclei, blood vessels, and skull; puncture procedures may correspond to tissues such as blood vessels and skull.

[0097] Step 204 : for each tissue to be segmented, determine the modality corresponding to the tissue to be segmented according to a preset first corresponding relationship.

[0098] The first correspondence includes the correspondence between different tissues and different modalities, and the segmentation effect of the tissue in the medical image of the corresponding modality satisfies the segmentation condition. For example, in the first correspondence, one tissue may correspond to one modality with the best segmentation effect, and of course, one tissue may also correspond to multiple modalities with good segmentation effects.

[0099] For example, blood vessels can correspond to modalities such as time of flight (TOF), phase contrast (PC), and coronary computed tomography angiography (CTA); the skull can correspond to modalities such as CT and CTA; nuclei can correspond to modalities such as T1 (longitudinal relaxation time) and T2 (Spin-Spin (transverse) relaxation time) under MR; the brain parenchyma can correspond to modalities such as T2 Fluid Attenuated Inversion Recovery (T2-flair), T1, and T2.

[0100] In the above example, a tissue may correspond to multiple modalities, and the segmentation effect of the tissue in each corresponding modality is good, satisfying certain segmentation conditions. In an optional implementation, when the tissue corresponds to multiple modalities with good segmentation effects, further, a modality with the best segmentation effect can be determined from the multiple modalities with good segmentation effects corresponding to the tissue, thereby establishing a correspondence between the tissue and the modality with the best segmentation effect, and obtaining a first correspondence. Based on this, the modality with the best segmentation effect corresponding to the tissue can be determined based on the first correspondence, so that the tissue can be segmented based on the medical image corresponding to the modality with the best segmentation effect, thereby obtaining a more accurate tissue segmentation result.

[0101] For example, the doctor may determine the correspondence between the modality and the tissue based on experience, thereby obtaining the first correspondence, and store the first correspondence in a computer device.

[0102] Exemplarily, for each tissue, the computer device may also use medical images of different modalities to perform segmentation processing on the tissue respectively based on a large number of medical images of different modalities, so as to determine one or more image modalities whose segmentation effect meets the segmentation conditions after segmentation of the tissue, thereby establishing a correspondence between the tissue and the modality; in the same way, the computer device can obtain the modalities matched by different tissues respectively, and finally generate a first correspondence between different tissues and different modalities.

[0103] Exemplarily, based on the first corresponding relationship, each tissue to be segmented of the object to be tested can be matched with a modality with a better segmentation effect corresponding to the tissue to be segmented, and then the tissue can be segmented based on the medical image corresponding to the modality to obtain a better tissue segmentation result.

[0104] It should be noted that, since in the first corresponding relationship, one tissue may correspond to one or more modalities, for multiple tissues to be segmented of the object to be tested, the modalities corresponding to the tissues to be segmented may be the same or different.

[0105] Step 206 : determining a target medical image corresponding to the tissue to be segmented from the plurality of medical images according to the modality corresponding to the tissue to be segmented.

[0106] Exemplarily, after determining the modality corresponding to each tissue to be segmented, the computer device can, for each tissue to be segmented, filter out the medical image corresponding to the modality from multiple medical images of the object to be tested according to the modality corresponding to the tissue to be segmented, as the target medical image corresponding to the tissue to be segmented.

[0107] Exemplarily, when there are multiple modalities corresponding to the tissue to be segmented, a medical image corresponding to each modality can be determined from multiple medical images, and the medical image corresponding to each modality can be used as a target medical image corresponding to the tissue to be segmented.

[0108] Exemplarily, in the case where a modality corresponding to the tissue to be segmented corresponds to multiple medical images, the computer device can also determine multiple candidate medical images corresponding to the modality from the multiple medical images, and use the multiple candidate medical images as target medical images corresponding to the tissue to be segmented.

[0109] For example: when the tissue to be segmented is tissue x, the modalities corresponding to tissue x include modality a and modality b, and modality a includes medical image a1 and medical image a2, and modality b includes medical image b1, medical image b2, and medical image b3, the target medical images corresponding to tissue x may include medical image a1, medical image a2, medical image b1, medical image b2, and medical image b3.

[0110] It should be noted that, when the object to be detected includes multiple tissues to be segmented, and there are at least two tissues to be segmented with the same modality, then the target medical images corresponding to the at least two tissues to be segmented may also be the same.

[0111] Step 208 : segment the tissue to be segmented in the target medical image to obtain a tissue segmentation result of the target medical image.

[0112] Exemplarily, after the computer device has respectively determined the target medical image corresponding to each tissue to be segmented of the object to be tested, it can use the target medical image corresponding to the tissue to be segmented to perform segmentation processing on the tissue to be segmented in the target medical image for each tissue to be segmented, thereby obtaining the tissue segmentation result of the tissue to be segmented in the target medical image.

[0113] Exemplarily, when there are multiple target medical images corresponding to the tissue to be segmented, the computer device may perform segmentation processing on the tissue to be segmented in each target medical image to obtain a tissue segmentation result for each target medical image.

[0114] Exemplarily, when there are multiple target medical images corresponding to the tissue to be segmented, the computer device may also output and display the multiple target medical images corresponding to the tissue to be segmented, and perform segmentation processing on the tissue to be segmented according to one or more target medical images selected by the user from the multiple target medical images, thereby obtaining the tissue segmentation results of the one or more target medical images selected by the user.

[0115] For example, when the tissues to be segmented include nuclei, blood vessels, and skulls, it is assumed that the target medical image corresponding to the nuclei is a T1 image, the target medical image corresponding to the blood vessels is a PET image, and the target medical image corresponding to the skulls is a CT image; then, by segmenting the nuclei in the T1 image, the nuclei segmentation result of the T1 image can be obtained; by segmenting the blood vessels in the PET image, the blood vessel segmentation result of the PET image can be obtained; by segmenting the skull in the CT image, the skull segmentation result of the CT image can be obtained.

[0116] Exemplarily, when multiple tissues to be segmented correspond to the same target medical image, the tissue segmentation results of different tissues to be segmented in the target medical image can be obtained respectively; for example, the tissue segmentation result of A1 in the target medical image can be obtained by segmenting the tissue to be segmented A1; and the tissue segmentation result of A2 in the target medical image can be obtained by segmenting the tissue to be segmented A2. In other words, the tissue segmentation result of the target medical image can include the sub-tissue segmentation results of multiple different tissues in the target medical image.

[0117] Exemplarily, in the case where multiple tissues to be segmented correspond to the same target medical image, when different sub-tissue segmentation results of the target medical image are obtained, the different sub-tissue segmentation results of the target medical image can also be fused to obtain a fused tissue segmentation result of the target medical image. For example, the A1 tissue segmentation result of the target medical image and the A2 tissue segmentation result of the target medical image are fused to obtain a fused tissue segmentation result of the target medical image containing both tissue A1 and tissue A2.

[0118] In the above tissue segmentation method, the computer device obtains multiple medical images of the object to be tested and at least one tissue to be segmented of the object to be tested; wherein, the modalities of at least some of the multiple medical images are different; then, for each tissue to be segmented, the modality corresponding to the tissue to be segmented is determined according to a preset first correspondence relationship; and according to the modality corresponding to the tissue to be segmented, a target medical image corresponding to the tissue to be segmented is determined from the multiple medical images; and then the tissue to be segmented in the target medical image is segmented to obtain a tissue segmentation result of the target medical image. The first correspondence relationship includes the correspondence between different tissues and different modalities, and the segmentation effect of the tissue in the medical image of the corresponding modality meets the segmentation condition. That is to say, in the embodiment of the present application, for the situation where multiple modal images are used to segment different tissues, the computer device can automatically match the tissue to be segmented with the modal image to determine the target medical image corresponding to the modality with the best segmentation effect for the tissue to be segmented; using this method, there is no need for the user to match the tissue to be segmented with the corresponding modal image based on subjective judgment or experience judgment, which can not only reduce the user's human participation in the tissue segmentation process, but also improve the matching degree between the tissue and the modal image, and avoid the problem of poor tissue segmentation effect when the user selects a modal image with poor segmentation effect for tissue segmentation due to lack of experience; that is, using the method of the present application, not only can the segmentation effect of different tissues in multi-modal images be improved, but also the efficiency of tissue segmentation can be improved.

[0119] In an exemplary embodiment, Figure 3 As shown, the process of "obtaining multiple medical images of the object to be tested" in the above step 202 may include steps 302 to 306. Among them:

[0120] Step 302: Obtain a target technique of the object to be tested selected by a user from at least one candidate technique.

[0121] Exemplarily, a display interface of a computer device may provide a user with multiple candidate formula options, and the user may select a target formula for the object to be tested through the multiple candidate formula options on the display interface; the computer device may obtain the target formula for the object to be tested selected by the user from at least one candidate formula by responding to a trigger operation of the user on the display interface.

[0122] Step 304: Determine multiple candidate modalities corresponding to the target procedure according to the preset second corresponding relationship.

[0123] Among them, the second correspondence includes the correspondence between different procedures and different modalities, and one procedure can correspond to multiple modalities. Exemplarily, the second correspondence can be pre-set and stored in the computer device based on the user's requirements for modal images of different procedures. For example: SEEG procedures can correspond to T1, T2-Flair, Phase Contrast Angiography (PCA), TOF, PET, CT and other modalities; DBS procedures can correspond to T1, T2, TOF, CT and other modalities; puncture procedures can correspond to T1, T2, CT and other modalities.

[0124] It should be noted that the multiple modalities corresponding to different procedures may be the same or different. In addition, for the modalities corresponding to each procedure in the second correspondence, the user can also add a custom modality for the procedure; for example, if a CTA modality is added to a puncture procedure, the modalities corresponding to the puncture procedure may include T1, T2, CT and CTA.

[0125] Exemplarily, after obtaining the target technique for the object to be tested, the computer device can determine multiple candidate modalities corresponding to the target technique based on the second corresponding relationship. That is, the computer device can automatically match the required modalities for different techniques to avoid the user from determining the required modality for the target technique by himself, which can not only improve the intelligence of the software system, but also improve the convenience of user operation, reduce the steps of manual operation of the user, and reduce the tediousness of manual operation.

[0126] Step 306: Based on the multiple candidate modalities, obtain a medical image corresponding to each candidate modality of the object to be tested.

[0127] Exemplarily, the computer device may obtain medical images corresponding to each candidate modality of the object to be tested from a preset database based on multiple candidate modalities corresponding to the target surgical procedure of the object to be tested.

[0128] Exemplarily, the computer device may also generate an image input component corresponding to each candidate modality based on the multiple candidate modalities, and display the image input component corresponding to each candidate modality in the display interface; based on this, the user can import the medical image of the object to be tested corresponding to each candidate modality based on each image input component. In other words, for the image input component corresponding to each candidate modality, the computer device can obtain the medical image of the candidate modality of the object to be tested corresponding to the image input component in response to the user's image input operation on the image input component.

[0129] Exemplarily, the computer device obtains a medical image of a candidate modality of the object to be tested corresponding to the image input component in response to an image input operation of the user on the image input component, which may include: obtaining the candidate medical image input by the user in response to the image input operation of the user on the image input component; and when the modality of the candidate medical image is consistent with the candidate modality corresponding to the image input component, importing the candidate medical image as the candidate modality of the medical image of the object to be tested corresponding to the image input component. That is to say, for each image input component, after the user selects the medical image of the object to be tested, the computer device can also match the modality of the medical image selected by the user with the modality corresponding to the image input component; if the match is successful, it can be said that the modality of the medical image selected by the user is consistent with the modality corresponding to the image input component, and the medical image can be imported at this time; if the match is unsuccessful, it means that the import has failed. At this time, the computer device can also output a reminder message of import failure, or modality mismatch / inconsistency, to remind the user to re-select the medical image of the modality corresponding to the image input component for import, so as to ensure that the medical images imported by the user are medical images of the modalities required for the target procedure, thereby improving the matching degree between the medical image and the procedure.

[0130] Exemplarily, for a modality, it can support the input of multiple medical images corresponding to the modality. Based on this, for the image input component corresponding to the candidate modality, it can include at least one image input subcomponent of the candidate modality, wherein the image input subcomponent is used to import the medical image of the candidate modality. Exemplarily, a medical image of a candidate modality can be imported through an image input subcomponent, then multiple image input subcomponents can import multiple medical images of the candidate modality.

[0131] Exemplarily, the image input component corresponding to the candidate modality may also include an adding subcomponent, which may be used to add an image input subcomponent of the candidate modality. That is, when the number of medical images of the candidate modality that the user wants to import is greater than the image input subcomponent of the candidate modality displayed in the display interface, the user may also use the adding subcomponent to add additional image input subcomponents, so that the user can import more medical images of the object to be tested under the candidate modality to meet the user's needs.

[0132] Exemplarily, for each candidate modality, a preset number of image input subcomponents, such as one or two, can be displayed by default, and the user can adaptively increase the image input subcomponents according to the number of medical images they want to import. This not only avoids excessive display of image input subcomponents that would otherwise occupy too large a portion of the interface, but also allows for flexible operation and settings according to different scenario needs. This improves interface utilization while also meeting user needs and enhancing user experience.

[0133] In this embodiment, by obtaining the target procedure of the object to be measured selected by the user from at least one candidate procedure, multiple candidate modalities corresponding to the target procedure are determined according to a preset second correspondence relationship; then, based on the multiple candidate modalities, medical images corresponding to each candidate modality of the object to be measured are obtained; wherein the second correspondence relationship includes the correspondence between different procedures and different modalities; that is, in this embodiment, the computer device can not only provide selection operations for different procedures, but also automatically match the image modality corresponding to the target procedure based on the target procedure selected by the user, and further obtain the medical image of the object to be measured corresponding to each image modality corresponding to the target procedure; that is, in this embodiment, the computer device can not only automatically guide the acquisition of multimodal images based on different procedures, but also accurately match multimodal images with different tissues, so as to segment the best tissue from the multimodal image and improve the tissue segmentation effect and efficiency.

[0134] In an exemplary embodiment, the first corresponding relationship may also be generated based on a plurality of medical images of the object to be tested in different modalities; based on the above embodiment, the method may further include: obtaining the first corresponding relationship; Figure 4 As shown, the step of acquiring the first corresponding relationship may include steps 402 to 406. Among them:

[0135] Step 402 , performing tissue recognition on a plurality of medical images respectively to obtain tissue recognition results of each medical image.

[0136] The tissue recognition result includes the tissue recognized from the corresponding medical image.

[0137] For different detection objects, since the display effects of various tissues in medical images of different objects in different modalities may be different, the display effects of the same tissue in medical images of different objects in different modalities will also be different. For example, the display effect of tissue a of object A is the best in the medical image of the first modality, but the display effect of tissue a of object B is the best in the medical image of the second modality. If the first correspondence is set according to experience or historical image learning, it is possible that the match between a certain tissue and the modality is not optimal for a certain object.

[0138] Based on this, in order to solve this problem, in this embodiment, the precise matching relationship between different tissues and modalities is determined based on multiple medical images of the object to be tested in different modalities. Exemplarily, when the computer device obtains multiple medical images of different modalities of the object to be tested, it can perform tissue recognition on each medical image respectively, so as to obtain the tissue recognition results of each medical image; wherein, the tissue recognition result of the medical image can be a fused tissue segmentation result including multiple tissues, or it can be multiple tissue segmentation results, each of which only includes one tissue of the object to be tested.

[0139] Step 404: determine the tissue corresponding to each medical image based on the tissue identification results and the segmentation conditions.

[0140] Exemplarily, for each tissue of the object to be tested, the computer device can determine one or more tissue identification results corresponding to the tissue from the tissue identification results of each medical image; when there is only one tissue identification result for the tissue, the medical image corresponding to the tissue identification result can be determined as the target medical image corresponding to the tissue.

[0141] Exemplarily, in the case where a tissue corresponds to multiple tissue recognition results, the computer device may select a tissue recognition result with the best tissue recognition effect from the multiple tissue recognition results, and determine the medical image corresponding to the tissue recognition result with the best tissue recognition effect as the target medical image corresponding to the tissue. For example, in the case where a blood vessel corresponds to a blood vessel recognition result of a CT image and a blood vessel recognition result of an MR image, assuming that the recognition effect of the blood vessel recognition result of the CT image is better than that of the blood vessel recognition result of the MR image, then the CT image may be used as the target medical image of the blood vessel.

[0142] Next, when the target medical images corresponding to the tissues are determined, the tissues corresponding to the medical images can be determined from the perspective of the medical images. For example, when the blood vessels correspond to the CT images, the skulls correspond to the CT images, and the nuclei correspond to the T1 images, it can be determined that the tissues corresponding to the CT images include the blood vessels and the skulls, and the tissues corresponding to the T1 images include the nuclei.

[0143] Step 406: Establish a correspondence between the modality of the medical image and the corresponding tissue to obtain a first correspondence.

[0144] That is to say, when the computer device determines the tissue corresponding to each medical image of the object to be tested, it can establish a mutual correspondence between the modality of each medical image and the tissue corresponding to the medical image, thereby obtaining a first correspondence between different tissues and corresponding different modalities.

[0145] In this embodiment, the computer device performs tissue recognition on multiple medical images respectively to obtain tissue recognition results of each medical image; the tissue recognition results include tissues recognized from the corresponding medical images; then, based on the tissue recognition results and segmentation conditions, the tissues corresponding to each medical image are determined, and a correspondence between the modality of the medical image and the corresponding tissue is established to obtain a first correspondence. That is, in this embodiment, when the computer device matches tissues with images, it matches the modality image with the best segmentation effect for each tissue based on the tissues recognized from multiple medical images of the object to be tested and the recognition effect of the tissue in each medical image; this method can adapt to the differences between different modality images of different objects, ensure that the tissue is matched to the modality image with the best segmentation effect, so as to improve the segmentation effect of the subsequent segmentation of the tissue based on the corresponding modality image.

[0146] In an exemplary embodiment, Figure 5 As shown, before executing the above step 208, steps 502 to 504 may also be included. Among them:

[0147] Step 502, displaying at least one group of segmentation options; each group of segmentation options includes a tissue to be segmented and a corresponding target medical image.

[0148] That is, when the computer device determines the target medical image corresponding to each tissue to be segmented, it can output and display each tissue to be segmented and the corresponding target medical image to show the user the target medical image matched to each tissue to be segmented.

[0149] Exemplarily, a computer device may display at least one set of segmentation options to a user, wherein the segmentation options include not only information such as the tissue to be segmented and the corresponding target medical image, but also selection attributes so that the user can choose to perform tissue segmentation processing on the tissue to be segmented, or not to perform tissue segmentation processing on the tissue to be segmented.

[0150] Exemplarily, for the information displayed in the segmentation option, the information of the tissue to be segmented may include at least one of an identification, an image, etc. of the tissue to be segmented; the information of the target medical image corresponding to the tissue to be segmented may include at least one of the target medical image, the identification or the name of the target medical image, etc.; for example: the segmentation option may display the identification of the tissue to be segmented and the identification of the corresponding target medical image, such as: a vascular-CT image; the segmentation option may also display the image of the tissue to be segmented and the corresponding target medical image, wherein the image of the tissue to be segmented may be a tissue schematic diagram of the tissue to be segmented.

[0151] Step 504: Obtain a target segmentation option selected by the user from at least one group of segmentation options.

[0152] Accordingly, the above step 208 may include:

[0153] Step 506 , segmenting the tissue to be segmented in the target medical image corresponding to the target segmentation option to obtain a tissue segmentation result of the target medical image.

[0154] That is, the computer device can perform segmentation processing on the target segmentation option selected by the user, using the target medical image in the target segmentation option, and the tissue to be segmented in the target segmentation option, thereby obtaining the tissue segmentation result of the tissue to be segmented in the target medical image. For the unselected segmentation option, the tissue segmentation processing process is not performed on the tissue to be segmented in the unselected segmentation option.

[0155] Exemplarily, in the case where the tissue to be segmented displayed in the target segmentation option corresponds to multiple target medical images, an autonomous selection function can also be provided to the user, so that the user can select the target medical image to be segmented according to needs; that is, for the segmentation option, it not only has a selection attribute, but also the medical images corresponding to the tissue to be segmented displayed in the segmentation option also have a selection attribute. The user can freely select the medical image for tissue segmentation, so as to provide the user with more functional choices and flexible operations.

[0156] Exemplarily, for each target segmentation option, the computer device can obtain at least one selected target medical image selected by the user from the multiple target medical images displayed in the target segmentation option; and segment the tissue to be segmented in each selected target medical image corresponding to the target segmentation option to obtain the tissue segmentation result of each selected target medical image. This allows the computer device to perform tissue segmentation on all target medical images corresponding to the tissue to be segmented, but only segment the tissue to be segmented in the required target medical image according to the user's selection, which can not only reduce the algorithm processing amount of the computer device, but also meet the needs of the user and improve the segmentation efficiency.

[0157] Exemplarily, when the computer device does not detect a user's selection operation on multiple target medical images corresponding to the target segmentation option, it can determine a default target medical image from the multiple target medical images, and segment the tissue to be segmented in the default target medical image to obtain a tissue segmentation result of the default target medical image.

[0158] Exemplarily, the default target medical image may be a preset number of target medical images determined according to different screening strategies; for example: arbitrarily selecting a target medical image from a plurality of target medical images as the default target medical image; or, according to the arrangement order of the plurality of target medical images, selecting the first target medical image as the default target medical image, etc.

[0159] In an optional implementation, multiple target medical images can be sorted in descending order according to the segmentation effect, that is, the segmentation effect of the target medical images closer to the front is better than that of the target medical images closer to the back; based on this, when the computer device determines the default target medical image from the multiple target medical images, the first target medical image in the descending order can be used as the default target medical image, that is, the target medical image with the best segmentation effect can be used as the default target medical image.

[0160] Exemplarily, in the case where the tissue to be segmented corresponds to multiple target medical images, from the perspective of interface usage and aesthetics, multiple target medical images can also be displayed in the form of a drop-down list, that is, the first target medical image is displayed in the target segmentation option, and other target medical images except the first target medical image are hidden; accordingly, the target segmentation option can also include a drop-down component, and the computer device can display other target medical images hidden in the drop-down list in response to the user's triggering operation on the drop-down component. Exemplarily, in the case of a large number of target medical images, the drop-down list can also display multiple target medical images in a sliding form to avoid the drop-down list being too long and affecting the surrounding displayed content, thereby improving the interface display effect and increasing the interface aesthetics.

[0161] In this embodiment, when the computer device matches the corresponding target medical image for each tissue to be segmented, the computer device can display segmentation options including the tissue to be segmented and the corresponding target medical image to the user, so that the user can select the target segmentation option that requires tissue segmentation from at least one group of segmentation options displayed, and then the computer device performs segmentation processing on the tissue to be segmented in the target medical image corresponding to the selected target segmentation option, thereby obtaining the tissue segmentation result of the target medical image; the method in this embodiment can provide the user with selectable tissue segmentation options, and combine with user operations to perform segmentation processing on the tissue required by the user, thereby improving the flexibility and diversity of tissue segmentation.

[0162] In an exemplary embodiment, the computer device can also display the segmentation process and segmentation results of the tissue to be segmented in real time during the process of segmenting each tissue to be segmented; for example: the computer device can display the segmentation response corresponding to the target segmentation option at a preset position on the target segmentation option; wherein the segmentation response may include the segmentation process and / or the segmentation result, and the segmentation result may include segmentation success or segmentation failure.

[0163] Exemplarily, the segmentation response can be displayed at any position on the corresponding target segmentation option, or at any position around the corresponding target segmentation option; for example: when the segmentation option includes at least one of an image of the tissue to be segmented and a corresponding target medical image, the segmentation response can be displayed at any position in the area where the image is located; for another example: when the segmentation option only includes the identification of the tissue to be segmented and the identification of the corresponding target medical image, the segmentation response can also be displayed at any position around the segmentation option.

[0164] For example, for the segmentation process of the tissue to be segmented, a progress bar can be used to dynamically display the segmentation progress of the tissue to be segmented; for the segmentation results of the tissue to be segmented, preset icons or graphics corresponding to different segmentation results can be used for static display. For example: for the result of successful segmentation, a check mark (√) can be used to indicate it, and for the result of failed segmentation, a wrong mark (×) can be used to indicate it; or other graphics and colors that are easy to distinguish between success and failure can be used to indicate it, which is not specifically limited in the embodiments of the present application.

[0165] In this embodiment, the segmentation option can not only display the tissue to be segmented and the corresponding target medical image to show the user the matching results of the tissue and the image; it can also display the segmentation status of the tissue to be segmented so that the user can timely grasp the real-time segmentation status of each tissue to be segmented; that is, the method in this embodiment can provide the user with stronger information feedback and improve the user's grasp of the tissue segmentation process.

[0166] In an exemplary embodiment, the computer device can also fuse and display the tissue segmentation results of each target medical image. Exemplarily, the tissue segmentation results of each target medical image can include the tissue segmentation results of the target medical image corresponding to each tissue to be segmented, that is, the computer device can fuse the tissue segmentation results corresponding to each tissue to be segmented, and display the fused fused tissue segmentation result. Among them, the fused tissue segmentation result includes the segmented tissues to be segmented of the object to be tested. In addition, the fused tissue segmentation result can be a tissue mask image including the tissues to be segmented of the object to be tested, and each tissue to be segmented can be marked with a different color.

[0167] In this embodiment, the computer device can also fuse the tissue segmentation results of each tissue to be segmented so that the user can simultaneously view the segmentation results of multiple tissues in one tissue mask image; using this method, the user can avoid viewing the segmentation results of different tissues separately, which can improve the display effect of multi-tissue segmentation.

[0168] In an exemplary embodiment, Figure 6 As shown, before the above step 208, steps 602 to 604 may also be included. Among them:

[0169] Step 602: Determine the target initialization operation corresponding to the target technique according to the preset fourth corresponding relationship.

[0170] The fourth corresponding relationship may include the corresponding relationship between different techniques and different initialization operations. One technique may correspond to one or more initialization operations; and the initialization operations corresponding to different techniques may be the same or different.

[0171] Step 604, perform target initialization operation.

[0172] That is, before the computer executes the tissue segmentation step, one or more operations that need to be performed before tissue segmentation can be determined according to the target procedure, and these operations can be automatically performed by the computer device or guided by the user.

[0173] Exemplarily, after acquiring the target formula selected by the user, the computer device may determine the target initialization operation corresponding to the target formula and execute the target initialization operation.

[0174] Exemplarily, the computer device may also determine a target initialization operation corresponding to the target procedure and execute the target initialization operation before the user determines to segment the tissue.

[0175] Of course, the computer device can also perform a target initialization operation corresponding to the target procedure at any stage before performing tissue segmentation; the embodiment of the present application does not limit the timing of performing the target initialization operation before tissue segmentation.

[0176] In this embodiment, the computer device can automatically match the necessary target initialization operation based on the target procedure, and automatically execute or guide the user to execute the target initialization operation before tissue segmentation; it can avoid the situation where the user misses the necessary steps before tissue segmentation, resulting in tissue segmentation failure, thereby improving the reliability of tissue segmentation. In addition, by determining the target initialization operation corresponding to the target procedure and executing the target initialization operation, the user's mastery of the procedure function can be further reduced, that is, the difficulty and professionalism of software operation can be further reduced.

[0177] In an exemplary embodiment, Figure 7 As shown, a software system operation process is provided, including the following steps:

[0178] Step 1: Select the target technique.

[0179] For different techniques, the key functions they focus on are also different. For example, after the user selects the target technique, the software system can filter out at least one technique function corresponding to the target technique from all the functions indicated by the software system according to the target technique, and hide other functions. That is, in this example, the computer device can provide different technique functions for different technique selections to avoid interference of other technique functions on the user's selection.

[0180] Step 2: Modality matching.

[0181] When the user selects a target procedure for the object to be measured, the computer device can automatically match the required image modality according to the target procedure, and automatically acquire or receive medical images of various modalities of the object to be measured input by the user.

[0182] Since different surgical procedures use different image modalities differently, the software system can automatically match the image modality corresponding to the target surgical procedure based on the target surgical procedure selected by the user; that is, different modality images are selected for import according to different surgical procedures. For example, the SEEG procedure generally requires CT modality images. Through the surgical modality matching operation, it is possible to avoid non-standard data (i.e., modality images not required for surgical procedure-related functions) from being unable to be recognized by the system in the initial stage of the operation, resulting in the inability to use surgical procedure-related functions normally during the operation, thereby leading to surgical failure, thereby improving the fault tolerance of the software system.

[0183] Exemplarily, the computer device may also display input components corresponding to the candidate modalities in the display interface when multiple candidate modalities corresponding to the target procedure are determined, so as to provide the user with a space for independent selection based on the automatic recognition of the modality, so that the user can input medical images of the corresponding modality according to the needs, thereby improving the flexibility and diversity of user operations. Exemplarily, on this basis, a custom input component may also be added to the display interface so that the user can add additional images of other modalities to meet the user's usage needs.

[0184] Step 3: Initialize the procedure function.

[0185] According to different techniques selected by the user, the software system will perform different initialization operations. Exemplarily, the computer device can determine the target initialization operation corresponding to the target technique according to the target technique, and perform the target initialization operation.

[0186] In order to further reduce the user's operation, before entering the tissue segmentation page, the system can automatically guide and perform initialization operations; for example, for DBS procedures, the system's initialization operations may include procedure template selection, registration and fusion of multiple medical images, ACPC correction guidance, and automatic extraction of cortical markers to assist in the normal operation of subsequent functions. During the initialization process, if ACPC is an active guidance operation performed by the system, the cortical markers can be automatically extracted by the system based on the existing ACPC correction information, and no manual operation by the user is required in this process.

[0187] When the computer device obtains multiple medical images of the object to be tested in different modalities, it can automatically jump or guide the user to enter the interface for automatic registration of multiple images. In this interface, the multiple medical images are first registered and fused to obtain a fused medical image; then, after the fusion operation is completed, it can automatically jump or guide the user to enter the ACPC correction interface, and then the initial ACPC correction can be performed. After the correction is completed, the corresponding cortical marker point information is automatically extracted according to the ACPC information. After a series of initialization operations are completed, the tissue segmentation process can be entered.

[0188] The initialization operation not only realizes the automation of certain functions, but also realizes the standardized operation of related steps, reducing the dependence on the doctor's prior knowledge.

[0189] Step 4: Intelligently identify the segmentable tissues and the corresponding optimal sequences.

[0190] Different modal images can extract different optimal tissues. The software system will automatically identify the sequence (i.e., modal images) and corresponding tissues that can achieve the best recognition effect, and group the corresponding tissues and images for display.

[0191] Exemplarily, the computer device can perform tissue recognition on multiple medical images respectively to obtain tissue recognition results of each medical image; determine the tissue corresponding to each medical image based on each tissue recognition result and segmentation condition; and then establish a correspondence between the modality of the medical image and the corresponding tissue to obtain a first correspondence. Based on the first correspondence, modality matching can be performed on at least one to-be-segmented tissue corresponding to the target procedure, thereby determining the target medical image corresponding to each to-be-segmented tissue from multiple medical images of the subject to be tested.

[0192] Furthermore, the computer device can group the tissues to be segmented and the corresponding target medical images, arrange them in sequence in the form of segmentation options and display them to the user, as shown in FIG8(a); wherein the segmentation options may include a first area at the top and a second area at the bottom, the first area being used to display the graphics / image of the tissue to be segmented, and the second area being used to display the identification of the target medical image corresponding to the tissue to be segmented.

[0193] The user can select a target segmentation option for tissue segmentation from multiple segmentation options; illustratively, referring to FIG8( a ), a check mark (√) is marked in the upper right corner of the selected target segmentation option.

[0194] Exemplarily, for each segmentation option, the user may also modify the target medical image corresponding to the tissue to be segmented in the segmentation option to meet the user's specialized needs.

[0195] Step 5: Automatically split with one click.

[0196] After the user triggers the determination component in the display interface, the computer device can perform tissue segmentation processing on the tissue to be segmented in the target segmentation option selected by the user, and display the segmentation progress at the center of the first area on each target segmentation option, as shown in Figure 8(b). Exemplarily, during the process of tissue segmentation, the determination component in the display interface can be switched from a triggerable state to a trigger-disabled state. In other words, for multiple target segmentation options selected by the user, different tissues on different target medical images can be extracted and displayed at the same time through automatic one-click extraction, avoiding the user from segmenting different tissues separately, which can reduce the user's operation steps.

[0197] Step 6: Segmentation is complete.

[0198] After the segmentation is completed, the computer device can also display the tissue segmentation results of each tissue to be segmented, that is, display the segmentation response corresponding to the segmentation option, including segmentation success or segmentation failure; as shown in Figure 8(c), the computer device can display the response of segmentation success or segmentation failure at the center of the first area of ​​the target segmentation option. In this example, by providing strong feedback on the results of automatic segmentation, it is also possible to avoid the user from performing repeated segmentation operations without noticing the prompt that the segmentation is completed.

[0199] Exemplarily, for the tissue to be segmented that fails to be segmented, the user can also change the target medical image corresponding to the tissue to be segmented that fails to be segmented, so as to instruct the computer device to re-segment the tissue to be segmented based on the changed medical image; of course, for the tissue to be segmented that fails to be segmented, the user can also manually segment the tissue to be segmented that fails to be segmented based on any medical image of multiple medical images of the object to be tested.

[0200] Exemplarily, when all the tissues to be segmented have been segmented, the computer device can also fuse and display the tissue segmentation results of the tissues to be segmented; that is, multiple segmented tissues are merged and fused to display the multiple tissues in a fused segmented image at the same time; the fused segmented image can be a mask image including multiple tissues, or it can be a medical image in which multiple tissue masks are merged and displayed in a fused image.

[0201] Exemplarily, the computer device may also fuse the tissue segmentation results of the medical images belonging to the same modality for each medical image under different modalities, that is, obtain a fused segmented image corresponding to the medical image of the same modality. The fused segmented image may be a mask image of one or more tissues after tissue segmentation of the medical image of the modality, or may be a fused mark of one or more tissue masks in the medical image of the modality; exemplarily, when one or more tissue masks are fused in the medical image of the modality, other tissues in the medical image of the modality may be hidden, and only one or more tissues segmented by the medical image may be displayed.

[0202] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0203] Based on the same inventive concept, the embodiment of the present application also provides a tissue segmentation device for implementing the tissue segmentation method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more tissue segmentation device embodiments provided below can refer to the limitations on the tissue segmentation method above, and will not be repeated here.

[0204] In an exemplary embodiment, Fig. 9 As shown, a tissue segmentation device is provided, comprising: a first acquisition module 902, a first determination module 904, a second determination module 906 and a segmentation module 908, wherein:

[0205] The first acquisition module 902 is used to acquire multiple medical images of the object to be detected and to acquire at least one tissue to be segmented of the object to be detected; wherein at least some of the multiple medical images have different modalities.

[0206] The first determination module 904 is used to determine the modality corresponding to the tissue to be segmented according to a preset first correspondence relationship for each tissue to be segmented; the first correspondence relationship includes the correspondence between different tissues and different modalities, and the segmentation effect of the tissue in the medical image of the corresponding modality meets the segmentation condition.

[0207] The second determination module 906 is configured to determine a target medical image corresponding to the tissue to be segmented from among multiple medical images according to the modality corresponding to the tissue to be segmented.

[0208] The segmentation module 908 is used to perform segmentation processing on the tissue to be segmented in the target medical image to obtain a tissue segmentation result of the target medical image.

[0209] In one embodiment, the first acquisition module 902 includes:

[0210] A first acquisition submodule is used to acquire a target formula of the object to be tested selected by a user from at least one candidate formula;

[0211] A first determination submodule is used to determine multiple candidate modalities corresponding to the target procedure according to a preset second correspondence relationship; the second correspondence relationship includes a correspondence relationship between different procedures and different modalities;

[0212] The second acquisition submodule is used to acquire the medical image corresponding to each candidate modality of the object to be tested based on multiple candidate modalities.

[0213] In one embodiment, the second acquisition submodule includes:

[0214] A display unit, used to generate and display an image input component corresponding to each candidate modality based on multiple candidate modalities;

[0215] The acquisition unit is used to acquire, for each image input component, a medical image of a candidate modality corresponding to the image input component of the object to be tested in response to an image input operation of the user on the image input component.

[0216] In one of the embodiments, the acquisition unit is specifically used to obtain a candidate medical image input by the user in response to an image input operation of the user on the image input component; when the modality of the candidate medical image is consistent with the candidate modality corresponding to the image input component, the candidate medical image is used as the medical image of the candidate modality corresponding to the image input component of the object to be tested and imported.

[0217] In one of the embodiments, the image input component corresponding to the candidate modality includes at least one image input subcomponent of the candidate modality, and the image input subcomponent is used to import the medical image of the candidate modality.

[0218] In one of the embodiments, the image input component corresponding to the candidate modality also includes an adding subcomponent, and the adding subcomponent is used to increase the image input subcomponent of the candidate modality.

[0219] In one embodiment, the first acquisition module 902 further includes:

[0220] The second determination submodule is used to determine at least one to-be-segmented tissue corresponding to the target surgical procedure according to a preset third correspondence relationship; the third correspondence relationship includes a correspondence relationship between different surgical procedures and different tissues.

[0221] In one embodiment, the device further includes a second acquisition module; the second acquisition module is used to acquire the first corresponding relationship; the second acquisition module includes:

[0222] The identification submodule is used to perform tissue identification on the multiple medical images respectively to obtain tissue identification results of each medical image; the tissue identification results include the tissue identified from the corresponding medical image;

[0223] A third determination submodule is used to determine the tissue corresponding to each medical image based on the identification results of each tissue and the segmentation conditions;

[0224] The third acquisition submodule is used to establish a correspondence between the modality of the medical image and the corresponding tissue to obtain a first correspondence.

[0225] In one embodiment, the device further comprises:

[0226] A first display module, used for displaying at least one group of segmentation options before the segmentation module 908 performs segmentation processing on the tissue to be segmented in the target medical image and obtains the tissue segmentation result of the target medical image; each group of segmentation options includes the tissue to be segmented and the corresponding target medical image;

[0227] A third acquisition module is used to acquire a target segmentation option selected by a user from at least one group of segmentation options;

[0228] Correspondingly, the segmentation module 908 is specifically configured to segment the tissue to be segmented in the target medical image corresponding to the target segmentation option to obtain a tissue segmentation result of the target medical image.

[0229] In one embodiment, if the tissue to be segmented displayed by the target segmentation option corresponds to multiple target medical images, the segmentation module 908 includes:

[0230] A fourth acquisition submodule, used to acquire at least one selected target medical image selected by a user from a plurality of target medical images;

[0231] The segmentation submodule is used to segment the tissue to be segmented in each selected target medical image corresponding to the target segmentation option, and obtain the tissue segmentation result of each selected target medical image.

[0232] In one of the embodiments, the segmentation submodule is further used to determine a default target medical image from multiple target medical images when no user selection operation on multiple target medical images corresponding to the target segmentation option is detected, and to segment the tissue to be segmented in the default target medical image to obtain a tissue segmentation result of the default target medical image.

[0233] In one of the embodiments, a plurality of target medical images are sorted in descending order according to the segmentation effect, and the segmentation submodule is specifically configured to use the first target medical image in the descending order as a default target medical image.

[0234] In one embodiment, the first target medical image is displayed in the target segmentation option, and other target medical images except the first target medical image among the multiple target medical images are hidden; the target segmentation option also includes a drop-down component, and the device also includes:

[0235] The second display module is used to display other hidden target medical images in response to a user triggering operation on the pull-down component.

[0236] In one embodiment, the device further comprises:

[0237] The third display module is used to display a segmentation response corresponding to the target segmentation option at a preset position on the target segmentation option; the segmentation response includes a segmentation success or a segmentation failure.

[0238] In one embodiment, the device further comprises:

[0239] The fourth display module is used to fuse and display the tissue segmentation results of each target medical image.

[0240] In one embodiment, the device further comprises:

[0241] a third determination module, configured to determine, before the segmentation module 908 performs segmentation processing on the tissue to be segmented in the target medical image and obtains the tissue segmentation result of the target medical image, a target initialization operation corresponding to the target surgical procedure according to a preset fourth corresponding relationship; the fourth corresponding relationship includes a corresponding relationship between different surgical procedures and different initialization operations;

[0242] The execution module is used to execute the target initialization operation.

[0243] Each module in the above-mentioned tissue segmentation device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0244] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Fig.10As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer 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 input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a tissue segmentation method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0245] Those skilled in the art will understand that Fig.10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0246] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the tissue segmentation method in any of the above embodiments are implemented.

[0247] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the tissue segmentation method in any of the above embodiments are implemented.

[0248] In one embodiment, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the steps of the tissue segmentation method in any of the above embodiments are implemented.

[0249] 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 used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0250] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0251] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0252] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A tissue segmentation method, It is characterized in that The method comprises: Acquire multiple medical images of the object to be tested, and acquire at least one tissue to be segmented of the object to be tested; wherein at least some of the multiple medical images have different modalities; For each of the tissues to be segmented, the modality corresponding to the tissue to be segmented is determined according to a preset first correspondence relationship; the first correspondence relationship includes a correspondence relationship between different tissues and different modalities, and the segmentation effect of the tissue in the medical image of the corresponding modality meets the segmentation condition; Determining a target medical image corresponding to the tissue to be segmented from the multiple medical images according to the modality corresponding to the tissue to be segmented; The tissue to be segmented in the target medical image is segmented to obtain a tissue segmentation result of the target medical image.

2. The method according to claim 1, It is characterized in that The step of acquiring a plurality of medical images of the object to be tested comprises: Obtaining a target formula of the object to be tested selected by a user from at least one candidate formula; Determine multiple candidate modalities corresponding to the target procedure according to a preset second correspondence relationship; the second correspondence relationship includes a correspondence relationship between different procedures and different modalities; Based on the multiple candidate modalities, a medical image corresponding to each of the candidate modalities of the object to be tested is obtained.

3. The method according to claim 2, It is characterized in that The step of acquiring a medical image corresponding to each of the candidate modalities of the object to be tested based on the multiple candidate modalities includes: Based on the multiple candidate modalities, generate and display an image input component corresponding to each of the candidate modalities; For each of the image input components, in response to an image input operation of the user on the image input component, a medical image of the candidate modality corresponding to the image input component of the object to be tested is obtained.

4. The method according to claim 3, It is characterized in that The step of obtaining the candidate modality medical image corresponding to the image input component of the object to be tested in response to the user's image input operation on the image input component comprises: In response to an image input operation of the user on the image input component, acquiring a candidate medical image input by the user; In the case where the modality of the candidate medical image is consistent with the candidate modality corresponding to the image input component, the candidate medical image is imported as the medical image of the candidate modality corresponding to the image input component of the object to be tested.

5. The method according to claim 2, It is characterized in that The step of obtaining at least one tissue to be segmented of the object to be detected comprises: According to a preset third correspondence, at least one to-be-segmented tissue corresponding to the target procedure is determined; the third correspondence includes a correspondence between different procedures and different tissues.

6. The method according to any one of claims 1 to 5, It is characterized in that Before segmenting the tissue to be segmented in the target medical image to obtain the tissue segmentation result of the target medical image, the method further includes: displaying at least one group of segmentation options; each group of segmentation options includes the tissue to be segmented and a corresponding target medical image; Obtaining a target segmentation option selected by a user from the at least one group of segmentation options; Accordingly, the segmenting of the tissue to be segmented in the target medical image to obtain a tissue segmentation result of the target medical image includes: The tissue to be segmented in the target medical image corresponding to the target segmentation option is segmented to obtain a tissue segmentation result of the target medical image.

7. The method according to claim 6, It is characterized in that If the tissue to be segmented displayed by the target segmentation option corresponds to a plurality of target medical images, segmenting the tissue to be segmented in the target medical image corresponding to the target segmentation option to obtain a tissue segmentation result of the target medical image includes: Acquire at least one selected target medical image selected by a user from the plurality of target medical images; The tissue to be segmented in each of the selected target medical images corresponding to the target segmentation option is segmented to obtain a tissue segmentation result of each of the selected target medical images.

8. The method according to claim 7, It is characterized in that The step of segmenting the tissue to be segmented in the target medical image corresponding to the target segmentation option to obtain a tissue segmentation result of the target medical image further includes: If no user selection operation on the multiple target medical images corresponding to the target segmentation option is detected, a default target medical image is determined from the multiple target medical images, and the tissue to be segmented in the default target medical image is segmented to obtain a tissue segmentation result of the default target medical image.

9. The method according to claim 6, It is characterized in that The method further comprises: A segmentation response corresponding to the target segmentation option is displayed at a preset position on the target segmentation option; the segmentation response includes segmentation success or segmentation failure.

10. The method according to claim 2, It is characterized in that Before segmenting the tissue to be segmented in the target medical image to obtain the tissue segmentation result of the target medical image, the method further includes: Determine the target initialization operation corresponding to the target technique according to a preset fourth corresponding relationship; the fourth corresponding relationship includes a corresponding relationship between different techniques and different initialization operations; The target initialization operation is performed.

11. A tissue segmentation device, It is characterized in that The device comprises: A first acquisition module is used to acquire multiple medical images of the object to be tested, and to acquire at least one tissue to be segmented of the object to be tested; wherein at least some of the multiple medical images have different modalities; A first determination module is used to determine, for each of the tissues to be segmented, a modality corresponding to the tissue to be segmented according to a preset first correspondence relationship; the first correspondence relationship includes a correspondence relationship between different tissues and different modalities, and the segmentation effect of the tissue in the medical image of the corresponding modality meets the segmentation condition; A second determination module, configured to determine, according to the modality corresponding to the tissue to be segmented, a target medical image corresponding to the tissue to be segmented from among the multiple medical images; The segmentation module is used to perform segmentation processing on the tissue to be segmented in the target medical image to obtain a tissue segmentation result of the target medical image.

12. A computer device comprising a memory and a processor, wherein the memory stores a computer program. It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 10 are implemented.

13. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images