Classification method and classification device for prostate remarkable cancer

Abstract

The invention discloses a classification method and a classification device for prostate significant cancer. The classification method comprises the following steps: acquiring a plurality of magnetic resonance images, and selecting a T2 sequence and a plurality of diffusion weighted image DWI sequences with different b values from the magnetic resonance images; processing the plurality of DWI sequences to obtain at least one apparent diffusion coefficient ADC sequence; extracting one sequence from the T2 sequence, the plurality of DWI sequences and the at least one ADC sequence as a basic segmentation sequence, and segmenting each image in the basic segmentation sequence by using a pre-trained deep learning model to segment prostate glands and prostate lesions; based on the segmented basic segmentation sequence, segmenting each image in other sequences; training a classification model by using all the segmented sequences to obtain a trained classification model; and inputting magnetic resonance imaging to be classified into the trained classification model for classification. The method can be used for accurately and efficiently detecting the prostate remarkable cancer.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular to a classification method, classification device, medium, electronic device, and computer program product for classifying significant prostate cancer. Background Art

[0002] Prostate cancer is one of the most common malignancies in men worldwide, and has a significant impact on men's quality of life and life expectancy. Currently, multiparametric magnetic resonance imaging (mpMRI) combined with the prostate imaging reporting and data system (PI-RADS) is the main imaging method for detecting clinically significant prostate cancer (csPCa). However, these methods have problems of high variability and low specificity, especially when using the PI-RADS scoring system, the diagnostic consistency between different doctors is poor, and PI-RADS has low specificity in detecting csPCa. In addition, the interpretation process of mpMRI is time-consuming, which increases the workload of doctors. Summary of the invention

[0003] Embodiments of the present application provide a method for classifying significant prostate cancer, a device for classifying significant prostate cancer, a medium, an electronic device, and a computer program product.

[0004] In a first aspect, an embodiment of the present application provides a classification method for prostate significant cancer, which is used in an electronic device, and the classification method includes:

[0005] An acquisition step of acquiring a plurality of magnetic resonance images, and selecting a T2 sequence and a plurality of diffusion weighted image DWI sequences with different b values ​​from each of the magnetic resonance images;

[0006] A processing step of processing the multiple DWI sequences to obtain at least one apparent diffusion coefficient ADC sequence;

[0007] A first segmentation step is to select a sequence from the T2 sequence, the multiple DWI sequences, and the at least one ADC sequence as a basic segmentation sequence, and use a pre-trained deep learning model to segment each image in the basic segmentation sequence to segment the prostate gland and the prostate lesion;

[0008] A second segmentation step, based on the segmented basic segmentation sequence, segmenting each image in other sequences to segment the prostate gland and the prostate lesion;

[0009] The training step uses all the segmented sequences to train the classification model to obtain the trained classification model;

[0010] The classification step is to input the magnetic resonance imaging to be classified into the trained classification model for classification.

[0011] In a second aspect, the present invention provides a classification device for prostate significant cancer, the classification device comprising:

[0012] An acquisition unit, which acquires magnetic resonance imaging of each of a plurality of patients, and selects a T2 sequence and a plurality of diffusion weighted image DWI sequences with different b values ​​from each of the magnetic resonance imaging;

[0013] A processing unit processes the multiple DWI sequences to obtain at least one apparent diffusion coefficient ADC sequence;

[0014] A first segmentation unit is configured to take a sequence from the T2 sequence, the multiple DWI sequences, and the at least one ADC sequence as a basic segmentation sequence, and to segment each image in the basic segmentation sequence to segment the prostate gland and the prostate lesion;

[0015] A second segmentation unit, based on the segmented basic segmentation sequence, segments each image in other sequences to segment the prostate gland and the prostate lesion;

[0016] A training unit uses all the segmented sequences to train the classification model to obtain a trained classification model;

[0017] The classification unit inputs the magnetic resonance imaging to be classified into the trained classification model for classification.

[0018] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having instructions stored thereon, and when the instructions are executed on a computer, the computer executes the method for classifying significant prostate cancer as described in any one of the first aspects.

[0019] In a fourth aspect, an embodiment of the present invention provides an electronic device, comprising: one or more processors; one or more memories; the one or more memories storing one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device executes the classification method for significant prostate cancer described in the first aspect.

[0020] In a fifth aspect, an embodiment of the present application provides a computer program product, comprising computer executable instructions, wherein the instructions are executed by a processor to implement the method for classifying significant prostate cancer described in the first aspect.

[0021] In the present invention, a classification model is constructed (trained) based on two parameters (ie, T2 sequence and DWI sequence) in multi-parameter magnetic resonance imaging, that is, a classification model is constructed based on dual parameters, so the process of constructing the classification model is simpler and less costly.

[0022] In the present invention, based on multi-parameter magnetic resonance imaging, the classification model constructed by the present invention can accurately and efficiently detect significant prostate cancer and reduce the workload of doctors.

[0023] In the present invention, in the process of constructing the classification model, T2 weighted image (T2WI) sequence and T2 fat suppression weighted (FS-T2WI) image can be used, so that the classification model is applicable to various multi-parameter magnetic resonance imaging, and the application is more flexible and extensive. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 According to an embodiment of the present application, a flow chart of a method for classifying significant prostate cancer is shown;

[0025] Figure 2 According to an embodiment of the present application, a structural diagram of a classification device for prostate significant cancer is shown;

[0026] Figure 3 According to an embodiment of the present application, a block diagram of an electronic device is shown. DETAILED DESCRIPTION

[0027] The illustrative embodiments of the present application include, but are not limited to, a method for classifying significant prostate cancer, an apparatus, a medium, an electronic device, and a computer program product for classifying significant prostate cancer.

[0028] The embodiments of the present application will be described in further detail below in conjunction with the accompanying drawings.

[0029] Figure 1 The classification method of significant prostate cancer in an embodiment of the present application is shown, which is used in an electronic device. Specifically, in the acquisition step S11, multiple magnetic resonance images are acquired, and a T2 sequence and multiple diffusion weighted image DWI sequences with different b values ​​are selected from each of the magnetic resonance images.

[0030] Magnetic resonance imaging of the prostate of multiple patients is acquired, and magnetic resonance imaging is also called multi-parametric (sequence) magnetic resonance imaging, which includes multiple sequences, such as T2 sequence, diffusion weighted image (DWI) sequence, proton density weighted imaging (PDWI) sequence, perfusion weighted imaging (PWI), etc. The sequence may also be called parameter.

[0031] In this embodiment, a T2 sequence and multiple DWI sequences are selected from multiple sequences of each magnetic resonance imaging, for example, three DWI sequences are selected, including a DWI1000 sequence, a DWI2000 sequence, and a DWI 3000 sequence, wherein the numbers 1000, 2000, and 3000 represent the b-values ​​of DWI. It is understandable that the present invention can randomly select DWI sequences with different b-values.

[0032] In the processing step S12, a plurality of DWI sequences are processed to obtain at least one apparent diffusion coefficient ADC sequence.

[0033] Specifically, any two DWI sequences with different b values ​​are processed to obtain corresponding ADC sequences, thereby obtaining at least one ADC sequence.

[0034] In this embodiment, the above DWI1000 sequence, DWI2000 sequence, and DWI3000 sequence are subtracted to obtain corresponding ADC sequences. For example, DWI2000-DWI1000=ADC1, DWI3000-DWI1000=ADC2, and DWI3000-DWI2000=ADC3. In this way, three ADC sequences can be obtained.

[0035] It can be understood that the ADC sequence is generated by performing mathematical operations and processing on the image data at different b values ​​obtained by DWI scanning. In DWI scanning, the signal intensity of the tissue is related to the diffusion movement of water molecules and the b value. By measuring the change in signal intensity at different b values, the ADC value of the tissue can be calculated and presented in the form of an image, namely an ADC sequence.

[0036] It can be understood that each of the above sequences T2, DWI1000, DWI2000, DWI 3000, ADC1, ADC2, and ADC3 includes multiple images.

[0037] In the first segmentation step S13, a sequence is taken from the T2 sequence, multiple DWI sequences, and at least one ADC sequence as a basic segmentation sequence, and each image in the basic segmentation sequence is segmented using a pre-trained deep learning model to segment the prostate gland and prostate lesions. The T2 sequence includes a T2 weighted image (T2WI) sequence and a T2 fat-suppressed weighted (FS-T2WI) image.

[0038] The deep learning model is, for example, a U-Net network. The following describes a process of pre-training the U-Net network.

[0039] First, multiple sample MRIs of multiple patients were obtained, and training data sets and validation data sets were formed from the multiple sample MRIs. Professional doctors performed delineation in each sample MRI in the training set to distinguish the glandular area and the lesion area of ​​the prostate in the MRI. The lesion area was delineated on the T2-weighted image (T2WI), diffusion-weighted image (DWI), and apparent diffusion coefficient (ADC) map. During the delineation process, multiple sequences were referenced and combined with radical surgical pathology or biopsy results. The delineation of the lesion was performed on the training data set and the validation data set.

[0040] In order to simplify the field of view and improve the computational efficiency of deep learning, the prostate region was centrally cropped to be suitable for prostate gland segmentation and prostate lesion segmentation tasks. The lesions themselves were also centrally cropped to be suitable for the subsequent classification module. Finally, normalization was performed: for each T2WI and DWI, all voxel values ​​were standardized by subtracting the mean and dividing by the standard deviation. In the case of ADC, an additional step of dividing the voxel values ​​by 1000 was performed to ensure that the clinical significance of the values ​​was preserved.

[0041] In order to enhance the compatibility of sample data, the training dataset contains both T2WI and FS-T2WI. The U-Net network is trained using the training dataset. The U-Net network has a symmetrical U-shaped structure, including an encoding network on the left and a decoding network on the right. The encoding network is responsible for extracting image features and context information, while the decoding network restores image resolution and positioning details. The two networks are connected by jump connections to promote feature fusion. The loss function uses Dice loss and cross entropy loss, which are defined as follows:

[0042] Loss = W 1 Ldice+W 2 Lc

[0043] Among them, Ldice is Dice loss, Lc is cross entropy loss, W 1 and W 2 is the weight coefficient. This model can also be implemented, for example, using nnU-Net, an automated and flexible deep learning framework for medical image segmentation.

[0044] In this embodiment, for example, the T2 sequence is used as the basic segmentation sequence, and a pre-trained U-Net model is used to segment each image in the T2 sequence to segment the prostate gland and the prostate lesion.

[0045] In the second segmentation step S14, based on the segmented basic segmentation sequence, each image in other sequences is segmented to segment the prostate gland and the prostate lesion.

[0046] Specifically, each image in the segmented T2 sequence is aligned with each image in the other sequences so as to segment each image in the other sequences.

[0047] In this embodiment, for example, the images in the T2 sequence are aligned with the images of each sequence of DWI1000, DWI2000, DWI 3000, ADC1, ADC2, and ADC3 using the world coordinates embedded in the DICOM file, so that each of the other sequences is segmented to obtain all the segmented sequences. That is, the prostate glands and prostate lesions in all images in all (for example, 7 in this embodiment) sequences are segmented.

[0048] It can be understood that for each patient's magnetic resonance imaging, all the segmented sequences can be obtained.

[0049] In the training step S15, all the segmented sequences are used to train the classification model to obtain a trained classification model.

[0050] In this embodiment, the classification model is, for example, a deep learning network, and the 7 segmented sequences of each patient are input into the deep learning network for training to obtain a trained classification model.

[0051] For each image in all the sequences after segmentation, a lesion mask is obtained. For the lesion mask, a center crop is performed, that is, the lesion itself is center cropped.

[0052] The lesion mask of each image is also used to train the classification model, that is, the lesion mask and the 7 sequences after segmentation are used as sample data and input into the deep learning network for training.

[0053] In the classification step S16, the magnetic resonance imaging to be classified is input into the trained classification model for classification. The magnetic resonance imaging to be classified includes prostate lesions, and the magnetic resonance imaging to be classified is input into the trained classification model to classify the prostate significant cancer in the prostate lesions.

[0054] It can be understood that prostate lesions include prostatitis, prostate hyperplasia, and prostate cancer. Through the classification model trained as described above, it is possible to determine whether the prostate lesion is significant prostate cancer (i.e., significant prostate cancer) based on the magnetic resonance imaging to be classified.

[0055] In the present invention, a classification model is constructed (trained) based on two parameters (ie, T2 sequence and DWI sequence) in multi-parameter magnetic resonance imaging, that is, a classification model is constructed based on dual parameters, so the process of constructing the classification model is simpler and less costly.

[0056] In the present invention, based on multi-parameter magnetic resonance imaging, the classification model constructed by the present invention can accurately and efficiently detect significant prostate cancer and reduce the workload of doctors.

[0057] In the present invention, in the process of constructing the classification model, T2 weighted image (T2WI) sequence and T2 fat suppression weighted (FS-T2WI) image can be used, so that the classification model is applicable to various multi-parameter magnetic resonance imaging, and the application is more flexible and extensive.

[0058] The present invention also provides a classification device for prostate significant cancer, characterized in that the classification device comprises:

[0059] An acquisition unit 201 acquires magnetic resonance images of multiple patients, and selects a T2 sequence and a plurality of diffusion weighted image DWI sequences with different b values ​​from each magnetic resonance image;

[0060] A processing unit 202 processes the multiple DWI sequences to obtain at least one apparent diffusion coefficient ADC sequence;

[0061] A first segmentation unit 203 is configured to take out a sequence from the T2 sequence, the multiple DWI sequences, and the at least one ADC sequence as a basic segmentation sequence, and to segment each image in the basic segmentation sequence to segment the prostate gland and the prostate lesion;

[0062] A second segmentation unit 204 segments each image in other sequences based on the segmented basic segmentation sequence to segment the prostate gland and the prostate lesion;

[0063] A training unit 205 uses all the segmented sequences to train a classification model to obtain a trained classification model;

[0064] The classification unit 206 inputs the magnetic resonance imaging to be classified into the trained classification model for classification.

[0065] It can be understood that the acquisition unit 201, the processing unit 202, the first segmentation unit 203, the second segmentation unit, the training unit 205, and the classification unit 206 can be Figure 3 The processor 102 in the electronic device 100 having the functions of these modules or units is implemented.

[0066] The present invention also provides a computer-readable storage medium, wherein the storage medium stores instructions, which, when executed on a computer, cause the computer to execute Figure 1 The method shown in .

[0067] The present invention also provides a computer program product, including computer executable instructions, which are executed by the processor 102 to implement Figure 1 The method shown in .

[0068] Reference now Figure 3 , Figure 3An example electronic device 1400 according to an embodiment of the present invention is schematically shown. In one embodiment, the electronic device 1400 may include one or more processors 1404, a system control logic unit 1408 connected to at least one of the processors 1404, a system memory 1412 connected to the system control logic unit 1408, a non-volatile memory (NVM) 1416 connected to the system control logic unit 1408, and a network interface 1420 connected to the system control logic unit 1408.

[0069] In some embodiments, the processor 1404 may include one or more single-core or multi-core processors. In some embodiments, the processor 1404 may include any combination of general-purpose processors and special-purpose processors (e.g., graphics processors, application processors, baseband processors, etc.). In embodiments where the electronic device 1400 uses an eNB (Evolved Node B, enhanced base station) or RAN (Radio Access Network, radio access network) controller, the processor 1404 may be configured to execute various embodiments, such as Figure 1 The embodiment shown.

[0070] In some embodiments, system control logic unit 1408 may include any suitable interface controller to provide any suitable interface to at least one of processors 1404 and / or any suitable device or component in communication with system control logic unit 1408 .

[0071] In some embodiments, the system control logic unit 1408 may include one or more memory controllers to provide an interface to the system memory 1412. The system memory 1412 may be used to load and store data and / or instructions. In some embodiments, the system memory 1412 of the electronic device 1400 may include any suitable volatile memory, such as a suitable dynamic random access memory (DRAM).

[0072] The non-volatile memory 1416 may include one or more tangible, non-transitory computer-readable media for storing data and / or instructions. In some embodiments, the non-volatile memory 1416 may include any suitable non-volatile memory such as flash memory and / or any suitable non-volatile storage device, such as at least one of a HDD (Hard Disk Drive), a CD (Compact Disc) drive, and a DVD (Digital Versatile Disc) drive.

[0073] The non-volatile memory 1416 may include a portion of storage resources on a device on which the electronic device 1400 is installed, or it may be accessible to the electronic device but is not necessarily a portion of the electronic device. For example, the non-volatile memory 1416 may be accessed over a network via the network interface 1420 .

[0074] In particular, the system memory 1412 and the non-volatile memory 1416 may include a temporary copy and a permanent copy of the instructions 1424, respectively. The instructions 1424 may include instructions that, when executed by at least one of the processors 1404, cause the electronic device 1400 to implement the following: Figure 1 In some embodiments, instructions 1424, hardware, firmware, and / or software components thereof may additionally / alternatively be located in system control logic unit 1408, network interface 1420, and / or processor 1404.

[0075] The network interface 1420 may include a transceiver for providing a radio interface for the electronic device 1400, thereby communicating with any other suitable device (such as a front-end module, an antenna, etc.) through one or more networks. In some embodiments, the network interface 1420 may be integrated with other components of the electronic device 1400. For example, the network interface 1420 may be integrated with at least one of the processor 1404, the system memory 1412, the non-volatile memory 1416, and a firmware device (not shown) having instructions. When at least one of the processors 1404 executes the instructions, the electronic device 1400 implements the following. Figure 1 The method shown.

[0076] The network interface 1420 may further include any suitable hardware and / or firmware to provide a multiple-input multiple-output radio interface. For example, the network interface 1420 may be a network adapter, a wireless network adapter, a telephone modem and / or a wireless modem.

[0077] In one embodiment, at least one of the processors 1404 may be packaged together with logic for one or more controllers of the system control logic unit 1408 to form a system in package (SiP). In one embodiment, at least one of the processors 1404 may be integrated on the same die with logic for one or more controllers of the system control logic unit 1408 to form a system on chip (SoC).

[0078] The electronic device 1400 may further include an input / output (I / O) device 1432. The I / O device 1432 may include a user interface to enable a user to interact with the electronic device 1400; the design of the peripheral component interface enables the peripheral components to interact with the electronic device 1400. In some embodiments, the electronic device 1400 further includes a sensor for determining at least one of an environmental condition and location information related to the electronic device 1400.

[0079] In some embodiments, the user interface may include, but is not limited to, a display (e.g., an LCD display, a touch screen display, etc.), a speaker, a microphone, one or more cameras (e.g., a still image camera and / or a video camera), a flashlight (e.g., an LED flash), and a keyboard.

[0080] The various embodiments of the mechanism disclosed in the present application can be implemented in hardware, software, firmware or a combination of these implementation methods. The embodiments of the present application can be implemented as a computer program or program code executed on a programmable system, which includes at least one processor, a storage system (including volatile and non-volatile memory and / or storage elements), at least one input device and at least one output device.

[0081] Program code can be applied to input instructions to perform the functions described in this application and generate output information. The output information can be applied to one or more output devices in a known manner. For the purposes of this application, a processing system includes any system having a processor such as, for example, a digital signal processor (DSP), a microcontroller, an application specific integrated circuit (ASIC), or a microprocessor.

[0082] Program code can be implemented with high-level programming language or object-oriented programming language to communicate with the processing system. When necessary, program code can also be implemented with assembly language or machine language. In fact, the mechanism described in this application is not limited to the scope of any specific programming language. In either case, the language can be a compiled language or an interpreted language.

[0083] In some cases, the disclosed embodiments may be implemented in hardware, firmware, software, or any combination thereof. The disclosed embodiments may also be implemented as instructions carried or stored on one or more temporary or non-temporary machine-readable (e.g., computer-readable) storage media, which may be read and executed by one or more processors. For example, instructions may be distributed over a network or through other computer-readable media. Therefore, a machine-readable medium may include any mechanism for storing or transmitting information in a machine (e.g., computer) readable form, including, but not limited to, a floppy disk, an optical disk, an optical disk, a read-only memory (CD-ROM), a magneto-optical disk, a read-only memory (ROM), a random access memory (RAM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic card or an optical card, a flash memory, or a tangible machine-readable memory for transmitting information (e.g., carrier wave, infrared signal digital signal, etc.) using the Internet in an electrical, optical, acoustic or other form of propagation signal. Therefore, a machine-readable medium includes any type of machine-readable medium suitable for storing or transmitting electronic instructions or information in a machine (e.g., computer) readable form.

[0084] In the accompanying drawings, some structural or method features may be shown in a specific arrangement and / or order. However, it should be understood that such a specific arrangement and / or order may not be required. Instead, in some embodiments, these features may be arranged in a manner and / or order different from that shown in the illustrative drawings. In addition, the inclusion of structural or method features in a particular figure does not mean that such features are required in all embodiments, and in some embodiments, these features may not be included or may be combined with other features.

[0085] It should be noted that the units / modules mentioned in the various device embodiments of the present application are all logical units / modules. Physically, a logical unit / module can be a physical unit / module, or a part of a physical unit / module, or can be implemented as a combination of multiple physical units / modules. The physical implementation method of these logical units / modules themselves is not the most important. The combination of functions implemented by these logical units / modules is the key to solving the technical problems proposed by the present application. In addition, in order to highlight the innovative part of the present application, the above-mentioned device embodiments of the present application do not introduce units / modules that are not closely related to solving the technical problems proposed by the present application, which does not mean that there are no other units / modules in the above-mentioned device embodiments.

[0086] It should be noted that in the examples and description of this patent, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including one" do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0087] Although the present application has been illustrated and described with reference to certain preferred embodiments thereof, it will be apparent to those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present application.

Claims

1. A method for classifying significant prostate cancer, for use in electronic equipment, characterized in that: The classification method includes: An acquisition step of acquiring a plurality of magnetic resonance images, and selecting a T2 sequence and a plurality of diffusion weighted image DWI sequences with different b values ​​from each of the magnetic resonance images; A processing step of processing the multiple DWI sequences to obtain at least one apparent diffusion coefficient ADC sequence; A first segmentation step is to select a sequence from the T2 sequence, the multiple DWI sequences, and the at least one ADC sequence as a basic segmentation sequence, and use a pre-trained deep learning model to segment each image in the basic segmentation sequence to segment the prostate gland and the prostate lesion; A second segmentation step, based on the segmented basic segmentation sequence, segmenting each image in other sequences to segment the prostate gland and the prostate lesion; The training step uses all the segmented sequences to train the classification model to obtain the trained classification model; The classification step is to input the magnetic resonance imaging to be classified into the trained classification model for classification.

2. The classification method according to claim 1, characterized in that: In the processing step, any two DWI sequences with different b values ​​are processed to obtain corresponding ADC sequences, thereby obtaining the at least one ADC sequence.

3. The classification method according to claim 1, characterized in that: The training data set used to train the deep learning model includes both the T2-weighted image sequence and the T2 fat-suppression weighted image in the T2 sequence.

4. The classification method according to claim 1, characterized in that: In the second segmentation step, each image in the segmented basic segmentation sequence is aligned with each image in other sequences so as to segment each image in other sequences.

5. The classification method according to claim 1, characterized in that: For each image in all sequences after segmentation, obtain the lesion mask, Wherein, in the training step, the lesion mask of each image is also used to train the classification model.

6. The classification method according to claim 1, characterized in that: In the classification step, the magnetic resonance imaging to be classified includes the prostate lesion, and the magnetic resonance imaging to be classified is input into the trained classification model, so as to classify the significant prostate cancer in the prostate lesion.

7. A classification device for prostate cancer, characterized in that: The classification device comprises: An acquisition unit, which acquires magnetic resonance imaging of each of a plurality of patients, and selects a T2 sequence and a plurality of diffusion weighted image DWI sequences with different b values ​​from each of the magnetic resonance imaging; A processing unit processes the multiple DWI sequences to obtain at least one apparent diffusion coefficient ADC sequence; A first segmentation unit is configured to take a sequence from the T2 sequence, the multiple DWI sequences, and the at least one ADC sequence as a basic segmentation sequence, and to segment each image in the basic segmentation sequence to segment the prostate gland and the prostate lesion; A second segmentation unit, based on the segmented basic segmentation sequence, segments each image in other sequences to segment the prostate gland and the prostate lesion; A training unit uses all the segmented sequences to train the classification model to obtain a trained classification model; The classification unit inputs the magnetic resonance imaging to be classified into the trained classification model for classification.

8. A computer-readable storage medium, characterized in that: The storage medium stores instructions, which, when executed on a computer, cause the computer to execute the method for classifying significant prostate cancer according to any one of claims 1 to 6.

9. An electronic device, characterized in that: include: one or more processors; One or more memories; the one or more memories store one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device executes the classification method for significant prostate cancer described in any one of claims 1 to 6.

10. A computer program product comprising computer executable instructions, characterized in that: The instructions are executed by a processor to implement the method for classifying significant prostate cancer according to any one of claims 1 to 6.

Images