Magnetic resonance imagery classification method and apparatus based on quantum convolutional neural network

Abstract

The present application relates to the technical field of image classification, and provides a magnetic resonance imagery classification method and apparatus based on a quantum convolutional neural network. The method comprises: acquiring an MRI image, and, on the basis of the MRI image, a convolutional layer and a pooling layer, acquiring corresponding MRI image feature data; inputting the MRI image feature data into a fully connected layer, to acquire corresponding MRI image category data; and inputting the MRI image category data into a quantum layer, to convert same into quantum state data, and performing image classification on the basis of the quantum state data by means of the quantum layer. The present application solves the problem in the related art of low image classification efficiency in magnetic resonance imaging.

Description

Magnetic resonance image classification method and device based on quantum convolutional neural network Technical Field

[0001] The present application relates to the field of image classification technology. Specifically, the present application relates to a magnetic resonance image classification method and device based on quantum convolutional neural network. Background Art

[0002] Magnetic resonance imaging (MRI) is a widely used method for obtaining high-quality medical images. It is a popular, painless, and noninvasive brain imaging technique. MRI offers a unique view of the brain, providing high levels of spatial and contrast resolution. It is used to diagnose a variety of diseases and provides high-quality information about the brain's internal structures. However, the sheer volume of data extracted from these images makes it difficult to draw a conclusive diagnosis based on the raw data. In such cases, various image analysis tools are needed to analyze MRI images and extract conclusive information to classify normal and abnormal brain conditions.

[0003] Numerous medical image analysis methods based on deep learning have been proposed for health monitoring and diagnosis using brain magnetic resonance imaging (MRI). Convolutional neural networks, a branch of deep learning, are commonly used to analyze visual information. Convolutional neural networks consist of many trainable layers, including input, convolutional, convolutional, and fully connected layers. These layers can be fine-tuned using hyperparameters. Different convolutional neural network architectures, combined with various transfer learning techniques, have achieved significant success in improving image classification performance. In this way, they have surpassed traditional machine learning models in recent years.

[0004] Although numerous brain MRI classification algorithms have been developed, existing methods still suffer from numerous shortcomings. Since many algorithms rely on human-selected lesion locations, they cannot be fully automated. Furthermore, convolutional neural networks rely on classical computing principles and are limited by classical computing hardware. Consequently, they can face challenges in effectively processing and representing high-dimensional image data, a limitation that can limit their ability to effectively learn and model complex image classification tasks. Convolutional neural networks are also prone to overfitting, and on smaller datasets, convolutional neural network models often struggle to perform well.

[0005] From the above, we can see that how to improve the low efficiency of image classification in magnetic resonance imaging still needs to be solved.

[0006] Summary of the Invention

[0007] This application provides a method, device, electronic device, and storage medium for magnetic resonance imaging classification based on quantum convolutional neural networks, which can solve the problem of low image classification efficiency in magnetic resonance imaging in related technologies. The technical solution is as follows:

[0008] According to one aspect of the present application, a magnetic resonance image classification method based on a quantum convolutional neural network includes: acquiring an MRI image, acquiring corresponding MRI image feature data based on the MRI image, a convolutional layer, and a pooling layer; inputting the MRI image feature data into a fully connected layer to acquire corresponding MRI image category data; inputting the MRI image category data into a quantum layer to convert it into quantum state data, and performing image classification based on the quantum state data through the quantum layer.

[0009] According to one aspect of the present application, a magnetic resonance image classification device based on a quantum convolutional neural network comprises:

[0010] A feature data acquisition module acquires an MRI image and is used to acquire corresponding MRI image feature data based on the MRI image, the convolution layer, and the pooling layer;

[0011] a category data acquisition module, inputting the MRI image feature data into a fully connected layer to obtain corresponding MRI image category data;

[0012] The classification module inputs the MRI image category data into the quantum layer to convert it into quantum state data, and performs image classification through the quantum layer based on the quantum state data.

[0013] In an exemplary embodiment, the device includes, but is not limited to:

[0014] a conversion module, based on inputting the MRI image category data into the ZZFeatureMap, wherein the ZZFeatureMap is used to convert the MRI image category data into quantum state data;

[0015] The quantum state data is input into the Ansatz layer for transformation and measurement.

[0016] In an exemplary embodiment, the device includes, but is not limited to:

[0017] The expression of the ZZFeatureMap is: U(θ)=∏ k U k (θ k ), where θ k =x i (2π-x j ) For i<j, U k (θ k) is a two-qubit ZZ gate.

[0018] In an exemplary embodiment, the device includes, but is not limited to:

[0019] The Ansatz layer uses a RealAmplitudes circuit, and the RealAmplitudes circuit is expressed as follows: Among them, Ry(φ i ) is a single-qubit Y rotation with an angle of φ i , CNOT is the entanglement gate.

[0020] In an exemplary embodiment, the device includes, but is not limited to:

[0021] Two of the convolutional layers, two of the pooling layers, two of the fully connected layers and one of the quantum layers, two of the convolutional layers, two of the pooling layers and two of the fully connected layers constitute a regular layer, and the regular layer is connected to the quantum layer. The internal arrangement of the regular layer is: convolutional layer, pooling layer, convolutional layer, pooling layer, fully connected layer and fully connected layer, and the second fully connected layer outputs MRI image category data.

[0022] In an exemplary embodiment, the device includes, but is not limited to:

[0023] Mapping module, used to map class labels {0, 1} to {1, -1} respectively;

[0024] The mean square error acquisition module is used to obtain the mean square error (MSE) of the prediction and class label. in, is the Pauli-Z expectation value of a qubit state extracted from the quantum convolutional neural network for the i-th training data, is the label of the corresponding training data;

[0025] State acquisition module, used to obtain the mixed state a i |0><0|+b i |1><1|, if the i-th training data is marked as 0, then the cost function of θ will correspond to a i Greater than b i , if it is marked as 1, then the cost function of θ will correspond to a i Less than or equal to b i .

[0026] According to one aspect of the present application, an electronic device includes at least one processor and at least one memory, wherein the memory stores computer-readable instructions; the computer-readable instructions are executed by one or more of the processors, so that the electronic device implements the magnetic resonance image classification method based on quantum convolutional neural network as described above.

[0027] According to one aspect of the present application, a storage medium stores computer-readable instructions thereon, and the computer-readable instructions are executed by one or more processors to implement the magnetic resonance image classification method based on quantum convolutional neural network as described above.

[0028] According to one aspect of the present application, a computer program product includes computer-readable instructions, which are stored in a storage medium. One or more processors of an electronic device read the computer-readable instructions from the storage medium, load and execute the computer-readable instructions, so that the electronic device implements the magnetic resonance image classification method based on quantum convolutional neural network as described above.

[0029] The technical solution provided in this application has the following beneficial effects: combining the principles of quantum computing with the functions of convolutional neural networks. By utilizing the characteristics of quantum computing, quantum convolutional neural networks can obtain richer information representation; and by utilizing quantum parallelism, the overfitting problem of convolutional neural networks can be potentially alleviated; in addition, quantum convolutional neural networks use quantum mechanical ideas such as entanglement, superposition and interference to provide faster and more powerful image data analysis and processing capabilities, thereby obtaining better classification effects and improving the efficiency of image classification in magnetic resonance imaging. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for describing the embodiments of the present application. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without inventive efforts.

[0031] FIG1 is a schematic diagram of an implementation environment according to the present application;

[0032] FIG2 is a flow chart showing a method for magnetic resonance image classification based on quantum convolutional neural network according to an exemplary embodiment;

[0033] FIG3 is a schematic diagram of a convolutional layer, a pooling layer, a fully connected layer, and a quantum layer in a magnetic resonance image classification method based on a quantum convolutional neural network according to an exemplary embodiment;

[0034] FIG4 is a flowchart from S121 to S122 in a magnetic resonance image classification method based on a quantum convolutional neural network according to an exemplary embodiment;

[0035] FIG5 is a flowchart of steps S130 to S150 in a magnetic resonance image classification method based on a quantum convolutional neural network according to an exemplary embodiment;

[0036] FIG6 is a structural block diagram of a magnetic resonance image classification device based on a quantum convolutional neural network according to an exemplary embodiment;

[0037] FIG7 is another structural block diagram of a magnetic resonance image classification device based on a quantum convolutional neural network according to an exemplary embodiment;

[0038] Fig. 8 is a structural block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0039] The following describes in detail embodiments of the present application, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and are not to be construed as limiting the present application.

[0040] It will be understood by those skilled in the art that, unless expressly stated otherwise, the singular forms "a", "an", "said" and "the" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the specification of the present disclosure refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, "connected" or "coupled" as used herein may include wireless connections or wireless couplings. The term "and / or" used herein includes all or any units and all combinations of one or more associated listed items.

[0041] Convolutional neural networks rely on classical computing principles and are limited by classical computing hardware. This limitation may restrict their ability to effectively learn and model complex image classification tasks. Convolutional neural networks are also prone to overfitting, and on smaller datasets, convolutional neural network models often struggle to perform well. As can be seen from the above, the related art still suffers from the low efficiency of magnetic resonance imaging classification based on quantum convolutional neural networks.

[0042] To this end, the magnetic resonance image classification method based on quantum convolutional neural network provided by the present application can effectively improve the accuracy of magnetic resonance image classification based on quantum convolutional neural network. Accordingly, the magnetic resonance image classification method based on quantum convolutional neural network is applicable to a magnetic resonance image classification device based on quantum convolutional neural network, and the magnetic resonance image classification device based on quantum convolutional neural network can be deployed in electronic equipment.

[0043] In order to make the objectives, technical solutions and advantages of this application clearer, the implementation methods of this application will be further described in detail below with reference to the accompanying drawings.

[0044] Figure 1 is a schematic diagram of an implementation environment involved in an image processing method. It should be noted that this implementation environment is only an example adapted for the present invention and should not be considered to provide any limitation on the scope of application of the present invention. The implementation environment includes a collection end and a service end.

[0045] Specifically, the acquisition end can also be considered as an image acquisition device, including but not limited to electronic devices with shooting functions such as cameras, cameras, and video recorders. For example, the acquisition end is an MRI image acquisition device.

[0046] The server side can be a desktop computer, laptop computer, server, or other electronic device. It can also be a computer cluster consisting of multiple servers, or even a cloud computing center consisting of multiple servers. The server side is used to provide background services, such as, but not limited to, magnetic resonance imaging classification services.

[0047] A network communication connection is established between the server and the acquisition end in advance through wired or wireless means, and data transmission between the server and the acquisition end is achieved through the network communication connection. The transmitted data includes but is not limited to: MRI images, etc.

[0048] In one application scenario, through the interaction between the acquisition end and the server end, after the acquisition end captures the MRI image, it can upload the MRI image to the server end to request the server end to provide magnetic resonance image classification service.

[0049] For the server, after receiving the MRI image uploaded by the acquisition end, it calls the magnetic resonance image classification service to classify the MRI image, thereby improving the classification efficiency of magnetic resonance imaging.

[0050] Please refer to Figure 2. An embodiment of the present application provides a magnetic resonance image classification method based on a quantum convolutional neural network. The method is applicable to an electronic device, which can be a server in the implementation environment shown in Figure 1.

[0051] In the following method embodiments, for ease of description, the execution subject of each step of the method is taken as an electronic device as an example for illustration, but this does not constitute a specific limitation.

[0052] As shown in FIG2 , the method may include the following steps:

[0053] S100, acquiring an MRI image, and acquiring corresponding MRI image feature data based on the MRI image, the convolution layer, and the pooling layer;

[0054] Among them, MRI images are magnetic resonance imaging. It should be pointed out that the convolution layer can extract local features in the image through convolution operations. The convolution operation can simulate the human eye's perception of the image and extract local features in the image by sliding the convolution kernel. Different convolution kernels can extract different features, such as edges, textures, colors, etc.; the pooling layer is used to reduce the dimension of the feature map, reduce the amount of calculation and the risk of overfitting. The pooling operation is usually performed after the convolution layer. It can aggregate local features in the feature map to obtain global features; by stacking multiple convolution layers and pooling layers, higher-level MRI image feature data can be gradually extracted from the MRI image. These MRI image feature data can be used for subsequent classification to help the model better understand and analyze MRI images.

[0055] In addition, in an embodiment of the present application, the structure of the applied quantum convolutional neural network is shown in Figure 3, including two convolutional layers, two pooling layers, two fully connected layers and one quantum layer, and the two convolutional layers, two pooling layers, and two fully connected layers constitute a conventional layer, and the conventional layer is connected to the quantum layer. In addition, the internal arrangement of the conventional layer is: convolution layer, pooling layer, convolution layer, pooling layer, fully connected layer and fully connected layer, and the second fully connected layer outputs MRI image category data to the quantum layer.

[0056] S110, inputting the MRI image feature data into a fully connected layer to obtain corresponding MRI image category data. In the embodiment of the present application, the fully connected layer has two layers;

[0057] The fully connected layer can integrate the extracted features and output the corresponding prediction results. Each neuron in the fully connected layer is connected to all neurons in the previous layer, so it can receive all feature information from the previous layer, that is, the MRI image feature data output by the pooling layer. In the fully connected layer, each neuron performs a weighted summation of the input features according to its weight and bias, and then obtains the output of the neuron through the activation function.

[0058] S120 , inputting the MRI image category data into the quantum layer to convert it into quantum state data, and performing image classification through the quantum layer based on the quantum state data.

[0059] In this embodiment of the present application, the quantum layer includes the ZZFeatureMap and Ansatz layers, as shown in FIG4 , and the method further includes:

[0060] S121, based on inputting the MRI image category data into the ZZFeatureMap, the ZZFeatureMap converts the MRI image category data into quantum state data;

[0061] In this application, the ZZFeatureMap is responsible for encoding classical data into the quantum realm. This quantum feature map represents the data within the quantum system, transforming the input data (x) into a quantum state. It creates complex quantum states by applying a series of structured single-qubit and two-qubit gates to the qubits. The ZZFeatureMap can be mathematically represented as follows:

[0062] U(θ)=∏ k U k (θ k ), where θ k =x i (2π-x j ) For i<j, U k (θ k ) is a two-qubit ZZ gate. After the MRI image category data is converted into quantum state data through ZZFeatureMap, the quantum state records the significant features of the MRI image.

[0063] S122, input the quantum state data into the Ansatz layer for transformation and measurement.

[0064] Among them, the Ansatz layer, also known as the variational layer, processes the quantum state data created by the ZZFeatureMap. In the embodiment of this application, the RealAmplitudes circuit is used as Ansatz. The RealAmplitudes circuit consists of a single-qubit Y rotation and an entanglement gate CNOT. By studying the Y rotation (controlled by a programmable parameter changed during training), the quantum convolutional neural network can adjust its behavior by learning the input data. The entanglement gate connects the qubits and realizes the interactions required for the complex quantum behavior required for quantum computing; according to mathematical principles, the RealAmplitudes circuit is expressed as follows:

[0065] Among them, Ry(φ i ) is a single-qubit Y rotation with an angle of φ i , CNOT is the entanglement gate.

[0066] In addition, before the quantum convolutional neural network is put into use, it needs to be trained, as shown in Figure 5, including:

[0067] S130, mapping the class labels {0, 1} to {1, -1} respectively; by using the {1, -1} labels, the number of certain quantum gate operations can be reduced, thereby accelerating the calculation process.

[0068] S140, obtain the mean square error (MSE) of the prediction and class label as

[0069] in, is the Pauli-Z expectation value of a qubit state extracted from the quantum convolutional neural network for the i-th training data, are the labels of the corresponding training data.

[0070] By calculating the mean square error, we can use the mean square error as an optimization target and minimize the mean square error by adjusting the model parameters. By minimizing the mean square error, we can improve the prediction accuracy of the quantum convolutional neural model and make the quantum convolutional neural model perform better.

[0071] S150, obtain the mixed state a i |0><0|+b i |1><1|, if the i-th training data is marked as 0, then the cost function of θ will correspond to a i Greater than b i , at this time, in order to make the model's prediction result closer to the actual label 0, if it is labeled as 1, the cost function of θ will correspond to a i Less than or equal to b i , so that the model's prediction results are closer to the actual label 1.

[0072] This enables the quantum convolutional neural model to better learn and generalize different categories of data, improving the performance and generalization ability of the quantum convolutional neural model.

[0073] The following are embodiments of the apparatus of the present application, which can be used to implement the magnetic resonance image classification method based on quantum convolutional neural network involved in this application. For details not disclosed in the apparatus embodiments of the present application, please refer to the method embodiments of the magnetic resonance image classification method based on quantum convolutional neural network involved in this application.

[0074] Referring to FIG6 , an embodiment of the present application provides a magnetic resonance image classification device based on a quantum convolutional neural network, including but not limited to: a feature data acquisition module 200 , a category data acquisition module 210 , and a classification module 220 ;

[0075] A feature data acquisition module 200 acquires an MRI image and, based on the MRI image, a convolutional layer and a pooling layer, is used to acquire corresponding MRI image feature data;

[0076] The category data acquisition module 210 inputs the MRI image feature data into a fully connected layer to obtain corresponding MRI image category data;

[0077] The classification module 220 inputs the MRI image category data into the quantum layer to convert it into quantum state data, and performs image classification through the quantum layer based on the quantum state data.

[0078] In an exemplary embodiment, referring to FIG. 7 , the apparatus includes but is not limited to:

[0079] A conversion module 300 is configured to convert the MRI image category data into quantum state data based on inputting the MRI image category data into the ZZFeatureMap;

[0080] The quantum state data is input into the Ansatz layer for transformation and measurement.

[0081] In an exemplary embodiment, the device includes, but is not limited to:

[0082] The expression of the ZZFeatureMap is: U(θ)=∏ k U k (θ k ), where θ k =x i (2π-x j ) For i<j, U k (θ k ) is a two-qubit ZZ gate.

[0083] In an exemplary embodiment, the device includes, but is not limited to:

[0084] The Ansatz layer uses a RealAmplitudes circuit, and the RealAmplitudes circuit is expressed as follows: Among them, Ry(φ i ) is a single-qubit Y rotation with an angle of φ i , CNOT is the entanglement gate.

[0085] In an exemplary embodiment, the device includes, but is not limited to:

[0086] Two of the convolutional layers, two of the pooling layers, two of the fully connected layers and one of the quantum layers, two of the convolutional layers, two of the pooling layers and two of the fully connected layers constitute a regular layer, and the regular layer is connected to the quantum layer. The internal arrangement of the regular layer is: convolutional layer, pooling layer, convolutional layer, pooling layer, fully connected layer and fully connected layer, and the second fully connected layer outputs MRI image category data.

[0087] In an exemplary embodiment, the device includes, but is not limited to:

[0088] A mapping module 400 is used to map the class labels {0, 1} to {1, -1} respectively;

[0089] The mean square error acquisition module 410 is used to obtain the mean square error (MSE) of the prediction and class label. in, is the Pauli-Z expectation value of a qubit state extracted from the quantum convolutional neural network for the i-th training data, is the label of the corresponding training data;

[0090] State acquisition module 420, used to obtain the mixed state a i |0><0|+b i |1><1|, if the i-th training data is marked as 0, then the cost function of θ will correspond to a i Greater than b i , if it is marked as 1, then the cost function of θ will correspond to a i Less than or equal to b i .

[0091] It should be noted that the magnetic resonance image classification device based on quantum convolutional neural network provided in the above embodiment only uses the division of the above-mentioned functional modules as an example when performing magnetic resonance image classification based on quantum convolutional neural network. In actual application, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the magnetic resonance image classification device based on quantum convolutional neural network will be divided into different functional modules to complete all or part of the functions described above.

[0092] In addition, the magnetic resonance image classification device based on quantum convolutional neural network provided in the above embodiment and the magnetic resonance image classification method based on quantum convolutional neural network belong to the same concept, and the specific manner in which each module performs operations has been described in detail in the method embodiment and will not be repeated here.

[0093] Please refer to FIG8 . An embodiment of the present application provides an electronic device 4000 , which may include a desktop computer, a laptop computer, a server, etc.

[0094] In FIG. 8 , the electronic device 4000 includes at least one processor 4001 and at least one memory 4003 .

[0095] Data exchange between the processor 4001 and the memory 4003 can be achieved through at least one communication bus 4002. The communication bus 4002 may include a path for transmitting data between the processor 4001 and the memory 4003. The communication bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 4002 can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is used in FIG8 , but this does not mean that there is only one bus or one type of bus.

[0096] Optionally, the electronic device 4000 may further include a transceiver 4004, which may be used for data exchange between the electronic device and other electronic devices, such as data transmission and / or data reception. It should be noted that in actual applications, the number of transceivers 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present application.

[0097] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0098] The memory 4003 can be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program instructions or codes in the form of instructions or data structures and can be accessed by the electronic device 400, but is not limited to these.

[0099] Computer-readable instructions are stored in the memory 4003 , and the processor 4001 can read the computer-readable instructions stored in the memory 4003 through the communication bus 4002 .

[0100] The computer-readable instructions are executed by one or more processors 4001 to implement the magnetic resonance image classification method based on quantum convolutional neural network in the above embodiments.

[0101] In addition, an embodiment of the present application provides a storage medium on which computer-readable instructions are stored. The computer-readable instructions are executed by one or more processors to implement the magnetic resonance image classification method based on quantum convolutional neural network as described above.

[0102] In an embodiment of the present application, a computer program product is provided. The computer program product includes computer-readable instructions, which are stored in a storage medium. One or more processors of an electronic device read the computer-readable instructions from the storage medium, load and execute the computer-readable instructions, so that the electronic device implements the magnetic resonance image classification method based on quantum convolutional neural network as described above.

[0103] Compared with related technologies, quantum convolutional neural networks combine quantum computing principles with the functions of convolutional neural networks. By utilizing the characteristics of quantum computing, quantum convolutional neural networks can obtain richer information representation; and by utilizing quantum parallelism, they can potentially alleviate the overfitting problem of convolutional neural networks. In addition, quantum convolutional neural networks use quantum mechanical ideas such as entanglement, superposition and interference to provide faster and more powerful image data analysis and processing capabilities, thereby obtaining better classification effects and improving the efficiency of image classification in magnetic resonance imaging.

[0104] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0105] The above description is only part of the implementation methods of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A magnetic resonance image classification method based on quantum convolutional neural network, characterized in that: include: Acquire an MRI image, and acquire corresponding MRI image feature data based on the MRI image, the convolution layer, and the pooling layer; Inputting the MRI image feature data into a fully connected layer to obtain corresponding MRI image category data; The MRI image category data is input into a quantum layer and converted into quantum state data, and image classification is performed through the quantum layer based on the quantum state data.

2. The method according to claim 1, wherein In the method of inputting the MRI image category data into a quantum layer for conversion into quantum state data, the quantum layer includes a ZZFeatureMap and an Ansatz layer, and the method further includes: Based on inputting the MRI image category data into the ZZFeatureMap, the ZZFeatureMap converts the MRI image category data into quantum state data; The quantum state data is input into the Ansatz layer for transformation and measurement.

3. The method according to claim 2, wherein The method further comprises: The expression of the ZZFeatureMap is: U(θ)=∏ k U k (θ k ), where θ k =x i (2π-x j ) For i<j, U k (θ k ) is a two-qubit ZZ gate.

4. The method according to claim 2, wherein The method further comprises: The Ansatz layer uses a RealAmplitudes circuit, and the RealAmplitudes circuit is expressed as follows: Among them, Ry(φ i ) is a single-qubit Y rotation with an angle of φ i , CNOT is the entanglement gate.

5. The method according to claim 1, wherein The method further comprises: Two of the convolutional layers, two of the pooling layers, two of the fully connected layers and one of the quantum layers, two of the convolutional layers, two of the pooling layers and two of the fully connected layers constitute a regular layer, and the regular layer is connected to the quantum layer. The internal arrangement of the regular layer is: convolutional layer, pooling layer, convolutional layer, pooling layer, fully connected layer and fully connected layer, and the second fully connected layer outputs MRI image category data.

6. The method according to claim 1, wherein In the training method of the quantum convolutional neural network, the method further includes: Map class labels {0, 1} to {1, -1} respectively; Get the mean squared error (MSE) of the predictions and class labels as in, is the Pauli-Z expectation value of a qubit state extracted from the quantum convolutional neural network for the i-th training data, is the label of the corresponding training data; Get the mixed state a i |0><0|+b i |1><1|, if the i-th training data is marked as 0, then the cost function of θ will correspond to a i Greater than b i , if it is marked as 1, then the cost function of θ will correspond to a i Less than or equal to b i .

7. A magnetic resonance image classification device based on quantum convolutional neural network, characterized in that: include: A feature data acquisition module acquires an MRI image and is used to acquire corresponding MRI image feature data based on the MRI image, the convolution layer, and the pooling layer; a category data acquisition module, inputting the MRI image feature data into a fully connected layer to obtain corresponding MRI image category data; The classification module inputs the MRI image category data into the quantum layer and converts it into quantum state data. The quantum state data is used to perform image classification through the quantum layer.

8. The device according to claim 7, wherein The quantum layer includes a ZZFeatureMap and an Ansatz layer, and the device further includes: a conversion module, based on inputting the MRI image category data into the ZZFeatureMap, wherein the ZZFeatureMap is used to convert the MRI image category data into quantum state data; The quantum state data is input into the Ansatz layer for transformation and measurement.

9. An electronic device, characterized in that: include: at least one processor and at least one memory, wherein: The memory has computer-readable instructions stored thereon; The computer-readable instructions are executed by one or more of the processors, so that the electronic device implements the magnetic resonance image classification method based on quantum convolutional neural network according to any one of claims 1 to 6.

10. A storage medium having computer-readable instructions stored thereon, characterized in that: The computer-readable instructions are executed by one or more processors to implement the magnetic resonance image classification method based on quantum convolutional neural network according to any one of claims 1 to 6.

Images