CN117710761A - Quantum convolution neural network-based magnetic resonance image classification method and device - Google Patents
Quantum convolution neural network-based magnetic resonance image classification method and device Download PDFInfo
- Publication number
- CN117710761A CN117710761A CN202410169637.7A CN202410169637A CN117710761A CN 117710761 A CN117710761 A CN 117710761A CN 202410169637 A CN202410169637 A CN 202410169637A CN 117710761 A CN117710761 A CN 117710761A
- Authority
- CN
- China
- Prior art keywords
- quantum
- layer
- data
- mri image
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 13
- 238000011176 pooling Methods 0.000 claims abstract description 25
- 238000006243 chemical reaction Methods 0.000 claims abstract description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 49
- 239000002096 quantum dot Substances 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 4
- 238000002595 magnetic resonance imaging Methods 0.000 abstract description 64
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000004891 communication Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 210000004556 brain Anatomy 0.000 description 4
- 230000006872 improvement Effects 0.000 description 4
- 230000003993 interaction Effects 0.000 description 4
- 210000005036 nerve Anatomy 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000004590 computer program Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000002610 neuroimaging Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000005610 quantum mechanics Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003703 image analysis method Methods 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000013526 transfer learning Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The application provides a magnetic resonance image classification method and device based on a quantum convolution neural network, and relates to the technical field of image classification. Wherein the method comprises the following steps: acquiring an MRI image, and acquiring corresponding MRI image characteristic data based on the MRI image, a convolution layer and a pooling layer; inputting the MRI image characteristic data into a full-connection layer to obtain corresponding MRI image category data; the MRI image class data is input to a quantum layer for conversion to quantum state data, and image classification is performed by the quantum layer based on the quantum state data. The method and the device solve the problem of low image classification efficiency in magnetic resonance imaging in the related technology.
Description
Technical Field
The application relates to the technical field of image classification, in particular to a magnetic resonance image classification method and device based on a quantum convolution neural network.
Background
Magnetic resonance imaging is a widely used method for obtaining high quality medical images, which is one of the popular, painless, noninvasive brain imaging techniques. For brain imaging, magnetic resonance imaging presents a unique view by providing a high level of spatial and contrast resolution. It is used not only for diagnosing various diseases but also for providing high quality information images about the internal structures of the brain. However, the amount of data extracted from these images is enormous, and it is difficult to draw conclusive diagnosis from the raw data. In this case, we need to analyze the magnetic resonance image with various image analysis tools to extract conclusive information to classify the brain normal and abnormal.
Many medical image analysis methods based on deep learning techniques have been proposed for health monitoring and diagnosis of brain magnetic resonance images. Convolutional neural networks are a branch of deep learning, commonly used to analyze visual information. Convolutional neural networks consist of many trainable layers including an input layer, a convolutional layer, a connection layer, and a full connection layer. The convolution layer and the connection layer may be fine-tuned by super-parameters. Different convolutional neural network architectures are combined with different transfer learning techniques, and have achieved great success due to their improved performance in terms of image classification. In this way, they have exceeded the traditional machine learning model in the last few years.
Although many brain magnetic resonance image classification algorithms have been developed, the existing methods still suffer from a number of drawbacks, as many algorithms rely on human-selected lesion locations and thus they cannot be fully automated. Furthermore, convolutional neural networks rely on classical computing principles and are limited by classical computing hardware. As a result, they may be challenged to efficiently process and represent high-dimensional image data, and such limitations may limit their ability to efficiently learn and model complex image classification tasks; convolutional neural networks also have the problem of being prone to overfitting, and convolutional neural network models often perform poorly on smaller data sets.
From the above, how to improve the image classification efficiency in the magnetic resonance imaging is still to be solved.
Disclosure of Invention
The application provides a magnetic resonance image classification method, a device, electronic equipment and a storage medium based on a quantum convolution neural network, which can solve the problem of low image classification efficiency in magnetic resonance imaging in the related technology. The technical scheme is as follows:
according to one aspect of the application, a magnetic resonance image classification method based on a quantum convolutional neural network comprises: acquiring an MRI image, and acquiring corresponding MRI image characteristic data based on the MRI image, a convolution layer and a pooling layer; inputting the MRI image characteristic data into a full-connection layer to obtain corresponding MRI image category data; the MRI image class data is input to a quantum layer for conversion to quantum state data, and image classification is performed by the quantum layer based on the quantum state data.
According to one aspect of the present application, a magnetic resonance image classification apparatus based on a quantum convolutional neural network includes:
the feature data acquisition module is used for acquiring MRI images and acquiring corresponding MRI image feature data based on the MRI images, the convolution layer and the pooling layer;
the category data acquisition module inputs the MRI image characteristic data into the full-connection layer for acquiring corresponding MRI image category data;
the classification module inputs the MRI image category data to a quantum layer to be converted into quantum state data, and is used for classifying images through the quantum layer based on the quantum state data.
In an exemplary embodiment, the apparatus includes, but is not limited to:
a conversion module for converting the MRI image class data into quantum state data based on inputting the MRI image class data into the zzzfeaturemap;
the quantum state data is input to the Ansatz layer for transformation and measurement.
In an exemplary embodiment, the apparatus includes, but is not limited to:
the expression of the ZZFeatureMap is as follows:wherein->For i<j，Is a two qubit ZZ gate.
In an exemplary embodiment, the apparatus includes, but is not limited to:
the Ansatz layer employs a RealAmpliforms circuit, and the RealAmpliforms circuit is represented as follows:wherein->Is a single quantum bit Y rotation with an angle of +.>CNOT is the entanglement gate.
In an exemplary embodiment, the apparatus includes, but is not limited to:
the two convolution layers, the two pooling layers, the two full-connection layers and the one quantum layer form a conventional layer, the conventional layer is connected with the quantum layer, and the arrangement mode inside the conventional layer is as follows: the MRI image processing system comprises a reel layer, a pooling layer, a convolution layer, a pooling layer, a full connection layer and a full connection layer, wherein the second full connection layer outputs MRI image category data.
In an exemplary embodiment, the apparatus includes, but is not limited to:
a mapping module, configured to map class labels {0, 1} to {1, -1} respectively;
a mean square error acquisition module for acquiring a Mean Square Error (MSE) of the prediction and class labels asThe method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Pauli-Z expectation of one qubit state extracted from the quantum convolutional neural network for the ith training data,/one>Is a label of the corresponding training data;
a state acquisition module for acquiring a mixed stateIf the ith training data is marked 0, the cost function of θ will correspond to +.>Is greater than->If it is marked 1, the cost function of θ will correspond toLess than or equal to->。
According to one aspect of the application, an electronic device comprises 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 to cause an electronic device to implement a magnetic resonance image classification method based on a quantum convolutional neural network as described above.
According to one aspect of the application, a storage medium has stored thereon computer readable instructions that are executed by one or more processors to implement the method of magnetic resonance image classification based on a quantum convolutional neural network as described above.
According to one aspect of the application, a computer program product includes computer-readable instructions stored in a storage medium, one or more processors of an electronic device reading the computer-readable instructions from the storage medium, loading and executing the computer-readable instructions, causing the electronic device to implement a quantum convolutional neural network-based magnetic resonance image classification method as described above.
The beneficial effects that this application provided technical scheme brought are: the quantum computing principle is combined with the function of the convolutional neural network, and the quantum convolutional neural network can obtain richer information representation by utilizing the characteristic of quantum computing; and by exploiting quantum parallelism, potentially alleviating the overfitting problem of convolutional neural networks; in addition, the quantum convolution neural network provides faster and stronger image data analysis processing capacity by utilizing ideas of quantum mechanics such as entanglement, superposition, interference and the like, thereby obtaining better classification effect and realizing improvement of image classification efficiency in magnetic resonance imaging.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings that are required to be used in the description of the embodiments of the present application will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained from these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic illustration of an implementation environment in accordance with the teachings of the present application;
FIG. 2 is a flow chart illustrating a method of classifying magnetic resonance images based on a quantum convolutional neural network, according to an exemplary embodiment;
FIG. 3 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 example embodiment;
FIG. 4 is a flowchart illustrating S121 through S122 in a method of classifying magnetic resonance images based on a quantum convolutional neural network, according to an exemplary embodiment;
FIG. 5 is a flowchart illustrating S130 through S150 of a method for classifying magnetic resonance images based on a quantum convolutional neural network, according to an exemplary embodiment;
FIG. 6 is a block diagram illustrating a magnetic resonance image classification apparatus based on a quantum convolutional neural network, according to an exemplary embodiment;
FIG. 7 is a further block diagram illustrating a magnetic resonance image classification apparatus based on a quantum convolutional neural network, according to an exemplary embodiment;
fig. 8 is a block diagram illustrating a configuration of an electronic device according to an exemplary embodiment.
Detailed Description
Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary only for the purpose of illustrating the present application and are not to be construed as limiting the present application.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless expressly stated otherwise, as understood by those skilled in the art. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification of this disclosure, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly coupled. The term "and/or" as used herein includes all or any element and all combination of one or more of the associated listed items.
Convolutional neural networks rely on classical computing principles and are limited by classical computing hardware, which may limit their ability to efficiently learn and model complex image classification tasks, and convolutional neural networks also suffer from the problem of being prone to overfitting, which is often difficult to perform well on smaller data sets. As can be seen from the above, the related art still has the defect of low magnetic resonance image classification efficiency based on the quantum convolutional neural network.
Therefore, the magnetic resonance image classification method based on the quantum convolution neural network can effectively improve the accuracy of magnetic resonance image classification based on the quantum convolution neural network, and is correspondingly suitable for a magnetic resonance image classification device based on the quantum convolution neural network, the magnetic resonance image classification device based on the quantum convolution neural network can be deployed in electronic equipment,
for the purpose of making the objects, technical solutions and advantages of the present application more apparent, the embodiments of the present application will be described in further detail below with reference to the accompanying drawings.
Fig. 1 is a schematic diagram of an implementation environment related to an image processing method. It should be noted that this implementation environment is only one example adapted to the present invention and should not be considered as providing any limitation to the scope of use of the present invention. The implementation environment comprises an acquisition end and a service end.
Specifically, the capturing end may also be considered as an image capturing device, including but not limited to, a video camera, and other electronic devices having a photographing function. For example, the acquisition end is an MRI image acquisition device.
The server can be electronic equipment such as a desktop computer, a notebook computer, a server and the like, and can also be a computer cluster formed by a plurality of servers, or even a cloud computing center formed by a plurality of servers. The server is used for providing background services, for example, background services include, but are not limited to, magnetic resonance image classification services and the like.
The network communication connection is pre-established between the server and the acquisition end in a wired or wireless mode, and data transmission between the server and the acquisition end is realized through the network communication connection. The data transmitted includes, but is not limited to: MRI images, and the like.
In an application scene, after the acquisition end shoots an MRI image through interaction between the acquisition end and the service end, the MRI image can be uploaded to the service end to request the service end to provide magnetic resonance image classification service.
For the service end, after receiving the MRI image uploaded by the acquisition end, the MRI image classification service is called to realize the classification of the MRI image, so that the classification efficiency of the MRI image is improved.
Referring to fig. 2, an embodiment of the present application provides a magnetic resonance image classification method based on a quantum convolutional neural network, where the method is applicable to an electronic device, and the electronic device may be a server in the implementation environment shown in fig. 1.
In the following method embodiments, for convenience of description, the execution subject of each step of the method is described as an electronic device, but this configuration is not particularly limited.
As shown in fig. 2, the method may include the steps of:
s100, acquiring an MRI image, and acquiring corresponding MRI image characteristic data based on the MRI image, the convolution layer and the pooling layer;
wherein the MRI image is a magnetic resonance image, it should be noted that the convolution layer may extract local features in the image through convolution operation, the convolution operation may simulate the human eye's perception process of the image, and the local features in the image may be extracted through sliding convolution kernels, and different convolution kernels may extract different features, such as edges, textures, colors, etc.; the pooling layer is used for reducing the dimension of the feature map, reducing the calculated amount and the risk of overfitting, and the pooling operation is usually carried out after the convolution layer, so that the local features in the feature map can be aggregated to obtain global features; by stacking multiple convolution and pooling layers, higher levels of MRI image feature data may be progressively extracted from the MRI image, which may be used for subsequent classification, helping the model to better understand and analyze the MRI image.
In addition, in the embodiment of the present application, as shown in fig. 3, the structure of the quantum convolutional neural network applied to the present application includes two convolutional layers, two pooling layers, two fully-connected layers and one quantum layer, where the two convolutional layers, the two pooling layers and the two fully-connected layers form a conventional layer, and the conventional layer is connected with the quantum layer, and in addition, the internal arrangement manner of the conventional layer is as follows: the MRI system comprises a reel layer, a pooling layer, a convolution layer, a pooling layer, a full connection layer and a full connection layer, wherein the second full connection layer outputs MRI image category data to the quantum layer.
S110, inputting the MRI image characteristic data into a full-connection layer to obtain corresponding MRI image category data, wherein in the embodiment of the application, the full-connection layer has two layers;
the full connection layer can integrate the extracted features and output corresponding prediction results. And each neuron in the fully connected layer is connected to all neurons in the previous layer so that it can receive all characteristic information from the previous layer, i.e., MRI image characteristic data output from the pooling layer, in which each neuron weights and sums the input characteristics according to its weight and bias, and then obtains the output of that neuron through an activation function.
S120, MRI image category data are input into a quantum layer to be converted into quantum state data, and image classification is carried out through the quantum layer based on the quantum state data.
In this embodiment, the quantum layer includes zzzfeaturemap and Ansatz layers, as shown in fig. 4, where the method further includes:
s121, converting MRI image class data into quantum state data based on inputting the MRI image class data into ZZFeatureMap, ZZFeatureMap;
in this application, zzzfeaturemap is responsible for encoding classical data into the quantum domain. This quantum feature map symbolizes data inside the quantum system, which converts the input data (x) into quantum states. It creates complex quantum states by applying a series of structured single and double qubit gates to the qubit. The zzzfeaturemap can be expressed mathematically as follows:
wherein->For i<j，/>Is two qubit ZZ gates, and after MRI image class data is converted into quantum state data by ZZFeatureMap, the quantum state records the salient features of MRI images.
S122, inputting the quantum state data into an Ansatz layer for transformation and measurement.
Wherein the Ansatz layer, also called a variational layer, processes quantum state data created by ZZFeatureMap, in this embodiment of the application, using a RealAmplitude circuit as Ansatz, the RealAmplitude circuit consists of a single qubit Y rotation and entanglement gate CNOT, by studying the Y rotation (controlled by programmable parameters altered during training), the quantum convolutional neural network can adjust its behavior by learning the input data. The entanglement gate connects the qubits and achieves the interactions required for complex quantum behavior required for quantum computation; according to the mathematical principle, the realganitudes circuit is represented as follows:
wherein->Is a single quantum bit Y rotation with an angle of +.>CNOT is the entanglement gate.
In addition, before the quantum convolutional neural network is put into use, the quantum convolutional neural network needs to be trained, as shown in fig. 5, including:
s130, mapping class labels {0, 1} to {1, -1} respectively; the number of certain quantum gate operations can be reduced by using the {1, -1} tag, thereby speeding up the computation process.
S140, obtaining the Mean Square Error (MSE) of the prediction and class labels as；
Wherein,Pauli-Z expectation of one qubit state extracted from the quantum convolutional neural network for the ith training data,/one>Is a label of the corresponding training data.
By calculating the mean square error, the mean square error can be used as an optimization target, and the mean square error can be minimized by adjusting the parameters of the model. By minimizing the mean square error, the prediction precision of the quantum convolution nerve model can be improved, and the performance of the quantum convolution nerve model is better.
S150, obtaining a mixed stateIf the ith training data is marked 0, the cost function of θ will correspond to +.>Is greater than->At this point in order to bring the prediction result of the model closer to the actual tag 0, if it is marked 1, the cost function of θ will correspond to +.>Less than or equal to->So as to bring the prediction result of the model closer to the actual tag 1.
Therefore, the quantum convolution nerve model can learn and generalize different types of data better, and the performance and generalization capability of the quantum convolution nerve model are improved.
The following is an embodiment of the apparatus of the present application, which may be used to perform the method for classifying magnetic resonance images based on the quantum convolutional neural network according to the present application. For details not disclosed in the device embodiments of the present application, please refer to a method embodiment of a magnetic resonance image classification method based on a quantum convolutional neural network.
Referring to fig. 6, in an embodiment of the present application, a magnetic resonance image classification apparatus based on a quantum convolutional neural network is provided, including but not limited to: a feature data acquisition module 200, a category data acquisition module 210, and a classification module 220;
the feature data acquisition module 200 is used for acquiring an MRI image and acquiring corresponding MRI image feature data based on the MRI image, the convolution layer and the pooling layer;
the category data acquisition module 210 inputs the MRI image feature data into the full connection layer for acquiring corresponding MRI image category data;
the classification module 220 converts the MRI image class data input to a quantum layer into quantum state data, and is used for image classification by the quantum layer based on the quantum state data.
In an exemplary embodiment, referring to fig. 7, the apparatus includes, but is not limited to:
a conversion module 300 for converting the MRI image class data into quantum state data based on inputting the MRI image class data into the zzzfeaturemap;
the quantum state data is input to the Ansatz layer for transformation and measurement.
In an exemplary embodiment, the apparatus includes, but is not limited to:
the expression of the ZZFeatureMap is as follows:wherein->For i<j，Is a two qubit ZZ gate.
In an exemplary embodiment, the apparatus includes, but is not limited to:
the Ansatz layer employs a RealAmpliforms circuit, and the RealAmpliforms circuit is represented as follows:wherein->Is a single quantum bit Y rotation with an angle of +.>CNOT is the entanglement gate.
In an exemplary embodiment, the apparatus includes, but is not limited to:
the two convolution layers, the two pooling layers, the two full-connection layers and the one quantum layer form a conventional layer, the conventional layer is connected with the quantum layer, and the arrangement mode inside the conventional layer is as follows: the MRI image processing system comprises a reel layer, a pooling layer, a convolution layer, a pooling layer, a full connection layer and a full connection layer, wherein the second full connection layer outputs MRI image category data.
In an exemplary embodiment, the apparatus includes, but is not limited to:
a mapping module 400, configured to map class labels {0, 1} to {1, -1} respectively;
a mean square error acquisition module 410 for acquiring a Mean Square Error (MSE) of the prediction and class labels asThe method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Pauli-Z expectation of one qubit state extracted from the quantum convolutional neural network for the ith training data,/one>Is a label of the corresponding training data;
a state acquisition module 420 for acquiring a mixed stateIf the ith training data is marked 0, the cost function of θ will correspond to +.>Is greater than->If it is marked 1, the cost function of θ will correspond to +.>Less than or equal to->。
It should be noted that, when the magnetic resonance image classification device based on the quantum convolutional neural network provided in the above embodiment performs magnetic resonance image classification based on the quantum convolutional neural network, only the division of the above functional modules is used for illustration, and in practical application, the above functional allocation may be completed by different functional modules according to needs, that is, the internal structure of the magnetic resonance image classification device based on the quantum convolutional neural network will be divided into different functional modules, so as to complete all or part of the functions described above.
In addition, the magnetic resonance image classification device based on the quantum convolutional neural network provided in the above embodiment and the embodiment of the magnetic resonance image classification method based on the quantum convolutional neural network belong to the same concept, wherein the specific manner of executing the operations of each module has been described in detail in the method embodiment, and will not be described in detail here.
Referring to fig. 8, in an embodiment of the present application, an electronic device 4000 is provided, where the electronic device 4000 may include: desktop computers, notebook computers, servers, etc.
In fig. 8, the electronic device 4000 includes at least one processor 4001 and at least one memory 4003.
Among other things, data interaction between the processor 4001 and the memory 4003 may be achieved by at least one communication bus 4002. The communication bus 4002 may include a path for transferring data between the processor 4001 and the memory 4003. The communication bus 4002 may be a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus or an EISA (Extended Industry Standard Architecture ) bus, or the like. The communication bus 4002 can be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in fig. 8, but not only one bus or one type of bus.
Optionally, the electronic device 4000 may further comprise a transceiver 4004, the transceiver 4004 may be used for data interaction between the electronic device and other electronic devices, such as transmission of data and/or reception of data, etc. It should be noted that, in practical applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 is not limited to the embodiment of the present application.
The processor 4001 may be a CPU (Central Processing Unit ), general purpose processor, DSP (Digital Signal Processor, data signal processor), ASIC (Application Specific Integrated Circuit ), FPGA (Field Programmable Gate Array, field programmable gate array) or other programmable logic device, transistor logic device, hardware components, or any combination thereof. Which may implement or perform the various exemplary logic blocks, modules, and circuits described in connection with this disclosure. The processor 4001 may also be a combination that implements computing functionality, e.g., comprising one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.
Memory 4003 may be, but is not limited to, ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, RAM (Random Access Memory ) or other type of dynamic storage device that can store information and instructions, EEPROM (Electrically Erasable Programmable Read Only Memory ), CD-ROM (Compact Disc Read Only Memory, compact disc Read Only Memory) or other optical disk storage, optical disk storage (including compact disc, laser disc, optical disc, digital versatile disc, blu-ray disc, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program instructions or code in the form of instructions or data structures and that can be accessed by electronic device 400.
The memory 4003 has computer readable instructions stored thereon, and the processor 4001 can read the computer readable instructions stored in the memory 4003 through the communication bus 4002.
The computer readable instructions are executed by the one or more processors 4001 to implement the method for magnetic resonance image classification based on quantum convolutional neural networks in the above embodiments.
Furthermore, in an embodiment of the present application, a storage medium is provided, on which computer readable instructions are stored, the computer readable instructions being executed by one or more processors to implement the method for classifying magnetic resonance images based on a quantum convolutional neural network as described above.
In an embodiment of the present application, a computer program product is provided, where the computer program product includes computer readable instructions, where the computer readable instructions are stored in a storage medium, and where 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 a magnetic resonance image classification method based on a quantum convolutional neural network as described above.
Compared with the related technology, the quantum computing principle is combined with the function of the convolutional neural network, and the quantum convolutional neural network can obtain richer information representation by utilizing the characteristic of quantum computing; and by exploiting quantum parallelism, potentially alleviating the overfitting problem of convolutional neural networks; in addition, the quantum convolution neural network provides faster and stronger image data analysis processing capacity by utilizing ideas of quantum mechanics such as entanglement, superposition, interference and the like, thereby obtaining better classification effect and realizing improvement of image classification efficiency in magnetic resonance imaging.
It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.
The foregoing is only a partial embodiment of the present application, and it should be noted that, for a person skilled in the art, several improvements and modifications can be made without departing from the principle of the present application, and these improvements and modifications should also be considered as the protection scope of the present application.
Claims (10)
1. A magnetic resonance image classification method based on a quantum convolution neural network is characterized by comprising the following steps:
acquiring an MRI image, and acquiring corresponding MRI image characteristic data based on the MRI image, a convolution layer and a pooling layer;
inputting the MRI image characteristic data into a full-connection layer to obtain corresponding MRI image category data;
the MRI image class data is input to a quantum layer for conversion to quantum state data, and image classification is performed by the quantum layer based on the quantum state data.
2. The method of claim 1, wherein in the method of converting the MRI image category data into quantum state data by inputting the quantum layer, the quantum layer comprises zzzfeaturemap and Ansatz layers, the method further comprising:
based on inputting the MRI image class data to the zzzfeaturemap, the zzzfeaturemap converts the MRI image class data to quantum state data;
the quantum state data is input to the Ansatz layer for transformation and measurement.
3. The method of claim 2, wherein the method further comprises:
the expression of the ZZFeatureMap is as follows:wherein->For i<j，/>Is a two qubit ZZ gate.
4. The method of claim 2, wherein the method further comprises:
the Ansatz layer employs a RealAmpliforms circuit, and the RealAmpliforms circuit is represented as follows:wherein->Is a single quantum bit Y rotation with an angle of +.>CNOT is the entanglement gate.
5. The method of claim 1, wherein the method further comprises:
the two convolution layers, the two pooling layers, the two full-connection layers and the one quantum layer form a conventional layer, the conventional layer is connected with the quantum layer, and the arrangement mode inside the conventional layer is as follows: the MRI image processing system comprises a reel layer, a pooling layer, a convolution layer, a pooling layer, a full connection layer and a full connection layer, wherein the second full connection layer outputs MRI image category data.
6. The method of claim 1, wherein in the method of training the quantum convolutional neural network, the method further comprises:
mapping class labels {0, 1} to {1, -1} respectively;
obtaining the Mean Square Error (MSE) of the prediction and class labels asThe method comprises the steps of carrying out a first treatment on the surface of the Wherein,Pauli-Z expectation of one qubit state extracted from the quantum convolutional neural network for the ith training data,/one>Is a label of the corresponding training data;
acquiring a mixed stateIf the ith training data is marked 0, the cost function of θ will correspond toIs greater than->If it is marked 1, the cost function of θ will correspond to +.>Less than or equal to->。
7. A magnetic resonance image classification device based on a quantum convolution neural network, comprising:
the feature data acquisition module is used for acquiring MRI images and acquiring corresponding MRI image feature data based on the MRI images, the convolution layer and the pooling layer;
the category data acquisition module inputs the MRI image characteristic data into the full-connection layer for acquiring corresponding MRI image category data;
the classification module inputs the MRI image category data to a quantum layer to be converted into quantum state data, and is used for classifying images through the quantum layer based on the quantum state data.
8. The apparatus of claim 7, wherein the quantum layers comprise zzzfeaturemap and Ansatz layers, the apparatus further comprising:
a conversion module for converting the MRI image class data into quantum state data based on inputting the MRI image class data into the zzzfeaturemap;
the quantum state data is input to the Ansatz layer for transformation and measurement.
9. An electronic device, comprising: 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 to cause an electronic device to implement the quantum convolutional neural network-based magnetic resonance image classification method of any one of claims 1-6.
10. A storage medium having stored thereon computer readable instructions executable by one or more processors to implement the quantum convolutional neural network-based magnetic resonance image classification method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410169637.7A CN117710761A (en) | 2024-02-06 | 2024-02-06 | Quantum convolution neural network-based magnetic resonance image classification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410169637.7A CN117710761A (en) | 2024-02-06 | 2024-02-06 | Quantum convolution neural network-based magnetic resonance image classification method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117710761A true CN117710761A (en) | 2024-03-15 |
Family
ID=90162940
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410169637.7A Pending CN117710761A (en) | 2024-02-06 | 2024-02-06 | Quantum convolution neural network-based magnetic resonance image classification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117710761A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110739070A (en) * | 2019-09-26 | 2020-01-31 | 南京工业大学 | brain disease diagnosis method based on 3D convolutional neural network |
| CN114169468A (en) * | 2021-12-31 | 2022-03-11 | 上海联影智能医疗科技有限公司 | Image classification method, computer device, storage medium, and computer program product |
| CN116664931A (en) * | 2023-05-29 | 2023-08-29 | 重庆师范大学 | Knee osteoarthritis grading method based on quantum-to-classical migration learning |
| CN116912556A (en) * | 2023-06-09 | 2023-10-20 | 中国长城科技集团股份有限公司 | Picture classification method and device, electronic equipment and storage medium |
| CN117437494A (en) * | 2023-12-20 | 2024-01-23 | 量子科技长三角产业创新中心 | Image classification method, system, electronic equipment and storage medium |
-
2024
- 2024-02-06 CN CN202410169637.7A patent/CN117710761A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110739070A (en) * | 2019-09-26 | 2020-01-31 | 南京工业大学 | brain disease diagnosis method based on 3D convolutional neural network |
| CN114169468A (en) * | 2021-12-31 | 2022-03-11 | 上海联影智能医疗科技有限公司 | Image classification method, computer device, storage medium, and computer program product |
| CN116664931A (en) * | 2023-05-29 | 2023-08-29 | 重庆师范大学 | Knee osteoarthritis grading method based on quantum-to-classical migration learning |
| CN116912556A (en) * | 2023-06-09 | 2023-10-20 | 中国长城科技集团股份有限公司 | Picture classification method and device, electronic equipment and storage medium |
| CN117437494A (en) * | 2023-12-20 | 2024-01-23 | 量子科技长三角产业创新中心 | Image classification method, system, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200311871A1 (en) | Image reconstruction method and device | |
| US11508146B2 (en) | Convolutional neural network processing method and apparatus | |
| CN109685819B (en) | Three-dimensional medical image segmentation method based on feature enhancement | |
| US11574198B2 (en) | Apparatus and method with neural network implementation of domain adaptation | |
| CN112464003B (en) | Image classification method and related device | |
| CN109410114B (en) | Compressed Sensing Image Reconstruction Algorithm Based on Deep Learning | |
| US20240078633A1 (en) | Image processing apparatus and method | |
| CN111914997B (en) | Method for training neural network, image processing method and device | |
| KR20200144398A (en) | Apparatus for performing class incremental learning and operation method thereof | |
| US11481608B2 (en) | Method and apparatus with neural network parameter quantization | |
| CN111291631B (en) | Video analysis method and related model training method, device and apparatus thereof | |
| US20210383203A1 (en) | Apparatus and method with neural network | |
| US11119507B2 (en) | Hardware accelerator for online estimation | |
| CN113066065A (en) | No-reference image quality detection method, system, terminal and medium | |
| Ren et al. | Robustness of transfer learning to image degradation | |
| WO2022179606A1 (en) | Image processing method and related apparatus | |
| WO2023207531A1 (en) | Image processing method and related device | |
| Bricman et al. | CocoNet: A deep neural network for mapping pixel coordinates to color values | |
| CN117710761A (en) | Quantum convolution neural network-based magnetic resonance image classification method and device | |
| KR102432766B1 (en) | Magnetic resonance image analysis system and method for alzheimer's disease classification | |
| CN113256556A (en) | Image selection method and device | |
| Khanzadi et al. | Robust fuzzy rough set based dimensionality reduction for big multimedia data hashing and unsupervised generative learning | |
| US20230148319A1 (en) | Method and device with calculation for driving neural network model | |
| Patel et al. | Deep Learning in Medical Image Super-Resolution: A Survey | |
| WO2022133814A1 (en) | Omni-scale convolution for convolutional neural networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |