Intelligent dynamic contrast enhanced magnetic resonance parameter imaging method and device

Abstract

The invention relates to the technical field of medical magnetic resonance imaging, and discloses an intelligent dynamic contrast enhanced magnetic resonance parameter imaging method and device.According to the method, a pixel-level visual converter is integrated at the connection position of an encoder and a decoder of a generator so as to enhance the learning ability of non-local modes and time sequence information in medical images; the parameter estimation precision in a noise variable environment can be improved. Gradient penalty is introduced into a loss function to optimize the performance of a discriminator, so that real and generated images can be distinguished more effectively. According to the CycleGAN-like model, spatial features can be fully extracted and utilized while time sequence analysis is carried out, a more comprehensive and fine PK parameter diagram is provided, and powerful technical support is provided for improving the accuracy of diagnosis of diseases such as cervical cancer.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical magnetic resonance imaging, in particular to an intelligent dynamic contrast-enhanced magnetic resonance parameter imaging method and device. BACKGROUND

[0002] Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is an advanced medical imaging technique that plays a crucial role in diagnosing and treatment planning for diseases such as cervical cancer. DCE-MRI involves repeatedly acquiring T1-weighted images after contrast agent injection and capturing the transport of the contrast agent within the region of interest based on pharmacokinetic (PK) models, such as the Tofts model and its extended versions (ETK). This non-invasive microvascular parameter quantification and analysis, such as the forward volume transfer constant (K trans ) and extracellular space volume (v e ), enables a deep understanding of the physiological characteristics of tissues.

[0003] In traditional DCE-MRI data analysis, solving these PK model parameters usually relies on methods such as nonlinear least squares (NLLS) and Bayesian estimation. Although these methods are effective in certain scenarios, they are sensitive to noise, computationally expensive, and require accurate estimation of the arterial input function (AIF), which often poses a challenge in practical applications. Therefore, these traditional methods may lead to inaccurate and inconsistent parameter estimates when dealing with large volumes or high-noise data.

[0004] In recent years, deep learning methods have shown great potential in handling DCE-MRI data. Various advanced deep learning architectures, such as convolutional neural networks (CNNs) and long short-term memory networks (LSTMs), have been used to automate and accurately estimate the PK parameters of DCE-MRI. CNNs are good at extracting spatial features, but cannot fully capture the dynamic information of time series data. LSTMs are suitable for processing time series data and learning long-term dependencies, but often ignore key spatial feature extraction in the process of mapping contrast agent concentration time curves and arterial input functions to PK parameters, thereby limiting the overall accuracy of parameter estimation. Therefore, deep learning methods still face challenges in handling complex spatiotemporal dependencies and high-dimensional spatial features. SUMMARY

[0005] The present application provides an intelligent dynamic contrast-enhanced magnetic resonance parameter imaging method to solve the problem that deep learning methods still face challenges in handling complex spatiotemporal dependencies and high-dimensional spatial features in the prior art.

[0006] Correspondingly, the application further provides an intelligent dynamic contrast-enhanced magnetic resonance parameter imaging device, an electronic device, and a computer readable storage medium, which are used to ensure the implementation and application of the method.

[0007] To solve the above technical problems, the application discloses an intelligent dynamic contrast-enhanced magnetic resonance parameter imaging method, which comprises the following steps:

[0008] A CycleGAN-like model is constructed, the CycleGAN-like model comprising a generator and a discriminator, the generator comprising an encoder and a decoder, and a pixel-level visual transformer being arranged at a connection position of the encoder and the decoder;

[0009] The dynamic contrast-enhanced magnetic resonance image sequence is input into the encoder to obtain an encoded image;

[0010] The pixel-level visual transformer is used to construct the encoded image into an image sequence, and a feature sequence is extracted based on the image sequence; the image sequence comprises pixel information and position information;

[0011] The feature sequence is input into the decoder to obtain a multi-parameter map;

[0012] The CycleGAN-like model uses a loss function to optimize the performance of the discriminator; the loss function is constructed based on a least squares GAN loss and a gradient penalty enhancement loss.

[0013] The application further discloses an intelligent dynamic contrast-enhanced magnetic resonance parameter imaging device, which comprises:

[0014] A model construction module is configured to construct a CycleGAN-like model; the CycleGAN-like model comprises a generator and a discriminator, the generator comprising an encoder and a decoder, and a pixel-level visual transformer being arranged at a connection position of the encoder and the decoder;

[0015] An encoding module is configured to input a dynamic contrast-enhanced magnetic resonance image sequence into the encoder to obtain an encoded image;

[0016] A time sequence learning module is configured to use a pixel-level visual transformer to construct the encoded image into an image sequence, and extract a feature sequence based on the image sequence; the image sequence comprises pixel information and position information;

[0017] A decoding module is configured to input the feature sequence into the decoder to obtain a multi-parameter map;

[0018] The CycleGAN-like model uses a loss function to optimize the performance of the discriminator; the loss function is constructed based on a least squares GAN loss and a gradient penalty enhancement loss.

[0019] The application further discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the method in one or more of the embodiments of the application when executing the program.

[0020] The application further discloses a computer readable storage medium, and the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method in one or more of the embodiments of the application.

[0021] In the application, a pixel-level visual transformer is integrated at the connection position of the encoder and the decoder of the generator to enhance the learning ability of non-local patterns and time sequence information in medical images, and the parameter estimation accuracy in a variable noise environment can be improved. Gradient penalty is introduced into the loss function to optimize the performance of the discriminator, so that it can more effectively distinguish between real and generated images. The CycleGAN model can fully extract and utilize spatial features while performing time series analysis, providing more comprehensive and detailed PK parameter maps, and providing strong technical support for improving the accuracy of cervical cancer and other disease diagnosis.

[0022] Additional aspects and advantages of the application will be described in the following description part, which will become apparent from the following description or be understood by practicing the application. BRIEF DESCRIPTION OF DRAWINGS

[0023] The above and / or additional aspects and advantages of the application will become apparent and more readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:

[0024] Figure 1 A flowchart of an intelligent dynamic contrast enhancement magnetic resonance parameter imaging method provided by an embodiment of the application is shown in the figure;

[0025] Figure 2 A Unet generator structure diagram of the integrated pixel-level visual transformer provided by the embodiment of the application is shown in the figure;

[0026] Figure 3 A pixel-level visual transformer module structure diagram provided by the embodiment of the application is shown in the figure;

[0027] Figure 4 A whole network structure diagram provided by the embodiment of the application is shown in the figure;

[0028] Figure 5 An experimental result diagram of cervical cancer DCE MRI parameter imaging provided by the embodiment of the application is shown in the figure;

[0029] Figure 6 A multi-parameter map profile analysis result diagram provided by the embodiment of the application is shown in the figure;

[0030] Figure 7A structural schematic diagram of an intelligent dynamic contrast-enhanced magnetic resonance parameter imaging device provided by an embodiment of the present application is shown in FIG. 1.

[0031] Figure 8 A structural schematic diagram of an electronic device provided by an embodiment of the present application is shown in FIG. 4. DETAILED DESCRIPTION

[0032] Embodiments of the present application are described in detail below with reference to the attached drawing figures, wherein the same or like reference numerals are used throughout the drawing figures to refer to the same or like elements or to like functions performed by like elements. The embodiments described below are illustrative of the present application, and are not meant to be limiting of the present application.

[0033] It can be understood by those skilled in the art that, unless specifically stated otherwise, singular forms such as "a," "an," and "the" are intended to include plural forms as well. It will be further understood that the terms "includes," "including," "comprises," and / or "comprising," when used in this specification, 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 can be present. In addition, the word "coupling" or "coupled" as used herein means the coupling or connection of two or more elements, which can either be direct or through intervening elements. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0034] Those skilled in the art can understand that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It should also be understood that the terms, such as those defined in a generally used dictionary, should be interpreted as having a meaning that is consistent with the meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless specifically so defined herein.

[0035] The scheme provided by the embodiments of the present application can be executed by any electronic device, which can be a terminal device or a server. The server can be a physical server, a server cluster composed of multiple physical servers or a distributed system, or a cloud server providing cloud computing services. The terminal can be a smartphone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited thereto. The terminal and the server can be connected directly or indirectly through wired or wireless communication, which is not limited in the present application. The intelligent dynamic contrast enhancement magnetic resonance parameter imaging method and device provided by the present application can solve at least one of the technical problems in the prior art.

[0036] The technical scheme of the present application and how the technical scheme solves the above technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described again in some embodiments. The embodiments of the present application will be described below with reference to the drawings.

[0037] The embodiments of the present application provide a possible implementation manner, as shown in Figure 1 A flowchart of an intelligent dynamic contrast enhancement magnetic resonance parameter imaging method is provided, which can be executed by any electronic device, and can be executed on a server side or a terminal device.

[0038] As shown in Figure 1 The method can include the following steps:

[0039] Step 101, constructing a CycleGAN-like model; the CycleGAN-like model includes a generator and a discriminator, the generator includes an encoder and a decoder, and the connection position of the encoder and the decoder is provided with a pixel-level visual transformer.

[0040] The embodiments of the present application propose a CycleGAN-like model, which is composed of a generator and a discriminator to form an adversarial generation network, so as to realize the transformation from a dynamic contrast enhancement magnetic resonance image sequence to a multi-parameter map. The generator is used to learn the mapping from the dynamic contrast enhancement magnetic resonance image sequence to the multi-parameter map, and then generate the multi-parameter map. The discriminator is used to determine whether the multi-parameter map generated by the generator is a real image.

[0041] In the generator, a pixel-level visual transformer is introduced to improve the ability of the model in non-local pattern recognition and time series information extraction. The structure of the encoder and the decoder can be a UNet structure, or a UNet structure with a convolution block attention module (CBAM), or a UNet structure with other attention mechanisms, to enhance the feature extraction ability of the model. The structure of the encoder and the decoder can also be other network structures, such as a deep residual network (ResNet) or a densely connected network (DenseNet), which are not specifically limited in the embodiments of the present application.

[0042] In step 102, a dynamic contrast-enhanced magnetic resonance image sequence is input into an encoder to obtain an encoded image.

[0043] In the embodiments of the present application, features in a dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) image are extracted by an encoder to obtain an encoded image.

[0044] In step 103, the encoded image is constructed into an image sequence by using a pixel-level visual transformer, and a feature sequence is extracted based on the image sequence; wherein the image sequence includes pixel information and position information.

[0045] Through the powerful time series processing capability of the pixel-level visual transformer, the parameter estimation accuracy of the model in a variable noise environment can be improved. And with the deep time series analysis function of the pixel-level visual transformer, the time series dynamics of the image sequence can be more comprehensively captured to realize more accurate PK parameter estimation, which is suitable for processing dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) sequences.

[0046] In step 104, the feature sequence is input into a decoder to obtain a multi-parameter map.

[0047] The feature sequence not only includes the spatial features extracted in the encoder, but also includes the analysis of the time series in the pixel-level visual transformer. By processing the feature sequence using the decoder, a more comprehensive and detailed PK parameter map can be output.

[0048] Among them, the CycleGAN-like model uses a loss function to optimize the performance of the discriminator; the loss function is constructed based on the least squares GAN loss and the gradient penalty enhancement loss.

[0049] The CycleGAN-like model in the embodiments of the present application introduces a gradient penalty into the loss function to optimize the performance of the discriminator, so that the discriminator can more effectively distinguish between real and generated images.

[0050] In the embodiments of the present application, a pixel-level visual transformer is integrated at the connection position of the encoder and the decoder of the generator to enhance the learning ability of non-local patterns and temporal information in medical images, thereby improving the parameter estimation accuracy in a variable noise environment. Gradient penalty is introduced into the loss function to optimize the performance of the discriminator, so that it can more effectively distinguish between real and generated images. The CycleGAN-like model can simultaneously analyze time series and fully extract and utilize spatial features, providing more comprehensive and detailed PK parameter maps, and providing strong technical support for improving the accuracy of cervical cancer and other disease diagnosis.

[0051] In an optional embodiment, the encoder and the decoder are UNet encoding-decoding structures, and the encoder includes a plurality of down-sampling encoding blocks;

[0052] The dynamic contrast-enhanced magnetic resonance image sequence is input into the encoder to obtain an encoded image, including:

[0053] The dynamic contrast-enhanced magnetic resonance image sequence is preprocessed to convert it into a three-dimensional tensor;

[0054] The three-dimensional tensor is dimensionally reduced layer by layer using a plurality of down-sampling encoding blocks to obtain an encoded image.

[0055] As shown in Figure 2 The down-sampling encoding block includes a basic convolution block and a down-sampling convolution block with a step size of 2. The input dynamic contrast-enhanced magnetic resonance image sequence is first converted into a three-dimensional tensor through a preprocessing layer, and then dimensionally reduced layer by layer through alternating basic convolution blocks and down-sampling convolution blocks with a step size of 2 in the encoder until an encoded image with a size and dimension suitable for ViT (Vision Transformer) processing is obtained. In the embodiments of the present application, the width w and the height h are halved each time the down-sampling is performed, so as to gradually reduce the image size and increase the depth f of the features. For each down-sampling convolution block, the feature dimension is doubled to better capture complex image features and form a feature representation with sufficient information for subsequent pixel-level ViT processing.

[0056] Exemplarily, the encoder includes four down-sampling encoding blocks, and the size of the dynamic contrast-enhanced magnetic resonance image sequence is (256, 256, 40). The three-dimensional tensor (w0=h0=256, f0=48) after preprocessing is input into the basic convolution block of the first down-sampling encoding block to obtain a feature with a size of (256, 256, 48); after the down-sampling processing of the down-sampling convolution block D4 of the first down-sampling encoding block, the feature is input into the basic convolution block A feature with a size of (128, 128, 96) is obtained; after the feature is down-sampled by a down-sampling convolutional block D3 of a third down-sampling encoding block, the feature is input into a basic convolutional block of a fourth down-sampling encoding block A feature with a size of (64, 64, 192) is obtained; after the feature is down-sampled by a down-sampling convolutional block D2 of a third down-sampling encoding block, the feature is input into a basic convolutional block of a fourth down-sampling encoding block A feature with a size of (32, 32, 384) is obtained; after the feature is down-sampled by a down-sampling convolutional block D1 of a fourth down-sampling encoding block, a feature with a size (w = h = 16, f0 = 384) suitable for VIT processing is obtained.

[0057] In an optional embodiment, the pixel-level vision transformer comprises a Transformer encoder.

[0058] The encoded image is constructed into an image sequence by using the pixel-level vision transformer, and a feature sequence is extracted based on the image sequence, comprising:

[0059] A token sequence is constructed in units of each pixel point in the encoded image.

[0060] Position information of each pixel point in the token sequence is generated by means of two-dimensional Fourier position embedding.

[0061] An image sequence is generated based on pixel information and position information of each pixel point in the token sequence.

[0062] The Transformer encoder is used to extract features of the image sequence, and a feature sequence is obtained.

[0063] In order to accurately extract the time sequence and spatial features in the dynamic contrast-enhanced magnetic resonance image sequence, the pixel-level vision transformer (Vision Transformer, ViT) is introduced at the end of the encoder in the embodiment of the application, and the structure is as shown in Figure 3 This module constructs a token sequence representing the entire image sequence in units of each pixel point, each token carrying pixel-level information, and the token sequence also enhances the recognition ability of the model for the spatial position of the image through two-dimensional Fourier position embedding, thereby generating an image sequence with spatial features and position features. Through this design, the model can capture more fine local details and understand the time sequence changes of the image at a global level, thereby improving the generation quality of the multi-parameter map.

[0064] In an optional embodiment, the Transformer encoder is used to extract features of the image sequence, and a feature sequence is obtained, comprising:

[0065] The image sequence is processed multiple times by using residual connection and continuous layer normalization operation to obtain a feature sequence.

[0066] Optionally, as shown in Figure 3 ViT includes a position embedding layer, a linear layer and a Transformer encoder (i.e., a transformer encoding module). The Transformer encoder is provided with a linear layer before and after it. ViT first changes the shape of the image, i.e., flattens the image along the spatial dimension to form a token sequence with a length of 256, and each token in the sequence is a vector with a length of f. Then, the position embedding layer is used for position embedding to obtain an image sequence with spatial features and temporal features with a dimension of f+f p , and then the dimension of the result is linearly mapped to f v by the first linear layer, and the result is input into the Transformer encoder. The Transformer encoder includes a normalization layer, a multi-head self-attention mechanism and a feedforward neural network. The image sequence input into the first normalization layer after the first linear layer is subjected to layer normalization processing and then input into the multi-head self-attention mechanism to capture features in the image sequence. The features output by the multi-head self-attention mechanism are subjected to layer normalization processing by the second normalization layer, and then input into the feedforward neural network for nonlinear mapping to obtain a feature sequence. The feature sequence output by the Transformer encoder is linearly converted by the second linear layer to linearly project the output result back to the dimension f, and then output after changing the shape. The input of the first normalization layer is connected to the input of the second normalization layer by residual connection. The input of the second normalization layer is connected to the input of the second linear layer by residual connection. In the embodiment of the present application, the Transformer encoder can process the image sequence multiple times to enhance the ability of the model to capture sequence data features. Specifically, the image sequence can be processed multiple times by residual connection and continuous layer normalization operation.

[0067] In the embodiment of the present application, ReZero regularization and its learnable parameter a can be used to optimize the residual connection to enhance the stability of model training.

[0068] Optionally, other regularization techniques (such as LayerNormalization or BatchNormalization regularization techniques) can also be used in the embodiment of the present application to test the influence of different regularization methods on the training effect of the model.

[0069] In an optional embodiment, the decoder includes a plurality of up-sampling encoding blocks, and the up-sampling encoding blocks are connected to the corresponding down-sampling encoding blocks in the decoder by skip connection.

[0070] The feature sequence is input into the decoder to obtain a multi-parameter graph, including:

[0071] Based on the feature sequence and the features of the corresponding down-sampling encoding block of each up-sampling encoding block, the plurality of up-sampling encoding blocks are processed to obtain a multi-parameter map.

[0072] As shown in Figure 2 The up-sampling encoding block includes an up-sampling convolution block with a step of 2 and a basic convolution block. After the feature is up-sampled by the up-sampling convolution block, it is input into the basic convolution block. The input of the basic convolution block in the up-sampling encoding block is connected to the output of the basic convolution block in the down-sampling encoding block by a skip connection to fuse the features in the corresponding down-sampling encoding block. In the embodiments of the present application, the decoder is allowed to access the intermediate features in the encoder stage, which can improve the detail preservation capability in the image reconstruction process, so that the final output can accurately reconstruct the image through the post-processing layer.

[0073] For example, in combination with the example of the encoder part, the decoder includes four up-sampling encoding blocks. The size of the features in the feature sequence input after VIT processing is (16, 16, 384). The feature sequence is input into the up-sampling convolution block U1 of the first up-sampling encoding block for up-sampling processing to obtain a feature with a size of (32, 32, 384). The feature is combined with the feature output by the basic convolution block of the fourth down-sampling encoding block through a skip connection, and then input into the up-sampling convolution block of the first up-sampling encoding block for processing; the up-sampling convolution block outputs a feature with a size of (64, 64, 192). The feature is combined with the feature output by the basic convolution block of the third down-sampling encoding block through a skip connection, and then input into the up-sampling convolution block of the second up-sampling encoding block for processing; the up-sampling convolution block outputs a feature with a size of (128, 128, 96). The feature is combined with the feature output by the basic convolution block of the second down-sampling encoding block through a skip connection, and then input into the up-sampling convolution block of the third up-sampling encoding block for processing; the up-sampling convolution block outputs a feature with a size of (256, 256, 48). The feature is combined with the feature output by the basic convolution block of the fourth down-sampling encoding block through a skip connection, and then input into the up-sampling convolution block of the fourth up-sampling encoding block for processing; the up-sampling convolution block outputs a feature with a size of (512, 512, 24). ​​The data is processed to obtain a decoded feature with dimensions (256, 256, 48). This decoded feature is then transformed to obtain a multi-parameter map with dimensions (256, 256, 2).

[0074] In an optional embodiment, such as Figure 4 As shown, the generator includes a first generator A and a second generator B, and the discriminator includes a first discriminator. Second discriminator D B ;

[0075] The first generator A is used to convert a dynamic contrast-enhanced magnetic resonance image sequence into a multi-parameter map, and the second generator B is used to convert the multi-parameter map into a dynamic contrast-enhanced magnetic resonance image sequence.

[0076] First discriminator The second discriminator D is used to determine the authenticity of the converted dynamic contrast-enhanced magnetic resonance image sequence. B Used to determine the authenticity of the converted multi-parameter graph;

[0077] like Figure 4 As shown, the CycleGAN-like model interweaves two generator-discriminator pairs for unpaired image-to-image transformation. Let A and B represent two image domains, and the CycleGAN-like model uses a generator G... A→B Transform the image from A to B using generator G. B→A The image is transformed from B to A by the discriminator D. A Used to distinguish the image in A from the image converted from B. f Discriminator D B This is used to distinguish the image in B from the image converted from A. f The CycleGAN-like model utilizes identity loss and cycle consistency loss to ensure the fidelity of image content during the transformation process. Here, A represents the dynamic contrast-enhanced magnetic resonance image sequence domain, and B represents the multi-parameter image domain. Generator G... A→B As the first generator, it realizes the generation of multi-parameter images from a dynamic contrast-enhanced magnetic resonance image sequence domain. Generator G B→A The second generator enables the generation of dynamic contrast-enhanced magnetic resonance image sequences from a multi-parameter image domain.

[0078] Taking the second discriminator as an example, the loss function of the second discriminator is:

[0079]

[0080] in,

[0081]

[0082] in, Loss for the second discriminator; Loss representing the multi-parametric map obtained by converting from domain A corresponding to the dynamic contrast-enhanced magnetic resonance image sequence to domain B corresponding to the multi-parametric map, Loss representing the real multi-parametric map in domain B corresponding to the multi-parametric map; GP is a weight of the gradient penalty term, and γ is a target gradient size, represents the gradient of x, represents the expected value; represents the first discriminator; D B represents the second discriminator; represents the multi-parametric map obtained by converting from domain A corresponding to the dynamic contrast-enhanced magnetic resonance image sequence to domain B corresponding to the multi-parametric map.

[0083] The loss function of the first discriminator refers to the loss function of the second discriminator, and the calculation methods of the two are the same.

[0084] The discriminator loss function combines the LSGAN loss to reduce the oscillation in the training process, optimize the image quality after reconstruction, and especially perform better than the traditional GAN loss function in preserving image details. The introduction of the gradient penalty term in the loss function helps to balance the relationship between the generator and the discriminator during the training process, so that the model training is more stable, and the potential training oscillation is reduced. This loss function design aims to improve the performance of the discriminator in distinguishing real and generated images, thereby encouraging the generator to produce higher quality images and ensuring high consistency between the images and the real situation. Further, the model can generate multi-parametric maps that are more realistic and accurate in both visual and statistical characteristics.

[0085] In the embodiments of the present application, through the above loss function and in combination with the pixel-level ViT and the encoding-decoding architecture, efficient and accurate mapping from the dynamic contrast-enhanced magnetic resonance image sequence to the multi-parametric map is successfully achieved, and the parameter estimation capability of the dynamic contrast-enhanced MRI is significantly improved.

[0086] Optionally, other types of loss functions such as Wasserstein GAN loss or conditional adversarial network (Conditional GAN) loss can also be used in the embodiments of the present application to further improve the image quality and stability of the model.

[0087] In an optional embodiment, after constructing the CycleGAN-like model, the method further comprises:

[0088] The paired image blocks are extracted from the pre-obtained dynamic contrast-enhanced magnetic resonance image sequence and corresponding multi-parametric map data set, input into the CycleGAN-like model, and the Adam optimizer is used to optimize the loss function to obtain the optimized CycleGAN-like model.

[0089] Optionally, the discriminator can be optimized by introducing other forms of regularization or deeper network architecture to improve its ability to identify details, which is not limited in the embodiments of the present application.

[0090] The method in the embodiments of the present application is verified by experiments based on the cervical cancer DCE MRI sequence data set, and the experimental results are as shown in Figure 5

[0091] Among them, Figure 5 The first row is the experimental results of the method in the embodiments of the present application (i.e. Figure 5 the experimental results of the present application in), which respectively shows the generated K trans parameter map and its ROI zoomed-in view, the generated v e parameter map and its ROI zoomed-in view; the second row is the corresponding real reference image.

[0092] For K trans real image and v e real image profile analysis, the results are as shown in Figure 6 The left side shows the generated K trans real image and v e real image, where the white line marks the specific profile analyzed. The deep learning methods used for comparison with the embodiments of the present application include Pix2Pix model, CycleGAN model, UNet model and AttUnet model. The right side chart shows the parameter value changes along the white profile line position (from left to right), where the blue curve Ours represents the analysis results of the parameter map generated by the method in the embodiments of the present application, and the red curve GroundTruth represents the real results. The area framed by the red dotted line in the figure highlights the specific distribution of the profile values in the cervical and uterine regions (ROI), verifying the accuracy of the method in the embodiments of the present application in the key regions.

[0093] It should be noted that although the embodiments of the present application mainly aim at DCE-MRI parameter mapping, it is also applicable to other medical imaging techniques and disease types:

[0094] 1) Parameter image generation for other diseases: The method of the embodiments of the present application can be applied to parameter image generation for other diseases such as heart disease, brain tumor, etc., providing support for accurate diagnosis and treatment planning under different pathological conditions.

[0095] 2) Parameter generation for other modal medical images: For example, for positron emission tomography (PET) or computed tomography (CT) images, the embodiments of the present application can also be used to generate parameter images reflecting the functional state or pathological changes of tissues.

[0096] 3) Cross-modality medical image conversion: The model of the embodiments of the present application can be used for image conversion between different medical imaging modalities, for example, converting CT images into dynamic contrast-enhanced magnetic resonance images, or vice versa, thereby providing more rich clinical information.

[0097] The method in the embodiments of the present application can also be applied outside medical imaging and can be applied to image-to-image conversion tasks in other fields, such as satellite image processing, artistic style conversion, etc.

[0098] The method in the embodiments of the present application is also applicable to multi-modality imaging fusion: The method in the embodiments of the present application is also applicable to conversion between different imaging modalities, such as image fusion of PET and CT, MRI and CT, etc., to obtain more rich diagnostic information.

[0099] The method in the embodiments of the present application is applicable to image processing in non-medical fields: The improved design of the CycleGAN-like model can be applied to image processing tasks in non-medical fields, such as video frame interpolation, high-resolution imaging, and real-time image recognition systems in autonomous vehicles.

[0100] Based on the same principle as the method provided in the embodiments of the present application, the embodiments of the present application also provide an intelligent dynamic contrast-enhanced magnetic resonance parameter imaging device, as shown in Figure 7 The device comprises:

[0101] The model construction module 701 is configured to construct a CycleGAN-like model; the CycleGAN-like model comprises a generator and a discriminator, the generator comprises an encoder and a decoder, and a pixel-level visual transformer is arranged at a connection position of the encoder and the decoder;

[0102] The encoding module 702 is configured to input the dynamic contrast-enhanced magnetic resonance image sequence into the encoder to obtain an encoded image;

[0103] The time sequence learning module 703 is configured to construct the encoded image into an image sequence by using the pixel-level visual transformer, and extract a feature sequence based on the image sequence; the image sequence comprises pixel information and position information;

[0104] The decoding module 704 is configured to input the feature sequence into the decoder to obtain a multi-parameter map;

[0105] The CycleGAN-like model uses a loss function to optimize the performance of the discriminator; the loss function is constructed based on a least squares GAN loss and a gradient penalty enhancement loss.

[0106] In the embodiments of the present application, a pixel-level visual transformer is integrated at the connection position of the encoder and the decoder of the generator to enhance the learning ability of non-local patterns and time sequence information in medical images, which can improve the parameter estimation accuracy in a variable noise environment. Gradient penalty is introduced into the loss function to optimize the performance of the discriminator, so that it can more effectively distinguish between real and generated images. The CycleGAN-like model can fully extract and utilize spatial features while performing time series analysis, providing more comprehensive and detailed PK parameter maps, and providing strong technical support for improving the accuracy of disease diagnosis such as cervical cancer.

[0107] The intelligent dynamic contrast-enhanced magnetic resonance parameter imaging device provided by the embodiments of the present application can realize Figures 1 to 6 The various processes realized in the method embodiments are not repeated here to avoid repetition.

[0108] The intelligent dynamic contrast-enhanced magnetic resonance parameter imaging device of the embodiments of the present application can execute the intelligent dynamic contrast-enhanced magnetic resonance parameter imaging method provided by the embodiments of the present application, and the implementation principles are similar. The actions performed by each module and unit in the intelligent dynamic contrast-enhanced magnetic resonance parameter imaging device in the embodiments of the present application correspond to the steps in the intelligent dynamic contrast-enhanced magnetic resonance parameter imaging method in the embodiments of the present application. For detailed descriptions of the functions of each module of the intelligent dynamic contrast-enhanced magnetic resonance parameter imaging device, please refer to the descriptions of the corresponding intelligent dynamic contrast-enhanced magnetic resonance parameter imaging method provided in the foregoing embodiments, which will not be repeated here.

[0109] Based on the same principles as the methods shown in the embodiments of the present application, the embodiments of the present application also provide an electronic device, which can include but is not limited to: a processor and a memory; the memory is used to store a computer program; the processor is used to execute the intelligent dynamic contrast-enhanced magnetic resonance parameter imaging method shown in any optional embodiment of the present application by calling the computer program. Compared with the prior art, the intelligent dynamic contrast-enhanced magnetic resonance parameter imaging method provided by the present application integrates a pixel-level visual transformer at the connection position of the encoder and the decoder of the generator to enhance the learning ability of non-local patterns and time sequence information in medical images, which can improve the parameter estimation accuracy in a variable noise environment. Gradient penalty is introduced into the loss function to optimize the performance of the discriminator, so that it can more effectively distinguish between real and generated images. The CycleGAN-like model can fully extract and utilize spatial features while performing time series analysis, providing more comprehensive and detailed PK parameter maps, and providing strong technical support for improving the accuracy of disease diagnosis such as cervical cancer.

[0110] In an optional embodiment, an electronic device is also provided, as shown in Figure 8 Figure 8 ​The electronic device 800 shown can be a server, comprising a processor 801 and a memory 803. The processor 801 and the memory 803 are connected, for example, through a bus 802. Optionally, the electronic device 800 can further comprise a transceiver 804. It should be noted that the transceiver 804 is not limited to one in actual application, and the structure of the electronic device 800 does not constitute a limitation to the embodiments of the present application.

[0111] The processor 801 can be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), a FPGA (Field Programmable Gate Array) or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It can implement or execute various exemplary logical blocks, modules and circuits described in combination with the disclosure. The processor 801 can also be a combination of computing functions, such as one or more microprocessor combinations, combinations of DSP and microprocessor, etc.

[0112] The bus 802 can include a path for transmitting information between the above-mentioned components. The bus 802 can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 802 can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience, Figure 8 In the figure, only one thick line is used to represent the bus, but it does not mean that there is only one bus or only one type of bus.

[0113] The memory 803 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions; a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions; an EEPROM (Electrically Erasable Programmable Read-Only Memory), a CD-ROM (Compact Disc Read-Only Memory) or other optical disk storage, a magnetic disk storage or other magnetic storage devices, or any other medium capable of storing desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto.

[0114] The memory 803 is configured to store application program codes for implementing the solutions of the present application, and the processor 801 is configured to control the execution of the application program codes. The processor 801 is configured to execute the application program codes stored in the memory 803 to implement the content shown in the foregoing method embodiments.

[0115] The electronic device includes, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a digital broadcast receiver, a PDA (Personal Digital Assistant), a PAD (Tablet Personal Computer), a PMP (Portable Multimedia Player), a vehicle terminal (e.g., a car navigation terminal), and the like, and a stationary terminal such as a digital TV, a desktop computer, and the like. Figure 8 The electronic device shown is merely an example and should not impose any limitation on the functions and use range of the embodiments of the present application.

[0116] The server provided by the present application can be a stand-alone physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDNs, and basic cloud computing services such as big data and artificial intelligence platforms. The terminal can be a smartphone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, and the like, but is not limited thereto. The terminal and the server can be connected directly or indirectly through wired or wireless communication, which is not limited in the present application.

[0117] The embodiments of the present application provide a computer readable storage medium, which stores a computer program. When the computer program runs on a computer, the computer can execute the corresponding content in the foregoing method embodiments.

[0118] It should be understood that although the steps in the flowcharts of the drawings are shown in a sequential order, such processes can be practiced with the steps in different orders. Further, at least some of the steps in the flowcharts can include multiple sub-steps or multiple stages, which can be performed at different times or in different orders, and not necessarily in the order shown.

[0119] It should be understood that the computer-readable storage medium in the above application can also be a computer-readable signal medium or a combination of a computer-readable storage medium and a computer-readable signal medium. The computer-readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device or apparatus. In this application, the computer-readable signal medium can include a data signal carried in a baseband or as a part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take many forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium that can send, propagate or transmit a program for use by or in connection with an instruction execution system, device or apparatus. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to a wire, an optical fiber, an RF (radio frequency) or the like, or any suitable combination of the above.

[0120] The above computer-readable medium can be included in the above electronic device; or can exist separately and not be assembled into the electronic device.

[0121] The above computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the method shown in the above embodiments.

[0122] According to an aspect of the present application, a computer program product or computer program is provided, which comprises computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the intelligent dynamic contrast enhancement magnetic resonance parameter imaging method and device provided in the various optional implementation manners described above.

[0123] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object-oriented programming language such as Java, Smalltalk, C++, or conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on a user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0124] The flow diagrams and the block diagrams in the drawings are illustrations of possible architectures, functions, and operations for systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flow diagrams or block diagrams can represent a module, a segment, or a portion of code, which comprises one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flow diagrams, and combinations thereof, can be implemented by special purpose hardware-based systems that perform the specified functions or operations, or combinations of special purpose hardware and computer instructions.

[0125] The modules involved in the embodiments of the present application can be implemented in a software manner or in a hardware manner. Among them, the name of the module does not constitute a limitation to the module itself in some cases. For example, the model construction module can also be described as "a model construction module for constructing a CycleGAN model".

[0126] The above description is merely exemplary of the application and the principles thereof. It is to be understood that those skilled in the art will be able to devise various arrangements which, although not explicitly described or shown herein, embody the principles of the application and are included within its spirit and scope. Furthermore, there are several variations to the application described herein which have not been described but will be understood by those skilled in the art. For example, the features of the application described and shown can be combined with other features of the application described and shown (but not limited to) in the patent specification and drawings.

Claims

1. A method for intelligent dynamic contrast-enhanced magnetic resonance parametric imaging, characterized in that, The method includes: Construct a CycleGAN-like model; the CycleGAN-like model includes a generator and a discriminator, the generator includes an encoder and a decoder, and a pixel-level visual transformer is set at the connection position of the encoder and the decoder; The encoder is used to input a sequence of dynamic contrast-enhanced magnetic resonance images to obtain coded images; The coded image is constructed into an image sequence using the pixel-level visual transformer, and a feature sequence is extracted based on the image sequence; the image sequence includes pixel information and position information. The feature sequence is input into the decoder to obtain a multi-parameter map; The CycleGAN-like model uses a loss function to optimize the performance of the discriminator; the loss function is constructed based on least squares GAN loss and gradient penalty enhancement loss.

2. The intelligent dynamic contrast-enhanced magnetic resonance parametric imaging method according to claim 1, characterized in that, The encoder and decoder are UNet encoder-decoder structures, and the encoder includes multiple downsampled coding blocks; The step of inputting a dynamic contrast-enhanced magnetic resonance image sequence into the encoder to obtain an encoded image includes: The dynamic contrast-enhanced magnetic resonance image sequence is preprocessed and converted into a three-dimensional tensor. The 3D tensor is reduced in dimensionality layer by layer using multiple downsampling coding blocks to obtain the coded image.

3. The intelligent dynamic contrast-enhanced magnetic resonance parametric imaging method according to claim 1, characterized in that, The pixel-level visual transformer includes a Transformer encoder; The step of constructing an image sequence from the encoded image using the pixel-level visual transformer and extracting a feature sequence based on the image sequence includes: Construct a token sequence using each pixel in the encoded image as a unit; The position information of each pixel in the token sequence is generated by two-dimensional Fourier position embedding. The image sequence is generated based on the pixel information of each pixel in the token sequence and the position information; The feature sequence is obtained by extracting features from the image sequence using the Transformer encoder.

4. The intelligent dynamic contrast-enhanced magnetic resonance parametric imaging method according to claim 1, characterized in that, The image sequence is extracted using the Transformer encoder to obtain the feature sequence, including: The image sequence is processed multiple times using residual connections and consecutive layer normalization operations to obtain the feature sequence.

5. The intelligent dynamic contrast-enhanced magnetic resonance parametric imaging method according to claim 2, characterized in that, The decoder includes multiple upsampled coding blocks, and the upsampled coding blocks are skipped connections with the corresponding downsampled coding blocks in the decoder. The step of inputting the feature sequence into the decoder to obtain a multi-parameter map includes: Based on the feature sequence and the features of the downsampled coding block corresponding to each upsampled coding block, the multi-parameter map is obtained by processing multiple upsampled coding blocks.

6. The intelligent dynamic contrast-enhanced magnetic resonance parametric imaging method according to claim 1, characterized in that, The generator includes a first generator and a second generator, and the discriminator includes a first discriminator and a second discriminator; The first generator is used to convert a dynamic contrast-enhanced magnetic resonance image sequence into a multi-parameter map, and the second generator is used to convert the multi-parameter map into a dynamic contrast-enhanced magnetic resonance image sequence. The first discriminator is used to determine the authenticity of the converted dynamic contrast-enhanced magnetic resonance image sequence, and the second discriminator is used to determine the authenticity of the converted multi-parameter map; The loss function of the second discriminator is: in, in, This is the loss of the second discriminator; This represents the loss for transforming the multiparameter map obtained from domain A corresponding to a dynamic contrast-enhanced magnetic resonance image sequence to domain B corresponding to the multiparameter map. λ represents the loss of the true multiparameter graph in domain B corresponding to the multiparameter graph; GP Here, γ represents the weight of the gradient penalty term, and γ is the magnitude of the target gradient. Represents the gradient with respect to x. Indicates the expected value; D represents the first discriminator; B This represents the second discriminator; This indicates a transformation from domain A corresponding to a dynamic contrast-enhanced magnetic resonance image sequence to domain B corresponding to a multi-parameter map.

7. The intelligent dynamic contrast-enhanced magnetic resonance parametric imaging method according to claim 1, characterized in that, After constructing the CycleGAN-like model, the method further includes: Paired image patches are extracted from a pre-obtained dynamic contrast-enhanced magnetic resonance image sequence and the corresponding multi-parameter map dataset and input into the CycleGAN-like model. The loss function is then optimized using the Adam optimizer to obtain the optimized CycleGAN-like model.

8. An intelligent dynamic contrast-enhanced magnetic resonance parametric imaging device, characterized in that, The device includes: A model building module is used to build a CycleGAN-like model; the CycleGAN-like model includes a generator and a discriminator, the generator includes an encoder and a decoder, and a pixel-level visual transformer is set at the connection position of the encoder and the decoder; The encoding module is used to input the dynamic contrast-enhanced magnetic resonance image sequence into the encoder to obtain an encoded image; The temporal learning module is used to construct an image sequence from the encoded image using the pixel-level visual transformer, and to extract a feature sequence based on the image sequence; the image sequence includes pixel information and position information. The decoding module is used to input the feature sequence into the decoder to obtain a multi-parameter map; The CycleGAN-like model uses a loss function to optimize the performance of the discriminator; the loss function is constructed based on least squares GAN loss and gradient penalty enhancement loss.

9. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 7.

Images