Denoising method and system for low-dose CT image and electronic equipment

Abstract

The invention relates to the field of image processing, and provides a denoising method and system for a low-dose CT image and electronic equipment, and the method comprises the steps: obtaining a target image which is a to-be-processed low-dose CT image; inputting the target image into a pre-trained dual-channel neural network model for de-noising processing to obtain an output result; wherein the dual-channel neural network model comprises a first channel and a second channel, the first channel is a U-Net architecture formed based on a plurality of sparse Transform blocks, and the second channel is a U-Net architecture formed based on a plurality of residual attention modules; the output result is obtained through convolution processing after the first output of the first channel and the second output of the second channel are fused. According to the technical scheme, the low-dose CT image denoising method and device are used for overcoming the defects that low-dose CT images are large in noise and fuzzy in image detail, automatic denoising can be carried out on the low-dose CT images through the neural network, and the denoising effect is better.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a denoising method and system for low-dose CT images, and electronic equipment. Background Art

[0002] Computed Tomography (CT) technology plays a vital role in modern medical diagnosis. CT technology uses X-ray scanning to reconstruct cross-sectional images of objects and has become an important tool for non-invasive diagnosis.

[0003] Since CT technology uses X-rays, it may have certain effects on the human body. For infants or pregnant women, low-dose CT images become the choice of these groups. Low-dose CT reduces radiation by reducing X-rays, but it will also lead to a decrease in the number of photons received by the detector, thereby increasing the image noise. The noise sources of low-dose CT images include the reduction in radiation dose, the sensitivity of the detector, the working state of electronic components, and external mechanical vibrations. These factors may cause blurred image details and affect the accurate identification of lesions. Summary of the invention

[0004] The present invention provides a low-dose CT image denoising method and system, and electronic equipment, which are used to solve the defects of large noise and blurred image details in low-dose CT images. The solution of the present application can automatically denoise low-dose CT images through a neural network, and the denoising effect is better.

[0005] The present invention provides a low-dose CT image denoising method, comprising: Acquiring a target image, wherein the target image is a low-dose CT image to be processed; Inputting the target image into a pre-trained dual-channel neural network model for denoising to obtain an output result; Among them, the dual-channel neural network model includes a first channel and a second channel, the first channel is a U-Net architecture based on several sparse Transformer blocks, and the second channel is a U-Net architecture based on several residual attention modules; the output result is the first output of the first channel and the second output of the second channel fused together and obtained through convolution processing.

[0006] According to the denoising method for low-dose CT images provided by the present invention, the first channel includes a first sparse Transformer block, a second sparse Transformer block, a third sparse Transformer block, a fourth sparse Transformer block and a fifth sparse Transformer block, wherein the first sparse Transformer block, the second sparse Transformer block and the third sparse Transformer block constitute an encoder, and the fourth sparse Transformer block and the fifth sparse Transformer block constitute a decoder; The second channel includes a first residual attention module, a second residual attention module, a third residual attention module, a fourth residual attention module and a fifth residual attention module, wherein the first residual attention module, the second residual attention module and the third residual attention module constitute an encoder, and the fourth residual attention module and the fifth residual attention module constitute a decoder.

[0007] According to the denoising method for low-dose CT images provided by the present invention, the sparse Transformer block includes a mixed-scale feedforward network layer and a Top-K sparse attention layer; The mixed-scale feedforward network layer is used to capture and integrate multi-scale features and identify noise; The Top-K sparse attention layer is used to retain the feature information of the target image and reduce the interference of irrelevant information on the noise reduction process.

[0008] According to the low-dose CT image denoising method provided by the present invention, the residual attention module includes a residual block and a spatial attention module; The residual block includes several standard convolutional layers and activation function layers, and the residual block is used to extract local features; The spatial attention module is used to extract statistical information of the target image through global maximum pooling and global average pooling to supplement the global feature information.

[0009] According to the low-dose CT image denoising method provided by the present invention, the training process of the dual-channel neural network model includes: Inputting the training set data into a pre-built initial model to obtain a first output image, wherein the training set data includes a low-dose CT image and a standard-dose CT image, and the first output image is a denoised image of the low-dose CT image; Calculating the similarity between the first output image and the standard dose CT image by using a loss function; The initial model is updated based on the similarity.

[0010] According to the low-dose CT image denoising method provided by the present invention, the loss function includes reconstruction loss, perception loss and structural similarity loss; The reconstruction loss is used to calculate the pixel value similarity between the first output image and the standard dose CT image; The perceptual loss is used to calculate the feature similarity between the first output image and the standard dose CT image; The structural similarity loss is used to calculate the structural similarity between the first output image and the standard-dose CT image.

[0011] According to the low-dose CT image denoising method provided by the present invention, the loss function is a weighted sum of reconstruction loss, perception loss and structural similarity loss.

[0012] According to the low-dose CT image denoising method provided by the present invention, the target image is input into a pre-trained dual-channel neural network model for denoising to obtain an output result, and then the method further includes: The denoised image is evaluated using the preset evaluation index. If the evaluation fails, the denoising process is performed again.

[0013] The present invention also provides a low-dose CT image denoising system, comprising: An image loading module, used for loading a target image, wherein the target image is a low-dose CT image to be processed; A denoising module, used for inputting the target image into a pre-trained dual-channel neural network model for denoising to obtain an output result; Among them, the dual-channel neural network model includes a first channel and a second channel, the first channel is a U-Net architecture based on several sparse Transformer blocks, and the second channel is a U-Net architecture based on several residual attention modules; the output result is the first output of the first channel and the second output of the second channel fused together and obtained through convolution processing.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the above-mentioned low-dose CT image denoising methods is implemented.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the computer program implements any of the above-mentioned low-dose CT image denoising methods.

[0016] The present invention also provides a computer program product, comprising a computer program, which implements any of the above-mentioned low-dose CT image denoising methods when executed by a processor.

[0017] In the low-dose CT image denoising method provided by the present invention, the low-dose CT image can be automatically denoised based on a pre-trained neural network. At the same time, the applied neural network is a dual-channel neural network model, which integrates the advantages of high denoising efficiency of the sparse Transformer mechanism and high denoising accuracy of the residual attention mechanism. Furthermore, the first channel and the second channel are both U-Net architectures, which can exert excellent image data analysis effects even when the amount of data is small. Especially when applied to the CT image denoising field of the present application, accurate image denoising can be achieved with very little data, providing doctors with accurate diagnostic basis. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 is a flow chart of a method for denoising a low-dose CT image provided by an embodiment of the present invention; Figure 2 is a schematic diagram of the structure of a dual-channel neural network model provided by an embodiment of the present invention; Figure 3 is a schematic diagram of the structure of a sparse Transformer block provided by an embodiment of the present invention; Figure 4 is a schematic diagram of the structure of the residual attention module provided by an embodiment of the present invention; Figure 5 is a structural schematic diagram of a low-dose CT image denoising system provided by an embodiment of the present invention; Figure 6 It is a schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] Figure 1 It is a flowchart of a low-dose CT image denoising method provided by an embodiment of the present invention.

[0022] like Figure 1 As shown, this embodiment provides a low-dose CT image denoising method, comprising: Step 101, acquiring a target image, wherein the target image is a low-dose CT image to be processed; Step 102, inputting the target image into a pre-trained dual-channel neural network model for denoising to obtain an output result; Among them, the dual-channel neural network model includes a first channel and a second channel, the first channel is a U-Net architecture based on several sparse Transformer blocks, and the second channel is a U-Net architecture based on several residual attention modules; the output result is the first output of the first channel and the second output of the second channel fused together and obtained through convolution processing.

[0023] like Figure 2 As shown in the figure, the target image input into the dual-channel neural network model can first be processed by the Sobel operator (Sobel) to obtain an intermediate image. After the edge features are extracted by Sobel, the intermediate image can be cascaded and spliced ​​with the original image information. The cascaded information is then passed through a convolution layer for initial feature projection and then transmitted to the first channel and the second channel. This convolution layer can be a 3×3 convolution kernel, and the feature projection process can be 2→96 channels, that is, the 2-channel projection of the original image information is converted into 96 channels.

[0024] In practical applications, such as Figure 2 As shown, the first channel and the second channel can both be U-Net architectures composed of 5-level modules. The encoder in the U-Net architecture performs two downsampling operations through 3×3 convolution downsampling with a stride of 2, and the decoder restores the resolution through two 3×3 transposed convolutions with a stride of 2, and performs jump connections with the corresponding level features.

[0025] Further, such as Figure 2As shown in the figure, after being processed by five sparse Transformer blocks, the first channel can be projected from 96 channels to 1 channel through another convolution layer. Similarly, after being processed by five residual attention modules, the second channel can be projected from 96 channels to 1 channel through another convolution layer. The output of the first channel after the convolution layer projection and the output of the second channel after the convolution layer projection can be spliced ​​and fused, and the fused features are mapped to the single-channel output layer to effectively retain the low-frequency information of the original image, thereby ensuring that the denoised image not only has higher visual quality, but also can accurately reflect the pathological state.

[0026] From the perspective of cognitive resonance, the Residual Attention Block (RAB) and the Sparse Transformer (STB) interact synergistically to maximize the extraction of meaningful features, while balancing detail preservation and image integrity in low-dose CT image processing, thereby significantly improving the denoising effect and detail retention ability of medical images while improving computational efficiency.

[0027] Specifically, the first channel includes a first sparse Transformer block, a second sparse Transformer block, a third sparse Transformer block, a fourth sparse Transformer block, and a fifth sparse Transformer block, wherein the first sparse Transformer block, the second sparse Transformer block, and the third sparse Transformer block constitute an encoder, and the fourth sparse Transformer block and the fifth sparse Transformer block constitute a decoder; The second channel includes a first residual attention module, a second residual attention module, a third residual attention module, a fourth residual attention module and a fifth residual attention module, wherein the first residual attention module, the second residual attention module and the third residual attention module constitute an encoder, and the fourth residual attention module and the fifth residual attention module constitute a decoder.

[0028] Figure 3 It is a schematic diagram of the structure of a sparse Transformer block provided in an embodiment of the present invention.

[0029] like Figure 3 As shown, the sparse Transformer block includes a mixed-scale feed-forward network layer and a Top-K sparse attention layer; The mixed-scale feedforward network layer is used to capture and integrate multi-scale features and identify noise; The Top-K sparse attention layer is used to retain the feature information of the target image and reduce the interference of irrelevant information on the noise reduction process.

[0030] Specifically, the full name of the Sparse Transformer Block is Sparse Transformer Module (STB). With Top-k Sparse Attention as its core, STB focuses on important local features in the image through the sparse attention mechanism of the Top-K sparse attention layer, avoiding the interference of irrelevant information that may exist in traditional denoising methods. The sparse attention mechanism can dynamically select the most meaningful features for diagnosis and aggregate them, thereby effectively retaining key medical details during the denoising process. In order to further improve the denoising effect, a mixed-scale feedforward network layer (MSFN) is introduced. The multi-scale convolution kernels of MSFN are used to capture feature information of different scales, help accurately identify and remove high-frequency noise, and maintain the structural integrity of low-frequency artifacts. Finally, through the design of Top-k Sparse Attention and the double residual connection of MSFN, STB can achieve multi-scale feature fusion of local perception and global sparse attention.

[0031] Figure 4 It is a structural diagram of the residual attention module provided in an embodiment of the present invention.

[0032] like Figure 4 As shown, the residual attention module includes a residual block and a spatial attention module; The residual block includes several standard convolutional layers and activation function layers, and the residual block is used to extract local features; The spatial attention module is used to extract statistical information of the target image through global maximum pooling and global average pooling to supplement the global feature information.

[0033] The Residual Attention Block (RAB) combines residual learning with the attention mechanism. The spatial attention module can weight the feature map through global maximum pooling (GMP) and global average pooling (GAP) to extract the global context information of the image. Furthermore, the residual attention module strengthens the learning and enhancement of multi-level features through cross-level jump connections. These jump connections not only help capture deeper features, but also retain shallow information, allowing the network to more accurately adjust the feature weights of different scales, thereby achieving spatially adaptive feature fusion and noise suppression. Specifically, the statistical information extracted by the GMP and GAP modules helps the module focus on information-intensive areas in the image, and through residual learning, the information loss problem in traditional convolutional networks is avoided, and fine enhancement and adaptive optimization of features are achieved.

[0034] In an exemplary embodiment, the training process of the dual-channel neural network model includes: Inputting the training set data into a pre-built initial model to obtain a first output image, wherein the training set data includes a low-dose CT image and a standard-dose CT image, and the first output image is a denoised image of the low-dose CT image; Calculating the similarity between the first output image and the standard dose CT image by using a loss function; The initial model is updated based on the similarity.

[0035] In practical applications, the training set data can be the publicly released data set of the 2016 NIH-AAPM-Mayo Clinic Low-dose CT Grand Challenge, and the 3 mm slice data in this data set is used for training and testing. Specifically, the 3 mm slice data includes 2378 3 mm low-dose (quarter-dose) and standard-dose CT image slices from 10 anonymous patients. Slices are selected from the first 9 patients, a total of 2167 slices for training, and 211 slices are selected from the 10th patient for testing. Data expansion is performed by cutting into small pieces (patches) to reduce the computational burden.

[0036] In an exemplary embodiment, the loss function includes reconstruction loss, perceptual loss, and structural similarity loss; The reconstruction loss is used to calculate the pixel value similarity between the first output image and the standard dose CT image; The perceptual loss is used to calculate the feature similarity between the first output image and the standard dose CT image; The structural similarity loss is used to calculate the structural similarity between the first output image and the standard-dose CT image.

[0037] In an exemplary embodiment, the loss function is a weighted sum of reconstruction loss, perceptual loss and structural similarity loss.

[0038] In practical applications, reconstruction loss is the main loss function of the model, which is used to ensure the pixel-level similarity between the denoised result and the reference full-dose CT image. The mean square error (MSE) is used to measure the difference between the predicted image and the true image, which conforms to the following formula (1): (1) in, N is the total number of pixels in the image, and are the pixel values ​​of the normal dose CT image and the output image, respectively.

[0039] Perceptual loss is used to improve the visual quality of the output image. Perceptual loss captures more discernible features by calculating the difference between the output image and the normal dose CT image in the high-level feature space, which conforms to the following formula (2): (2)

[0040] in, Indicates that the pre-trained VGG network is l The feature extraction function of the layer, is the size of the feature map of this layer, l Take it as the 16th layer.

[0041] SSIM is an evaluation index for evaluating the similarity of images in terms of brightness, contrast, and structure. Its range is [-1, 1], where 1 means they are exactly the same. The SSIM loss conforms to the following formula (3): (3) in, and is an image and The average value of and is the variance, is the covariance. and is a small constant used for stability.

[0042] After respectively calculating the reconstruction loss, perceptual loss and structural similarity loss, the total loss function can be calculated by the following formula (4): (4) in, , as well as is the weight coefficient for adjusting the impact of each loss item. For example, , , .

[0043] In an exemplary embodiment, the target image is input into a pre-trained dual-channel neural network model for denoising to obtain an output result, and then the method further includes: The denoised image is evaluated using the preset evaluation index. If the evaluation is unqualified, the denoising process is repeated until a qualified denoised image is obtained.

[0044] In implementation, the evaluation indicators may include Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Root Mean Square Error (RMSE).

[0045] In practical applications, the denoising method provided by the solution of the present application can be compared with five existing methods, namely RED-CNN, WGAN-VGG, CTFormer, EDCNN and HFormer, and can also be compared with the initial low-dose CT images and standard-dose CT images.

[0046] Table 1 below shows the comparison results of the average PSNR, SSIM and RMSE values ​​of the present application and different models. As shown in Table 1, compared with the existing denoising methods, the method of the present application achieves the highest denoising effect, while also achieving a very high degree of retention of image detail structures. The method of the present invention provides higher reliability for clinical diagnosis.

[0047] Table 1 Denoising effect comparison table

[0048]

[0049] Table 2 below shows the loss ablation comparison of the solution of this application. It can be seen from Table 2 that compared with using reconstruction loss alone, the addition of SSIM loss and perceptual loss can better guide the learning of the network, achieve better denoising effect and retain more detailed structures.

[0050] Table 2 Loss ablation comparison table

[0051]

[0052] Table 3 below shows the structural ablation comparison of the solutions of this application. It can be seen from Table 3 that the effective fusion of the sparse attention block and the residual attention module further enhances the denoising ability of the network.

[0053] Table 3 Comparison of structural ablation

[0054]

[0055] The denoising system for low-dose CT images provided by the present invention is described below. The denoising system for low-dose CT images described below and the denoising method for low-dose CT images described above can be referred to each other.

[0056] Figure 5 Schematic diagram of the structure of a low-dose CT image denoising system provided by an embodiment of the present invention.

[0057] like Figure 5 As shown, the low-dose CT image denoising system provided in this embodiment includes: An image loading module 501 is used to load a target image, where the target image is a low-dose CT image to be processed; De-noising module 502, used for inputting the target image into a pre-trained dual-channel neural network model for denoising to obtain an output result; Among them, the dual-channel neural network model includes a first channel and a second channel, the first channel is a U-Net architecture based on several sparse Transformer blocks, and the second channel is a U-Net architecture based on several residual attention modules; the output result is the first output of the first channel and the second output of the second channel fused together and obtained through convolution processing.

[0058] In an exemplary embodiment, the first channel includes a first sparse Transformer block, a second sparse Transformer block, a third sparse Transformer block, a fourth sparse Transformer block, and a fifth sparse Transformer block, wherein the first sparse Transformer block, the second sparse Transformer block, and the third sparse Transformer block constitute an encoder, and the fourth sparse Transformer block and the fifth sparse Transformer block constitute a decoder; The second channel includes a first residual attention module, a second residual attention module, a third residual attention module, a fourth residual attention module and a fifth residual attention module, wherein the first residual attention module, the second residual attention module and the third residual attention module constitute an encoder, and the fourth residual attention module and the fifth residual attention module constitute a decoder.

[0059] In an exemplary embodiment, the sparse Transformer block includes a mixed-scale feed-forward network layer and a Top-K sparse attention layer; The mixed-scale feedforward network layer is used to capture and integrate multi-scale features and identify noise; The Top-K sparse attention layer is used to retain the feature information of the target image and reduce the interference of irrelevant information on the noise reduction process.

[0060] In an exemplary embodiment, the residual attention module includes a residual block and a spatial attention module; The residual block includes several standard convolutional layers and activation function layers, and the residual block is used to extract local features; The spatial attention module is used to extract statistical information of the target image through global maximum pooling and global average pooling to supplement the global feature information.

[0061] In an exemplary embodiment, the training process of the dual-channel neural network model includes: Inputting the training set data into a pre-built initial model to obtain a first output image, wherein the training set data includes a low-dose CT image and a standard-dose CT image, and the first output image is a denoised image of the low-dose CT image; Calculating the similarity between the first output image and the standard dose CT image by using a loss function; The initial model is updated based on the similarity.

[0062] In an exemplary embodiment, the loss function includes reconstruction loss, perceptual loss, and structural similarity loss; The reconstruction loss is used to calculate the pixel value similarity between the first output image and the standard dose CT image; The perceptual loss is used to calculate the feature similarity between the first output image and the standard dose CT image; The structural similarity loss is used to calculate the structural similarity between the first output image and the standard-dose CT image.

[0063] In an exemplary embodiment, the loss function is a weighted sum of reconstruction loss, perceptual loss and structural similarity loss.

[0064] In the exemplary embodiment, an evaluation module is also included. The evaluation module is specifically used to: evaluate the denoised image using a preset evaluation index. If the evaluation is unqualified, the denoising process is performed again.

[0065] Figure 6 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 6As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630 and a communication bus 640, wherein the processor 610, the communication interface 620 and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logic instructions in the memory 630 to execute the denoising method of the low-dose CT image, and the method includes: Acquiring a target image, wherein the target image is a low-dose CT image to be processed; Using a pre-trained dual-channel neural network model to perform denoising on the target image; Inputting the target image into a pre-trained dual-channel neural network model for denoising to obtain an output result; Among them, the dual-channel neural network model includes a first channel and a second channel, the first channel is a U-Net architecture based on several sparse Transformer blocks, and the second channel is a U-Net architecture based on several residual attention modules; the output result is the first output of the first channel and the second output of the second channel fused together and obtained through convolution processing.

[0066] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0067] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored in a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the low-dose CT image denoising method provided by the above methods, the method includes: Acquiring a target image, wherein the target image is a low-dose CT image to be processed; Inputting the target image into a pre-trained dual-channel neural network model for denoising to obtain an output result; Among them, the dual-channel neural network model includes a first channel and a second channel, the first channel is a U-Net architecture based on several sparse Transformer blocks, and the second channel is a U-Net architecture based on several residual attention modules; the output result is the first output of the first channel and the second output of the second channel fused together and obtained through convolution processing.

[0068] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for denoising a low-dose CT image provided by the above methods is implemented, the method comprising: Acquiring a target image, wherein the target image is a low-dose CT image to be processed; Inputting the target image into a pre-trained dual-channel neural network model for denoising to obtain an output result; Among them, the dual-channel neural network model includes a first channel and a second channel, the first channel is a U-Net architecture based on several sparse Transformer blocks, and the second channel is a U-Net architecture based on several residual attention modules; the output result is the first output of the first channel and the second output of the second channel fused together and obtained through convolution processing.

[0069] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative effort.

[0070] Through the description of the above implementation modes, those skilled in the art can clearly understand that each implementation mode can be implemented by means of software plus a necessary general hardware platform, or of course by hardware. Based on such an understanding, the above technical solution can essentially or in other words be embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for denoising low-dose CT images, characterized in that: include: Acquiring a target image, wherein the target image is a low-dose CT image to be processed; Inputting the target image into a pre-trained dual-channel neural network model for denoising to obtain an output result; Among them, the dual-channel neural network model includes a first channel and a second channel, the first channel is a U-Net architecture based on several sparse Transformer blocks, and the second channel is a U-Net architecture based on several residual attention modules; the output result is the first output of the first channel and the second output of the second channel fused together and obtained through convolution processing.

2. The method for denoising low-dose CT images according to claim 1, characterized in that: The first channel includes a first sparse Transformer block, a second sparse Transformer block, a third sparse Transformer block, a fourth sparse Transformer block, and a fifth sparse Transformer block, wherein the first sparse Transformer block, the second sparse Transformer block, and the third sparse Transformer block constitute an encoder, and the fourth sparse Transformer block and the fifth sparse Transformer block constitute a decoder; The second channel includes a first residual attention module, a second residual attention module, a third residual attention module, a fourth residual attention module and a fifth residual attention module, wherein the first residual attention module, the second residual attention module and the third residual attention module constitute an encoder, and the fourth residual attention module and the fifth residual attention module constitute a decoder.

3. The method for denoising low-dose CT images according to claim 1, characterized in that: The sparse Transformer block includes a mixed-scale feed-forward network layer and a Top-K sparse attention layer; The mixed-scale feedforward network layer is used to capture and integrate multi-scale features and identify noise; The Top-K sparse attention layer is used to retain the feature information of the target image and reduce the interference of irrelevant information on the noise reduction process.

4. The method for denoising low-dose CT images according to claim 1, characterized in that: The residual attention module includes a residual block and a spatial attention module; The residual block includes several standard convolutional layers and activation function layers, and the residual block is used to extract local features; The spatial attention module is used to extract statistical information of the target image through global maximum pooling and global average pooling to supplement the global feature information.

5. The method for denoising low-dose CT images according to claim 1, characterized in that: The training process of the dual-channel neural network model includes: Inputting the training set data into a pre-built initial model to obtain a first output image, wherein the training set data includes a low-dose CT image and a standard-dose CT image, and the first output image is a denoised image of the low-dose CT image; Calculating the similarity between the first output image and the standard dose CT image by using a loss function; The initial model is updated based on the similarity.

6. The method for denoising low-dose CT images according to claim 5, characterized in that: The loss function includes reconstruction loss, perceptual loss and structural similarity loss; The reconstruction loss is used to calculate the pixel value similarity between the first output image and the standard dose CT image; The perceptual loss is used to calculate the feature similarity between the first output image and the standard dose CT image; The structural similarity loss is used to calculate the structural similarity between the first output image and the standard-dose CT image.

7. The method for denoising low-dose CT images according to claim 6, characterized in that: The loss function is a weighted sum of reconstruction loss, perceptual loss and structural similarity loss.

8. The method for denoising low-dose CT images according to claim 1, characterized in that: The target image is input into a pre-trained dual-channel neural network model for denoising to obtain an output result, and then the method further includes: The denoised image is evaluated using the preset evaluation index. If the evaluation is unqualified, the denoising process is repeated until a qualified denoised image is obtained.

9. A denoising system for low-dose CT images, characterized in that: include: An image loading module, used for loading a target image, wherein the target image is a low-dose CT image to be processed; A denoising module, used for inputting the target image into a pre-trained dual-channel neural network model for denoising to obtain an output result; Among them, the dual-channel neural network model includes a first channel and a second channel, the first channel is a U-Net architecture based on several sparse Transformer blocks, and the second channel is a U-Net architecture based on several residual attention modules; the output result is the first output of the first channel and the second output of the second channel fused together and obtained through convolution processing.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the low-dose CT image denoising method according to any one of claims 1 to 8 is implemented.

Images