Retinal vessel image segmentation method based on parent loss integration

Abstract

The invention relates to a retinal blood vessel image segmentation method based on parent loss integration, comprising: inputting a retinal blood vessel image to be segmented into a PL-TransUNet model to obtain an image segmentation visualization result, the PL-TransUNet model being obtained based on training of a training set, the training set comprising the retinal blood vessel image and a corresponding true value; the process of training the PL-TransUNet model based on the training set comprises the steps of predicting a retinal blood vessel image through an initial PL-TransUNet model, calculating a loss channel according to a prediction result and a truth value of the retinal blood vessel image, performing reverse filling and feature enhancement according to the loss channel, obtaining a final enhanced feature map, and obtaining a final enhanced feature map. And obtaining a new prediction result by using the final enhanced feature map, carrying out iteration until convergence, and obtaining a PL-TransUNet model. According to the invention, the accuracy of retinal vessel image segmentation can be improved.

Description

Technical Field

[0001] The invention relates to the technical field of image segmentation, and in particular to a retinal blood vessel image segmentation method based on parent loss integration. Background Art

[0002] At present, machine vision-based methods are commonly used to automatically segment retinal blood vessels. Among them, the deep learning image segmentation method of U-Net is the representative, which has the advantages of simple structure, high efficiency, and multi-scale feature fusion, and is widely used. Although the encoding and decoding structure of U-Net has shown excellent performance in some medical image segmentation tasks, it also has some defects. For example, the encoder mainly uses convolutional neural networks, and its local characteristics of convolution lead to its limited ability in global feature representation. Methods such as Transformer can be integrated into the encoder for improvement. For example, Zhang et al. proposed an enhancement network using SwinTransformer as an encoder to enhance features by fusing features at different perceptual levels and improving the utilization of spatial and channel information of the image. Fan et al. used two parallel branches of CNN and Transformer in the encoder to extract local and global information, and used channel attention and spatial attention for feature fusion. However, the ability of these methods to integrate prior knowledge to guide segmentation needs to be further improved. In addition, for example, Zhang et al. integrated the spatial prior information of the pancreas into deep CNN to improve the accuracy of organ segmentation. Wang et al. proposed a progressive dual prior network, which first locates and crops the tumor area, and then gradually refines the tumor using weak semantic prior and cross-scale related prior knowledge. Zhang et al. proposed a Gaussian mixture prior method, which embeds prior information into the self-attention mechanism to guide segmentation. The above methods all introduce priors from the perspective of estimating the target position, but ignore the loss information.

[0003] In order to improve the prior knowledge integration and global feature enhancement capabilities of the codec architecture neural network, the present invention proposes a retinal vascular image segmentation method based on parent loss integration. Summary of the invention

[0004] The purpose of the present invention is to provide a retinal vascular image segmentation method based on parent loss integration, strengthen the automatic integration of prior knowledge and global feature enhancement capabilities of the encoding-decoding neural network to improve the segmentation accuracy of retinal vascular images.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] The retinal vascular image segmentation method based on parent loss integration includes:

[0007] Acquire a retinal blood vessel image to be segmented;

[0008] Inputting the retinal vascular image to be segmented into a PL-TransUNet model to obtain an image segmentation visualization result, wherein the PL-TransUNet model is obtained based on training of a training set, and the training set includes retinal vascular images and corresponding true values;

[0009] The process of training the PL-TransUNet model based on the training set includes: predicting the retinal vascular image through the initial PL-TransUNet model, calculating the loss channel according to the prediction result of the retinal vascular image and the true value, performing reverse filling and feature enhancement according to the loss channel, obtaining a final enhanced feature map, using the final enhanced feature map to obtain a new prediction result for iteration until convergence, and obtaining the PL-TransUNet model.

[0010] Optionally, the PL-TransUNet model includes an encoder, a decoder and a BPL module, wherein the encoder includes a CNN local encoder and an RLFE module, and the RLFE module includes several Transformer submodules;

[0011] The BPL module is used to calculate the loss channel according to the prediction result of the retinal vascular image and the true value;

[0012] The encoder is used to perform feature enhancement on the retinal vascular image based on the loss channel to obtain the final enhanced feature map;

[0013] The decoder is used to output the new prediction result based on the final enhanced feature map.

[0014] Optionally, calculating the loss channel according to the prediction result of the retinal vascular image and the true value includes: inputting the prediction result and the true value into the BPL module to calculate the loss matrix;

[0015] The maximum value of the loss matrix is ​​selected according to the channel and binarized to obtain the loss channel.

[0016] Optionally, the loss matrix can be calculated as:

[0017] L m = -α(1-p l ) γ log(p l )

[0018] Among them, L m is the loss matrix, α is the balance factor, p l is the prediction result, and γ is the focusing parameter.

[0019] Optionally, the method to obtain the loss channel is:

[0020]

[0021] C index = argmax c (L m )

[0022] Among them, Lc is the loss channel, max c (·) indicates taking the maximum value by channel, Cindex indicates the channel number of the maximum value, L m is the loss matrix, and (x, y) is the channel element coordinate.

[0023] Optionally, performing reverse filling and feature enhancement according to the loss channel to obtain a final enhanced feature map includes:

[0024] Reversely fill the loss channel to the idle channel to obtain the filled idle channel;

[0025] Taking the retinal blood vessel image and the padded idle channel as image input, inputting the image input and the loss channel into the CNN local encoder respectively, and calculating image embedding and loss embedding;

[0026] The image embedding and the loss embedding are input into the RLFE module for global feature enhancement to obtain the final enhanced feature map.

[0027] Optionally, inputting the image embedding and the loss embedding into the RLFE module for global feature enhancement includes:

[0028] Embedding the image into the N-1 layer Transformer submodule in the RLFE module for global encoding, embedding the globally encoded image and the loss into the feature enhancement submodule to obtain an enhanced feature map;

[0029] The enhanced feature map is input into the N-layer Transformer submodule for global encoding to obtain the final enhanced feature map.

[0030] Optionally, the method for obtaining the enhanced feature map is:

[0031] F enhanc =F embed +F L

[0032] F L =L embed ·(mask F ·Mask L )

[0033]

[0034] Among them, F enhanc is the enhanced feature map, F embed is the image embedding, F L is the enhancement component based on the loss channel, mask F Embed mask for image, mask L is the parent loss embedding mask, (x, y) is the channel element coordinate, L embed is the loss embedding.

[0035] The beneficial effects of the present invention are as follows: the method of the present invention reversely fills the parent loss (BPL) module, converts the loss information of the segmentation model into a loss channel and reversely fills it into the idle channel of the next generation training input, and participates in the next generation training as prior knowledge to guide segmentation. Then, the parent loss is continued to be used for reverse loss feature enhancement (RLFE), first using Transformer to globally encode the image embedding, and then based on the attention mechanism, the loss embedding and the processed image embedding are feature enhanced and fused, which improves the segmentation accuracy from the two aspects of prior integration and feature enhancement, and has better segmentation effect in retinal vascular image segmentation, especially low-contrast microvascular segmentation. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only 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.

[0037] Figure 1 Comparison of the training process of PL-TransUNet of the embodiment of the present invention and the traditional method;

[0038] Figure 2 This is a flow chart of a retinal vascular image segmentation method based on parent loss integration according to an embodiment of the present invention;

[0039] Figure 3 A flow chart of calculating loss channels of a BPL module according to an embodiment of the present invention;

[0040] Figure 4 is a structural diagram of a RLFE module according to an embodiment of the present invention;

[0041] Figure 5 1 is an architecture diagram of PL-TransUNet according to an embodiment of the present invention;

[0042] Figure 6This is a visualization comparison chart of segmentation results of different methods on different data sets in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0044] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0045] Inspired by the reflection effect in biology, this embodiment proposes a retinal vascular image segmentation method based on Backfilling parent loss (BPL) and Reverse loss feature enhancements (RLFE) on the basis of TransUNet, named PL-TransUNet. Figure 1 The figure shows the comparison between PL-TransUNet and traditional methods. Top: At the end of each round of training, the traditional model directly performs back propagation after calculating the loss. Bottom: After PL-TransUNet calculates the loss, it further converts it into a loss matrix, and performs prior integration and feature enhancement in the next round of model training.

[0046] This embodiment provides a retinal vascular image segmentation method based on parent loss integration, including:

[0047] Acquire a retinal blood vessel image to be segmented;

[0048] Input the retinal vascular image to be segmented into the PL-TransUNet model to obtain the image segmentation visualization result, wherein the PL-TransUNet model is trained based on the training set, and the training set includes the retinal vascular image and the corresponding true value;

[0049] The process of training the PL-TransUNet model based on the training set includes: predicting the retinal vascular image through the initial PL-TransUNet model, calculating the loss channel according to the prediction result and the true value of the retinal vascular image, performing reverse filling and feature enhancement according to the loss channel, obtaining the final enhanced feature map, and using the final enhanced feature map to obtain a new prediction result for iteration until convergence to obtain the PL-TransUNet model.

[0050] Furthermore, the PL-TransUNet model includes an encoder, a decoder and a BPL module, wherein the encoder includes a CNN local encoder and an RLFE module, and the RLFE module includes several Transformer submodules. The BPL module is used to calculate the loss channel according to the prediction result and the true value of the retinal vascular image; the encoder is used to enhance the features of the retinal vascular image based on the loss channel to obtain the final enhanced feature map; the decoder is used to output a new prediction result based on the final enhanced feature map.

[0051] like Figure 5 As shown, in this embodiment, based on the proposed BPL and RLFE modules, a PL-TransUNet model is constructed on the basis of TransUNet. The PL-TransUNet model consists of two parts: an encoder and a decoder. The encoder contains 2 inputs, namely image input (4-channel input) and a loss channel. The encoder is mainly composed of a CNN local encoder and an RLFE module, where CNN adopts ResNet-50, and the RLFE module contains 12 Transformer sub-modules. The decoder has the same decoder structure as TransUNet, and is mainly composed of 1x1 convolution, upsampling layer and convolution layer. After the decoder outputs the prediction result, the BPL module is used to perform reverse parent loss filling.

[0052] The overall process of the method in this embodiment is as follows: Figure 2 As shown. First, the prediction result is obtained through the forward propagation of PL-TransUNet. On the one hand, the loss is calculated and back-propagated based on the prediction result and the true value GT. On the other hand, it is passed to the BPL module to convert it into a loss channel and reversely filled into the idle channel. Next, the image and the filled idle channel (collectively referred to as 4-channel image input) are input into the CNN local encoder to calculate the image embedding, and the loss channel is also input into the CNN to obtain the loss embedding. After that, these two embeddings are input into the RLFE module for global feature enhancement. Finally, the enhanced feature map is input into the decoder to obtain the prediction result, and this process is repeated until the algorithm converges.

[0053] Specifically, in this embodiment, when PL-TransUNet is trained for the first time, the two inputs are 4-channel input (i.e., the original image and an idle channel C0 of all zeros) and a loss channel of all zeros. The two inputs are first forward propagated independently, and after ResNet-50 encoding, image embedding and loss embedding are obtained respectively; then the two embeddings are input to the RLFE module for feature enhancement and fusion, and then enter the decoder to output the prediction results. In addition to participating in back propagation, the output also enters the BPL module to calculate the loss channel. In subsequent training, the loss channel of the last training is used to fill C0 and serve as another input. The subsequent process is the same as the first training until the convergence condition is reached and the algorithm stops.

[0054] Further, calculating the loss channel according to the prediction result and the true value of the retinal vascular image includes: inputting the prediction result and the true value into the BPL module to calculate the loss matrix;

[0055] According to the channel, the maximum value of the loss matrix is ​​selected and binarized to obtain the loss channel.

[0056] Specifically, in this embodiment, in order to carry GT information in the forward propagation process, a reverse padding parent loss method is proposed. Inspired by the biological reflection mechanism, the traditional training process can be regarded as a simple reflection, and the model only makes simple predictions and adjustments based on the input data; while the BPL method is similar to a complex reflection, and the model can use additional GT information to adjust its own parameters.

[0057] like Figure 3 As shown, the BPL module takes GT and prediction results as input and first calculates the loss matrix L m :

[0058] L m = -α(1-p l ) γ log(p l )

[0059] Among them, L m is the loss matrix, the balance factor α∈[0,1], the balance factor is α for positive samples and 1-α for negative samples, p l To predict the results, γ is the focusing parameter (γ ≥ 0), which controls the number of easy or hard samples.

[0060] Next, for L m Take the maximum value by channel and perform binarization to obtain the loss channel L carrying GT information c :

[0061]

[0062] Cindex = argmax c (L m )

[0063] Among them, Lc is the loss channel, max c (·) indicates taking the maximum value by channel, and Cindex indicates the channel number taking the maximum value. index = 1, the maximum loss value at this position is the value of the first channel (foreground), thereby retaining the loss information (L c Set to 1); otherwise, L c Set to 0, no loss information is retained, (x, y) is the coordinate of the channel element.

[0064] Furthermore, reverse filling and feature enhancement are performed according to the loss channel to obtain the final enhanced feature map including:

[0065] Reversely fill the loss channel to the idle channel to obtain the filled idle channel;

[0066] The retinal vascular image and the padded idle channel are used as image input, the image input and loss channel are input into the CNN local encoder respectively, and the image embedding and loss embedding are calculated;

[0067] The image embedding and loss embedding are input into the RLFE module for global feature enhancement to obtain the final enhanced feature map.

[0068] Specifically, in this embodiment, the image and the padded idle channels (collectively referred to as 4-channel image input) are input into the CNN local encoder to calculate the image embedding, and the loss channel is also input into the CNN to obtain the loss embedding. Afterwards, these two embeddings are input into the RLFE module for global feature enhancement.

[0069] Furthermore, the image embedding and loss embedding are input into the RLFE module for global feature enhancement including:

[0070] The image is embedded into the N-1 layer Transformer submodule in the input RLFE module for global encoding, and the globally encoded image embedding and loss are embedded into the input feature enhancement submodule to obtain the enhanced feature map;

[0071] The enhanced feature map is input into the N-layer Transformer submodule for global encoding to obtain the final enhanced feature map.

[0072] Specifically, in this embodiment, the loss channel contains rich GT information. In order to make full use of this information to guide feature extraction, a reverse loss feature enhancement (RLFE) method is proposed. Figure 4 As shown, the RLFE module contains two inputs: image embedding F embed and loss embedding Lembed First, F embed The input is sent to the N-1 layer Transformer submodule for global encoding and then combined with L embed Input to the feature enhancement submodule to calculate the enhanced feature map F enhanc ::

[0073] F enhanc =F embed +F L

[0074] F L =L embed ·(mask F ·Mask L )

[0075]

[0076] Among them, F enhanc is the enhanced feature map, F embed is the image embedding, F L For the enhancement component based on the loss channel, special attention is paid to the area where the loss is not 0 but the feature is 0, so that the features of this area are enhanced. F Embed mask for image, mask L Embedding mask for the parent loss.

[0077] like Figure 1 As shown in the figure, compared with the traditional segmentation model, PL-TransUNet further converts the loss into a loss channel after calculating the loss. On the one hand, it is used as a priori reverse filling to the idle channel of the next generation training input. On the other hand, the loss channel is converted into a loss embedding to further enhance the image features of the encoder. The main effects are as follows:

[0078] (1) A reverse padding parent loss module is designed. This module calculates an additional loss channel after forward propagation, and the loss channel is reversely filled into the idle channel of the next generation training input. This method enables the next generation training to obtain additional ground truth (GT) information and use the parent loss prior to guide segmentation.

[0079] (2) A reverse loss feature enhancement method based on parent loss is proposed. This module is integrated after the CNN local encoder of the model. First, the multi-layer Transformer module extracts global features from the image embedding. Then, the extracted image features are enhanced based on the attention mechanism using the loss embedding. Finally, another layer of Transformer module is used to perform a global feature fusion. This method effectively utilizes the GT information of the loss channel and further globally enhances the image features.

[0080] This embodiment is compared with other SOTA methods such as U-Net and TransUnet on three retinal blood vessel segmentation datasets: DRIVE, CHASE_DB1 and LES-AV. The results show that PL-TransUNet has a higher accuracy of approximately 98.94%, 98.48% and 98.30% respectively.

[0081] The experiment was implemented with PyTorch1.13 and ran on a computer with a 3.5GHz Intel i5-13600KF processor, 32GB of 3600MHz DDR4 RAM, and a 12GB NVIDIA RTX4070Ti. The system environment was Windows11. The experiment was conducted on three public retinal vascular datasets: DRIVE, CHASE_DB1, and LES-AV. First, the segmentation accuracy was compared with SOTA methods such as AM-Unet, DA-TransUNe, and RRWNet. Secondly, the effectiveness of BPL and RLFE was verified by ablation experiments. Finally, the segmentation results were visualized and discussed. The experimental parameters were set as follows: the optimizer used SGD, with a weight decay of 0.0001, a momentum of 0.9, an initial learning rate of 0.01, and an exponential decay method to update the learning rate. Max_Epoch = 200, batch size = 1, α = 0.5, γ = 2.

[0082] The relevant parameters of the dataset are shown in Table 1. The DRIVE dataset contains 40 retinal vascular images and their corresponding GTs with a resolution of 565×584. It is pre-divided into 20 samples for training and 20 samples for testing. The CHASE_DB1 dataset contains 28 retinal vascular images and their corresponding GTs with a resolution of 990×960. The training subset contains 20 samples, and the rest are used as test subsets. The LES-AV dataset contains 22 retinal vascular images and their corresponding GTs with an image resolution of 1620×1444. Its training set contains 16 images, and the remaining images are used as test sets. In order to alleviate the overfitting problem, 512×512 areas of the original image are randomly cut out as training samples during training.

[0083] Table 1

[0084]

[0085] The segmentation results are compared with the manually segmented GT. The sensitivity (SE), specificity (SP), accuracy (AC), Jaccard coefficient (JA), and Dice coefficient (DI, also known as F-measure) are calculated to measure the degree to which each pixel is correctly classified as background or blood vessel. The formulas for these indicators are as follows:

[0086]

[0087]

[0088] Among them, TP represents the number of true positive samples, FP represents the number of false positive samples, TN represents the number of true negative samples, and FN represents the number of false negative samples.

[0089] First, it is compared with ten methods including IterMiU-Net, HRD-Net, IMFF-Net, MILU-Net, AM-Unet, and DA-TransUNet on the DRIVE dataset. Table 2 shows the comparison results with existing methods on the DRIVE dataset. The bold values ​​indicate the best indicators. The AC and DI values ​​of PL-TransUNet rank first. AM-Unet adopts an attention mechanism and enhances the image contrast by enhancing the green channel. Its AC value ranks second. The AC value of DA-TransUNet ranks third. It is also based on the improved TransUNet structure and integrates spatial and channel attention. HRD-Net and MILU-Net enhance features by fusing branch outputs of different resolutions or fusing multi-scale detail features. Their AC and DI values ​​are very close and lower than DA-TransUNet. IterMiU-Net focuses on vascular information from multiple dimensions and enhances the ability to obtain local and global information. IMFF-Net adopts a four-layer encoder-decoder framework based on attention pool feature fusion. The AC and DI values ​​of these two methods are low.

[0090] Then, it is compared with nine methods including IterMiU-Net, HRD-Net, IMFF-Net, AM-Unet, DA-TransUNet, etc. on the CHASE_DB1 dataset. Table 3 shows the comparison results with existing methods on the CHASE_DB1 dataset. The bold values ​​indicate the best indicators. PL-TransUNet has the highest AC (98.48%), DI (85.30%), and JA (74.41%). HRD-Net ranks second in DI value. It fuses the outputs of branches with different resolutions through a global aggregation operation and integrates features using feature enhancement cascades. DA-TransUNet ranks third in DI value. It uses a dual attention mechanism in skip connections to capture finer features, but it easily leads to model overfitting. IterMiU-Net and IMFF-Net both have low DI values.

[0091] Furthermore, the eight methods including DA-TransUNet, FSE-Net, and RRWNet were compared on the LES-AV dataset. Table 4 shows the comparison results with existing methods on the LES-AV dataset. The bold values ​​indicate the best indicators. PL-TransUNet also has the highest AC (98.29%), DI (85.86%), and JA (75.24%). TransUNet ranks second in AC value, combining the self-attention mechanism of Transformer and the encoder-decoder structure of U-Net. DA-TransUNet uses a jump connection with a dual attention mechanism, and its AC value ranks third. FSE-Net ranks fourth in AC value, which processes the features of the distillation layer and the refinement layer through residual convolution and fuses multi-scale features. RRWNet is based on a fully convolutional neural network and has a lower AC value.

[0092] Table 2

[0093]

[0094] Table 3

[0095]

[0096] Table 4

[0097]

[0098]

[0099] Figure 6 The image segmentation visualization results of PL-TransUNet, TransUnet and DA-TransUNet on three datasets are shown. Figure 6 It can be seen that the compared methods can segment most of the blood vessel images well, but for the blood vessels in the low-contrast area, especially the micro-vessels, the segmentation results are different. Figure 6From the local magnified image, we can see that TransUnet only relies on the local and global features of the image itself for image segmentation. Its segmentation result is the worst, and most of the microvessels are classified as background. The segmentation result of DA-TransUnet is slightly better than that of TransUnet. It integrates the spatial and channel attention mechanisms on the basis of TransUnet for feature enhancement. Relatively obvious microvessels are segmented, but particularly small blood vessels cannot be segmented, resulting in more prominent discontinuity of blood vessels. PL-TransUNet has the best segmentation result, and most of the microvessels are correctly segmented. This is due to its parent loss integration mechanism, which improves the segmentation accuracy from the two aspects of prior integration and feature enhancement. It can be seen that PL-TransUNet has better segmentation effect in retinal vascular image segmentation, especially low-contrast microvessel segmentation.

[0100] The method of this embodiment adopts the backfill parent loss (BPL) module to convert the loss information of the segmentation model into a loss channel and backfill it into the idle channel of the next generation training input, and participate in the next generation training as prior knowledge to guide segmentation. At the same time, RLFE uses loss embedding to perform feature enhancement based on the attention mechanism on image embedding to extract more effective image features. Its effectiveness has been demonstrated through experimental comparisons on three retinal vascular datasets. The proposed BPL and RLFE modules are relatively independent and can be applied to different semantic segmentation models.

[0101] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A retinal vascular image segmentation method based on parent loss integration, characterized in that: include: Acquire a retinal blood vessel image to be segmented; Inputting the retinal vascular image to be segmented into a PL-TransUNet model to obtain an image segmentation visualization result, wherein the PL-TransUNet model is obtained based on training of a training set, and the training set includes retinal vascular images and corresponding true values; The process of training the PL-TransUNet model based on the training set includes: predicting the retinal vascular image through the initial PL-TransUNet model, calculating the loss channel according to the prediction result of the retinal vascular image and the true value, performing reverse filling and feature enhancement according to the loss channel, obtaining a final enhanced feature map, using the final enhanced feature map to obtain a new prediction result for iteration until convergence, and obtaining the PL-TransUNet model.

2. The retinal vascular image segmentation method based on parent loss integration according to claim 1, characterized in that: The PL-TransUNet model includes an encoder, a decoder and a BPL module, wherein the encoder includes a CNN local encoder and an RLFE module, and the RLFE module includes several Transformer submodules; The BPL module is used to calculate the loss channel according to the prediction result of the retinal vascular image and the true value; The encoder is used to perform feature enhancement on the retinal vascular image based on the loss channel to obtain the final enhanced feature map; The decoder is used to output the new prediction result based on the final enhanced feature map.

3. The retinal vascular image segmentation method based on parent loss integration according to claim 2, characterized in that: Calculating the loss channel according to the prediction result and the true value of the retinal vascular image includes: inputting the prediction result and the true value into the BPL module to calculate the loss matrix; The maximum value of the loss matrix is ​​selected according to the channel and binarized to obtain the loss channel.

4. The retinal vascular image segmentation method based on parent loss integration according to claim 3, characterized in that: The method to calculate the loss matrix is: L m =-α(1-p l ) γ log(p l ) Among them, L m is the loss matrix, α is the balance factor, p l is the prediction result, and γ is the focusing parameter.

5. The retinal vascular image segmentation method based on parent loss integration according to claim 3, characterized in that: The method to obtain the loss channel is: C index =argmax c (L m ) Among them, Lc is the loss channel, max c (·) indicates taking the maximum value by channel, Cindex indicates the channel number of the maximum value, L m is the loss matrix, and (x, y) is the channel element coordinate.

6. The retinal vascular image segmentation method based on parent loss integration according to claim 3, characterized in that: Reverse filling and feature enhancement are performed according to the loss channel to obtain the final enhanced feature map including: Reversely fill the loss channel to the idle channel to obtain the filled idle channel; Taking the retinal blood vessel image and the padded idle channel as image input, inputting the image input and the loss channel into the CNN local encoder respectively, and calculating image embedding and loss embedding; The image embedding and the loss embedding are input into the RLFE module for global feature enhancement to obtain the final enhanced feature map.

7. The retinal vascular image segmentation method based on parent loss integration according to claim 6, characterized in that: Inputting the image embedding and the loss embedding into the RLFE module for global feature enhancement includes: Embedding the image into the N-1 layer Transformer submodule in the RLFE module for global encoding, embedding the globally encoded image and the loss into the feature enhancement submodule to obtain an enhanced feature map; The enhanced feature map is input into the N-layer Transformer submodule for global encoding to obtain the final enhanced feature map.

8. The retinal vascular image segmentation method based on parent loss integration according to claim 7, characterized in that: The method to obtain the enhanced feature map is: F enhanc =F embed +F L F L =L embed ·(mask F ·mask L ) Among them, F enhanc is the enhanced feature map, F embed is the image embedding, F L is the enhancement component based on the loss channel, mask F Embed mask for image, mask L is the parent loss embedding mask, (x, y) is the channel element coordinate, L embed is the loss embedding.

Images