Training method and device of gastric cancer focus segmentation model, electronic equipment, storage medium and gastric cancer focus segmentation method

Abstract

The invention discloses a gastric cancer focus segmentation model training method and device, electronic equipment, a storage medium and a gastric cancer focus segmentation method. The invention relates to the technical field of image processing. The stomach cancer focus segmentation model training method comprises the steps that a three-dimensional computed tomography sample image is input into a semantic segmentation model to be trained, a stomach focus segmentation result is obtained, the semantic segmentation model comprises at least one residual error coding module, at least one bottleneck module and at least one decoding module, and the residual error coding module is used for coding the three-dimensional computed tomography sample image; the residual encoding module is in jump connection with the decoding module; and determining target model loss based on the stomach lesion segmentation result and the stomach lesion segmentation label, updating model parameters of the semantic segmentation model based on the target model loss until a model training stop condition is met, and obtaining a trained stomach cancer lesion segmentation model. According to the technical scheme, the model capable of realizing automatic segmentation of the gastric cancer lesion is obtained through training, so that the gastric cancer lesion area is segmented efficiently and accurately.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a training method, device, electronic equipment, storage medium and gastric cancer lesion segmentation method for a gastric cancer lesion segmentation model. Background Art

[0002] Early diagnosis and precise treatment of gastric cancer have become the research focus in the global medical field. Surgery is the main means of treating gastric cancer, and preoperative examination is crucial to the success of the operation.

[0003] Computed tomography (CT) can provide comprehensive preoperative imaging data, clearly showing the size and distribution of gastric cancer lesions, and providing an important basis for doctors to formulate surgical plans.

[0004] Although CT images play an important role in the diagnosis of gastric cancer, most of the current film reading work still relies on manual analysis by doctors. This method is affected by the doctors' experience and the accuracy of the results is unstable. In addition, the manual recognition process is time-consuming and labor-intensive, and it is difficult to meet the needs of rapid diagnosis. Summary of the invention

[0005] The present invention provides a training method, device, electronic device, storage medium and gastric cancer lesion segmentation method for a gastric cancer lesion segmentation model, so as to realize automatic segmentation of gastric cancer lesions and efficiently and accurately segment out the gastric cancer lesion area.

[0006] According to one aspect of the present invention, a method for training a gastric cancer lesion segmentation model is provided, comprising:

[0007] Acquire a three-dimensional computed tomography sample image and a gastric lesion segmentation label corresponding to the three-dimensional computed tomography sample image;

[0008] Inputting the three-dimensional computed tomography sample image into a semantic segmentation model to be trained to obtain a gastric lesion segmentation result, wherein the semantic segmentation model includes at least one residual encoding module, at least one bottleneck module and at least one decoding module, and the residual encoding module is jump-connected with the decoding module;

[0009] The target model loss is determined based on the gastric lesion segmentation result and the gastric lesion segmentation label, and the model parameters of the semantic segmentation model are updated based on the target model loss until the model training stop condition is met, thereby obtaining a trained gastric cancer lesion segmentation model.

[0010] According to another aspect of the present invention, a training device for a gastric cancer lesion segmentation model is provided, comprising:

[0011] A training sample acquisition module, used to acquire a three-dimensional computer tomography sample image and a gastric lesion segmentation label corresponding to the three-dimensional computer tomography sample image;

[0012] a gastric lesion segmentation result prediction module, used for inputting the three-dimensional computed tomography sample image into a semantic segmentation model to be trained to obtain a gastric lesion segmentation result, wherein the semantic segmentation model comprises at least one residual encoding module, at least one bottleneck module and at least one decoding module, and the residual encoding module is jump-connected with the decoding module;

[0013] A model parameter updating module is used to determine the target model loss based on the gastric lesion segmentation result and the gastric lesion segmentation label, and to update the model parameters of the semantic segmentation model based on the target model loss until the model training stop condition is met, thereby obtaining a trained gastric cancer lesion segmentation model.

[0014] According to another aspect of the present invention, a method for segmenting gastric cancer lesions is provided, comprising:

[0015] acquiring a computed tomography image to be segmented;

[0016] Inputting the computed tomography image to be segmented into a trained gastric cancer lesion segmentation model to obtain an initial gastric cancer lesion segmentation result, wherein the gastric cancer lesion segmentation model is obtained after training according to the training method for the gastric cancer lesion segmentation model described in any embodiment of the present invention;

[0017] The initial gastric cancer lesion segmentation result is subjected to morphological processing and / or maximum connected domain extraction to obtain a target gastric cancer lesion segmentation result.

[0018] According to another aspect of the present invention, a gastric cancer lesion segmentation device is provided, comprising:

[0019] A computer tomography image acquisition module, used for acquiring a computer tomography image to be segmented;

[0020] an initial gastric cancer lesion segmentation result prediction module, used for inputting the computed tomography image to be segmented into a trained gastric cancer lesion segmentation model to obtain an initial gastric cancer lesion segmentation result, wherein the gastric cancer lesion segmentation model is obtained after training according to the gastric cancer lesion segmentation model training method described in any embodiment of the present invention;

[0021] The target gastric cancer lesion segmentation result determination module is used to perform morphological processing and / or maximum connected domain extraction on the initial gastric cancer lesion segmentation result to obtain the target gastric cancer lesion segmentation result.

[0022] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0023] at least one processor;

[0024] and a memory communicatively coupled to the at least one processor;

[0025] In which, the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the training method of the gastric cancer lesion segmentation model or the gastric cancer lesion segmentation method described in any embodiment of the present invention.

[0026] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the training method for a gastric cancer lesion segmentation model or the gastric cancer lesion segmentation method described in any embodiment of the present invention when executed.

[0027] The technical solution of the embodiment of the present invention obtains a three-dimensional computed tomography sample image and a gastric lesion segmentation label corresponding to the three-dimensional computed tomography sample image, and then inputs the three-dimensional computed tomography sample image into the semantic segmentation model to be trained to obtain a gastric lesion segmentation result, wherein the semantic segmentation model includes at least one residual coding module, at least one bottleneck module and at least one decoding module, and the residual coding module and the decoding module are jump-connected, and then the target model loss is determined based on the gastric lesion segmentation result and the gastric lesion segmentation label, and the model parameters of the semantic segmentation model are updated based on the target model loss until the model training stop condition is met, thereby obtaining a trained gastric cancer lesion segmentation model. The above technical solution optimizes the feature extraction capability of the model through the residual coding module, effectively improves the accuracy of the gastric cancer lesion segmentation model, realizes the automatic segmentation of gastric cancer lesions through the gastric cancer lesion segmentation model, and efficiently and accurately segments the gastric cancer lesion area.

[0028] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0030] Figure 1 is a flow chart of a method for training a gastric cancer lesion segmentation model according to Embodiment 1 of the present invention;

[0031] Figure 2 is a flow chart of a gastric cancer lesion segmentation model training method provided according to Embodiment 2 of the present invention;

[0032] Figure 3 is a flow chart of a method for training a gastric cancer lesion segmentation model according to Embodiment 3 of the present invention;

[0033] Figure 4 is a schematic structural diagram of a gastric cancer lesion segmentation model provided according to an embodiment of the present invention;

[0034] Figure 5 is a structural schematic diagram of a first basic residual module provided according to an embodiment of the present invention;

[0035] Figure 6 is a structural schematic diagram of a second basic residual module provided according to an embodiment of the present invention;

[0036] Figure 7 is a flow chart of a gastric cancer lesion segmentation method provided according to Embodiment 4 of the present invention;

[0037] Figure 8 is a flow chart of a gastric cancer lesion segmentation method provided according to an embodiment of the present invention;

[0038] Fig. 9 is a schematic diagram of the structure of a training device for a gastric cancer lesion segmentation model provided according to Embodiment 5 of the present invention;

[0039] Fig.10 is a schematic structural diagram of a gastric cancer lesion segmentation device provided according to Embodiment 6 of the present invention;

[0040] Fig.11 It is a structural schematic diagram of an electronic device for implementing the training method of the gastric cancer lesion segmentation model according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 should fall within the scope of protection of the present invention.

[0042] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. The acquisition, storage, use, processing, etc. of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0043] Embodiment 1

[0044] Figure 1 This is a flow chart of a method for training a gastric cancer lesion segmentation model provided in Example 1 of the present invention. This embodiment is applicable to the case of identifying gastric cancer lesions in abdominal CT images. The method can be executed by a training device for a gastric cancer lesion segmentation model. The training device for the gastric cancer lesion segmentation model can be implemented in the form of hardware and / or software. The training device for the gastric cancer lesion segmentation model can be configured in electronic devices such as terminals and / or servers. Figure 1 As shown, the method includes:

[0045] S110, acquiring a three-dimensional computed tomography sample image and a gastric lesion segmentation label corresponding to the three-dimensional computed tomography sample image.

[0046] In the embodiment of the present invention, the 3D computed tomography sample image refers to a 3D CT image used for model training, for example, the 3D computed tomography sample image may be an abdominal region or other 3D CT image that can cover the stomach region. The gastric lesion segmentation label refers to a segmentation label of the stomach region used for model training, for example, the gastric lesion segmentation label may be a segmentation label of the complete stomach or the gastric lesion region, etc.

[0047] Exemplarily, multiple three-dimensional computed tomography sample images and the gastric lesion segmentation label corresponding to each three-dimensional computed tomography sample image can be read from a preset storage path of the electronic device, and multiple three-dimensional computed tomography sample images and the gastric lesion segmentation label corresponding to each three-dimensional computed tomography sample image can also be obtained from the cloud or other devices connected to the electronic device for communication, without specific limitation here.

[0048] S120. Input the three-dimensional computed tomography sample image into the semantic segmentation model to be trained to obtain a gastric lesion segmentation result, wherein the semantic segmentation model includes at least one residual encoding module, at least one bottleneck module and at least one decoding module, and the residual encoding module is jump-connected to the decoding module.

[0049] The semantic segmentation model is a deep learning model based on a residual encoder, including at least one residual encoding module, at least one bottleneck module and at least one decoding module, and the residual encoding module is jump-connected with the decoding module. The residual encoding module is a residual encoder, which can be composed of one or more residual basic modules. The bottleneck module is a bottleneck layer (Bottleneck), which is used to compress information and extract the most representative features. The decoding module is a decoder, which is used to reconstruct the compressed data to restore the integrity and accuracy of the original data.

[0050] It should be noted that by introducing the residual coding module and adding skip connections, the network depth and feature extraction capability of the semantic segmentation model are improved, while the gradient vanishing problem is alleviated.

[0051] In some optional embodiments, the semantic segmentation model may be a deep learning model improved by nnU-Net. Specifically, the encoder in nnU-Net may be replaced with a residual encoder, and other modules remain unchanged, thereby obtaining a semantic segmentation model.

[0052] In the embodiment of the present invention, the gastric lesion segmentation result is a segmentation result of a gastric cancer lesion predicted by a semantic segmentation model, which is a gastric cancer lesion region in a three-dimensional computed tomography sample image.

[0053] S130. Determine a target model loss based on the gastric lesion segmentation result and the gastric lesion segmentation label, and update the model parameters of the semantic segmentation model based on the target model loss until a model training stop condition is met, thereby obtaining a trained gastric cancer lesion segmentation model.

[0054] In an embodiment of the present invention, the target model loss refers to the loss value calculated by the loss function, which can be used to update the model parameters of the semantic segmentation model to achieve supervised training of the semantic segmentation model, thereby obtaining a gastric cancer lesion segmentation model.

[0055] Specifically, the target model loss can be calculated by a single loss function or a mixed loss function, which is not specifically limited here. The loss function may include but is not limited to a mean square error loss function, a Dice loss function, and a cross entropy loss function.

[0056] In some optional embodiments, a five-fold cross-validation method may be used for model training to ensure a comprehensive evaluation of the performance of the gastric cancer lesion segmentation model.

[0057] In some optional embodiments, in order to improve the global feature extraction capability of the gastric cancer lesion segmentation model, a shared parameter transfer learning strategy is introduced. Specifically, for convolution kernels of different sizes, when initializing the weights, the pre-trained convolution kernel parameters are converted to the corresponding size for initialization, such as using the three-way linear interpolation method to convert the original 3×3 convolution kernel parameters into 5×5 pre-trained parameters, helping the large convolution kernel network to overcome the performance saturation problem of limited data commonly seen in medical image segmentation.

[0058] The technical solution of the embodiment of the present invention obtains a three-dimensional computed tomography sample image and a gastric lesion segmentation label corresponding to the three-dimensional computed tomography sample image, and then inputs the three-dimensional computed tomography sample image into the semantic segmentation model to be trained to obtain a gastric lesion segmentation result, wherein the semantic segmentation model includes at least one residual coding module, at least one bottleneck module and at least one decoding module, and the residual coding module and the decoding module are jump-connected, and then the target model loss is determined based on the gastric lesion segmentation result and the gastric lesion segmentation label, and the model parameters of the semantic segmentation model are updated based on the target model loss until the model training stop condition is met, thereby obtaining a trained gastric cancer lesion segmentation model. The above technical solution optimizes the feature extraction capability of the model through the residual coding module, effectively improves the accuracy of the gastric cancer lesion segmentation model, realizes the automatic segmentation of gastric cancer lesions through the gastric cancer lesion segmentation model, and efficiently and accurately segments the gastric cancer lesion area.

[0059] Embodiment 2

[0060] Figure 2This is a flow chart of a method for training a gastric cancer lesion segmentation model provided in Embodiment 2 of the present invention. The method of this embodiment can be combined with each of the optional solutions in the method for training a gastric cancer lesion segmentation model provided in the above embodiments. The method for training a gastric cancer lesion segmentation model provided in this embodiment is further optimized. Optionally, the method of obtaining a three-dimensional computed tomography sample image and a gastric lesion segmentation label corresponding to the three-dimensional computed tomography sample image includes: obtaining a three-dimensional computed tomography image, performing a preprocessing operation on the three-dimensional computed tomography image, and obtaining a preprocessed three-dimensional computed tomography image, wherein the preprocessing operation includes: spatially redirecting the three-dimensional computed tomography image; resampling the three-dimensional computed tomography image; performing foreground segmentation on the three-dimensional computed tomography image; inputting the preprocessed three-dimensional computed tomography image into a pre-trained gastric lesion segmentation model to obtain a gastric lesion segmentation label; obtaining a preset computed tomography value range interval, and cropping the preprocessed three-dimensional computed tomography image based on the preset computed tomography value range interval to obtain a cropped three-dimensional computed tomography image; and performing voxel intensity normalization on the cropped three-dimensional computed tomography image to obtain a three-dimensional computed tomography sample image.

[0061] like Figure 2 As shown, the method includes:

[0062] S210, acquiring a three-dimensional computed tomography image, and performing a preprocessing operation on the three-dimensional computed tomography image to obtain a preprocessed three-dimensional computed tomography image, wherein the preprocessing operation includes: spatially redirecting the three-dimensional computed tomography image; resampling the three-dimensional computed tomography image; and performing foreground segmentation on the three-dimensional computed tomography image.

[0063] The three-dimensional computed tomography image may be a three-dimensional CT image that meets the inclusion criteria.

[0064] It should be noted that spatial redirection is used to adjust the direction of the 3D CT image to ensure the consistency of multiple 3D CT images. Resampling is used to adjust the spatial resolution of the 3D CT image to ensure the consistency of multiple 3D CT images. Foreground segmentation is used to extract the foreground area of ​​the stomach and surrounding tissues in the 3D CT image to reduce the interference of irrelevant background on model training.

[0065] S220, inputting the preprocessed three-dimensional computed tomography image into a pre-trained gastric lesion segmentation model to obtain a gastric lesion segmentation label.

[0066] Among them, the pre-trained gastric lesion segmentation model can be a pre-trained model obtained through self-supervised training of a large amount of data in an open source project. For example, the pre-trained gastric lesion segmentation model can be a U-Net or nnU-Net model, etc., which is not specifically limited here.

[0067] In some optional embodiments, after the gastric lesion segmentation label is obtained, the gastric lesion segmentation label may be checked and corrected to ensure the accuracy and consistency of the gastric lesion segmentation label.

[0068] S230: Obtain a preset computed tomography value range, and crop the preprocessed three-dimensional computed tomography image based on the preset computed tomography value range to obtain a cropped three-dimensional computed tomography image.

[0069] Among them, the preset computed tomography value range interval is a preset effective CT value range.

[0070] It should be noted that by cropping the preprocessed three-dimensional computed tomography image by a preset computed tomography value range, the effective CT value range of the preprocessed three-dimensional CT image can be captured, abnormal CT values ​​and artifacts can be removed, and the three-dimensional CT image can be enhanced.

[0071] S240 , normalizing the voxel intensity of the cropped three-dimensional computed tomography image to obtain a three-dimensional computed tomography sample image.

[0072] Specifically, mapping the voxel values ​​of the cropped three-dimensional CT image to a uniform range can eliminate the intensity differences between the three-dimensional CT images, thereby providing standardized and high-quality three-dimensional computed tomography sample images for model training.

[0073] S250. Input the three-dimensional computed tomography sample image into the semantic segmentation model to be trained to obtain a gastric lesion segmentation result, wherein the semantic segmentation model includes at least one residual encoding module, at least one bottleneck module and at least one decoding module, and the residual encoding module is jump-connected to the decoding module.

[0074] S260. Determine a target model loss based on the gastric lesion segmentation result and the gastric lesion segmentation label, and update the model parameters of the semantic segmentation model based on the target model loss until a model training stop condition is met, thereby obtaining a trained gastric cancer lesion segmentation model.

[0075] The technical solution of the embodiment of the present invention obtains a three-dimensional computed tomography image, performs a preprocessing operation on the three-dimensional computed tomography image, obtains a preprocessed three-dimensional computed tomography image, and then inputs the preprocessed three-dimensional computed tomography image into a pre-trained gastric lesion segmentation model to obtain a gastric lesion segmentation label, and then obtains a preset computed tomography value range interval, and based on the preset computed tomography value range interval, the preprocessed three-dimensional computed tomography image is cropped to obtain a cropped three-dimensional computed tomography image, and then the cropped three-dimensional computed tomography image is voxel-intensity normalized to obtain a three-dimensional computed tomography sample image. The above technical solution provides a high-quality data set for model training through operations such as data acquisition, preprocessing, image annotation, image enhancement, and image standardization.

[0076] Embodiment 3

[0077] Figure 3A flowchart of a method for training a gastric cancer lesion segmentation model provided in Embodiment 3 of the present invention, the method of this embodiment can be combined with the various optional schemes in the method for training a gastric cancer lesion segmentation model provided in the above embodiments. The training method for a gastric cancer lesion segmentation model provided in this embodiment is further optimized. Optionally, the three-dimensional computed tomography sample image is input into the semantic segmentation model to be trained to obtain a gastric lesion segmentation result, including: inputting the three-dimensional computed tomography sample image into the first residual coding module of the semantic segmentation model to be trained to obtain a first computed tomography coding feature image; inputting the first computed tomography coding feature image into the second residual coding module of the semantic segmentation model to be trained to obtain a second computed tomography coding feature image; inputting the second computed tomography coding feature image into the third residual coding module of the semantic segmentation model to be trained to obtain a third computed tomography coding feature image; inputting the third computed tomography coding feature image into the fourth residual coding module of the semantic segmentation model to be trained to obtain a fourth computed tomography coding feature image; inputting the fourth computed tomography coding feature image into the semantic segmentation model to be trained at least one bottleneck module of the semantic segmentation model to be trained to obtain a computed tomography bottleneck feature image; the computed tomography bottleneck feature image and the fourth computed tomography coding feature image are input into the fourth decoding module of the semantic segmentation model to be trained to obtain a fourth computed tomography decoding feature image; the fourth computed tomography decoding feature image and the third computed tomography coding feature image are input into the third decoding module of the semantic segmentation model to be trained to obtain a third computed tomography decoding feature image; the third computed tomography decoding feature image and the second computed tomography coding feature image are input into the second decoding module of the semantic segmentation model to be trained to obtain a second computed tomography decoding feature image; the second computed tomography decoding feature image and the first computed tomography coding feature image are input into the first decoding module of the semantic segmentation model to be trained to obtain a gastric lesion segmentation result.

[0078] like Figure 3 As shown, the method includes:

[0079] S301, obtaining a three-dimensional computed tomography sample image and a gastric lesion segmentation label corresponding to the three-dimensional computed tomography sample image.

[0080] S302: Input the three-dimensional computed tomography sample image into a first residual coding module of a semantic segmentation model to be trained to obtain a first computed tomography coding feature image.

[0081] S303: Input the first computed tomography encoding feature image into a second residual coding module of a semantic segmentation model to be trained to obtain a second computed tomography encoding feature image.

[0082] S304: Input the second computed tomography encoding feature image into a third residual coding module of a semantic segmentation model to be trained to obtain a third computed tomography encoding feature image.

[0083] S305. Input the third computed tomography encoding feature image into a fourth residual coding module of a semantic segmentation model to be trained to obtain a fourth computed tomography encoding feature image.

[0084] S306. Input the fourth computed tomography encoding feature image into at least one bottleneck module of the semantic segmentation model to be trained to obtain a computed tomography bottleneck feature image.

[0085] S307. Input the computed tomography bottleneck feature image and the fourth computed tomography encoding feature image into a fourth decoding module of the semantic segmentation model to be trained to obtain a fourth computed tomography decoding feature image.

[0086] S308. Input the fourth computed tomography decoding feature image and the third computed tomography encoding feature image into a third decoding module of the semantic segmentation model to be trained to obtain a third computed tomography decoding feature image.

[0087] S309. Input the third computed tomography decoding feature image and the second computed tomography encoding feature image into a second decoding module of the semantic segmentation model to be trained to obtain a second computed tomography decoding feature image.

[0088] S310. Input the second computed tomography decoding feature image and the first computed tomography encoding feature image into a first decoding module of a semantic segmentation model to be trained to obtain a gastric lesion segmentation result.

[0089] S311. Determine the target model loss based on the gastric lesion segmentation result and the gastric lesion segmentation label, and update the model parameters of the semantic segmentation model based on the target model loss until the model training stop condition is met, thereby obtaining a trained gastric cancer lesion segmentation model.

[0090] In an embodiment of the present invention, Figure 4This is a structural schematic diagram of a gastric cancer lesion segmentation model provided according to an embodiment of the present invention, wherein the first residual encoding module is jump-connected with the first decoding module, the second residual encoding module is jump-connected with the second decoding module, the third residual encoding module is jump-connected with the third decoding module, and the fourth residual encoding module is jump-connected with the fourth decoding module. The number of bottleneck modules can be one or more, which is not specifically limited here.

[0091] It should be noted that the four residual encoding modules, the bottleneck module and the four decoding modules jump-connected to the residual encoding module improve the network depth and feature extraction capability, alleviate the gradient vanishing problem, and effectively improve the segmentation accuracy of the gastric cancer lesion segmentation model.

[0092] Optionally, the residual coding module includes: a first basic residual module and at least one second basic residual module, the first basic residual module and the second basic residual module are spliced ​​and connected, the first basic residual module includes a first convolution layer, a second convolution layer, a third convolution layer and a downsampling layer, and the second basic residual module includes a fourth convolution layer, a fifth convolution layer and a sixth convolution layer.

[0093] In an embodiment of the present invention, Figure 5 is a structural schematic diagram of a first basic residual module provided according to an embodiment of the present invention; Figure 6 is a structural schematic diagram of a second basic residual module provided according to an embodiment of the present invention.

[0094] Exemplarily, the first convolution layer may be a 1×1 convolution, the second convolution layer may be a 3×3 convolution, the third convolution layer may be a 1×1 convolution, and the downsampling layer may be a 1×1 convolution. The fourth convolution layer may be a 1×1 convolution, the fifth convolution layer may be a 3×3 convolution, and the sixth convolution layer may be a 1×1 convolution. One or more second basic residual modules may be spliced ​​after the first basic residual module, and each layer in the first basic residual module and the second basic residual module may also include normalization processing and nonlinear processing.

[0095] Optionally, determining a target model loss based on the gastric lesion segmentation result and the gastric lesion segmentation label includes: determining a segmentation task loss based on the gastric lesion segmentation result and the gastric lesion segmentation label; determining a classification task loss based on the fourth computed tomography encoding feature image and the gastric lesion segmentation label; determining a target deep supervision loss based on the computed tomography bottleneck feature image, the fourth computed tomography decoding feature image, the third computed tomography decoding feature image, the second computed tomography decoding feature image and the gastric lesion segmentation label; determining a target model loss based on the segmentation task loss, the classification task loss and the target deep supervision loss.

[0096] In the embodiment of the present invention, the segmentation task loss includes Dice loss and cross entropy loss. The calculation formula of the segmentation task loss can be as follows:

[0097] ;

[0098] ;

[0099] ;

[0100] ;

[0101] ;

[0102] in, represents the gastric lesion segmentation result corresponding to the i-th 3D computed tomography sample image, represents the gastric lesion segmentation label corresponding to the i-th 3D computed tomography sample image; is a hyperparameter, which can be 1 or other values. represents the Dice loss, represents the cross entropy loss, represents the segmentation task loss, and N represents the number of 3D computed tomography sample images. It should be noted that the Dice loss and cross entropy loss are mixed to take into account both the integrity of regional segmentation and the classification accuracy.

[0103] The calculation formula for classification task loss can be as follows:

[0104] ;

[0105] in, represents the fourth computed tomography encoding feature image corresponding to the i-th three-dimensional computed tomography sample image, represents the classification task loss.

[0106] The target deep supervision loss is the loss calculated based on the deep supervision strategy. The calculation formula for the target model loss can be:

[0107] ;

[0108] in, represents the target model loss, represents the target deep supervision loss.

[0109] Optionally, determining the target deep supervision loss based on the computed tomography bottleneck feature image, the fourth computed tomography decoding feature image, the third computed tomography decoding feature image, the second computed tomography decoding feature image and the gastric lesion segmentation label includes: determining a first deep supervision loss based on the computed tomography bottleneck feature image and the gastric lesion segmentation label; determining a second deep supervision loss based on the fourth computed tomography decoding feature image and the gastric lesion segmentation label; determining a third deep supervision loss based on the third computed tomography decoding feature image and the gastric lesion segmentation label; determining a fourth deep supervision loss based on the second computed tomography decoding feature image and the gastric lesion segmentation label; and determining a target deep supervision loss based on the first deep supervision loss, the second deep supervision loss, the third deep supervision loss and the fourth deep supervision loss.

[0110] In an embodiment of the present invention, the calculation formula of the target depth supervision loss can be as follows:

[0111] ;

[0112] ;

[0113] ;

[0114] ;

[0115] ;

[0116] in, represents the first deep supervision loss, represents the second deep supervision loss, represents the third deep supervision loss, represents the fourth deep supervision loss, represents the computed tomography bottleneck feature image corresponding to the i-th 3D computed tomography sample image, represents the fourth computed tomography decoding feature image corresponding to the i-th three-dimensional computed tomography sample image, represents the third computed tomography decoding feature image corresponding to the i-th three-dimensional computed tomography sample image, It represents the second computed tomography decoding feature image corresponding to the i-th three-dimensional computed tomography sample image.

[0117] It should be noted that by adopting a deep supervision strategy, supervision is applied to each layer in the upsampling stage, and the loss is calculated layer by layer, so that the model can learn more diverse feature representations at different resolutions and scales, which not only improves the recognition of intermediate layer features, but also improves the diversity of global features. In addition, the model's adaptability to changes in input data is enhanced, and it shows higher robustness to uncertain situations such as noise and missing data.

[0118] The technical solution of the embodiment of the present invention improves the network depth and feature extraction capability and alleviates the gradient vanishing problem through four residual coding modules, a bottleneck module and four decoding modules jump-connected to the residual coding module, thereby effectively improving the accuracy of the gastric cancer lesion segmentation model.

[0119] Embodiment 4

[0120] Figure 7 This is a flow chart of a gastric cancer lesion segmentation method provided in Embodiment 4 of the present invention. This embodiment is applicable to the case of identifying gastric cancer lesions in abdominal CT images. The method can be performed by a gastric cancer lesion segmentation device. The gastric cancer lesion segmentation device can be implemented in the form of hardware and / or software. The gastric cancer lesion segmentation device can be configured in a computer terminal. Figure 7 As shown, the method includes:

[0121] S410: Acquire a computed tomography image to be segmented.

[0122] The computer tomography image to be segmented may be a CT image of a new patient.

[0123] S420. Input the computed tomography image to be segmented into a trained gastric cancer lesion segmentation model to obtain an initial gastric cancer lesion segmentation result, wherein the gastric cancer lesion segmentation model is obtained after training according to the training method for the gastric cancer lesion segmentation model described in any embodiment of the present invention.

[0124] S430, performing morphological processing and / or maximum connected domain extraction on the initial gastric cancer lesion segmentation result to obtain a target gastric cancer lesion segmentation result.

[0125] Among them, morphological processing can fill the holes in the initial gastric cancer lesion segmentation results through closing operations and remove artifacts through opening operations to optimize the coherence of the segmented regions. Maximum connected domain extraction can retain the largest connected region and filter out isolated regions.

[0126] For example, Figure 8 1 is a flow chart of a gastric cancer lesion segmentation method provided according to an embodiment of the present invention. The gastric cancer lesion segmentation method comprises:

[0127] In the first step, the CT images that meet the inclusion criteria are preprocessed to obtain preprocessed CT images. The preprocessing includes spatial redirection, resampling, and foreground segmentation.

[0128] In the second step, the pre-trained gastric lesion segmentation model is used to automatically annotate the pre-processed CT images to obtain gastric lesion segmentation labels.

[0129] The third step is to determine whether the labeling is incorrect. If there is an error in the labeling, the gastric lesion segmentation label is manually corrected; if there is no error in the labeling, proceed to the fourth step.

[0130] The fourth step is to construct a semantic segmentation model based on the residual coding module. The semantic segmentation model includes at least one residual coding module, at least one bottleneck module and at least one decoding module. The residual coding module is jump-connected to the decoding module.

[0131] The fifth step is to conduct multi-task deep supervised training. Specifically, by adopting a deep supervision strategy, supervision is applied to each layer in the upsampling stage, and the loss is calculated layer by layer, so that the model can learn more diverse feature representations at different resolutions and scales, which not only improves the recognition of intermediate layer features, but also improves the diversity of global features. In addition, the model's adaptability to changes in input data is enhanced, and it shows higher robustness to uncertainties such as noise and missing data.

[0132] Step 6: Model application and post-processing. The CT image to be segmented is input into the trained gastric cancer lesion segmentation model to obtain the initial gastric cancer lesion segmentation result, and then the initial gastric cancer lesion segmentation result is subjected to post-processing operations such as morphological processing and / or maximum connected domain extraction to obtain a high-precision target gastric cancer lesion segmentation result.

[0133] The technical solution of the embodiment of the present invention realizes automatic segmentation of gastric cancer lesions through a gastric cancer lesion segmentation model, thereby efficiently and accurately segmenting the gastric cancer lesion area.

[0134] Embodiment 5

[0135] Fig. 9 This is a schematic diagram of the structure of a training device for a gastric cancer lesion segmentation model provided in Embodiment 5 of the present invention. Fig. 9 As shown, the device comprises:

[0136] A training sample acquisition module 510 is used to acquire a three-dimensional computed tomography sample image and a gastric lesion segmentation label corresponding to the three-dimensional computed tomography sample image;

[0137] a gastric lesion segmentation result prediction module 520, configured to input the three-dimensional computed tomography sample image into a semantic segmentation model to be trained to obtain a gastric lesion segmentation result, wherein the semantic segmentation model comprises at least one residual encoding module, at least one bottleneck module and at least one decoding module, and the residual encoding module is jump-connected with the decoding module;

[0138] The model parameter updating module 530 is used to determine the target model loss based on the gastric lesion segmentation result and the gastric lesion segmentation label, and update the model parameters of the semantic segmentation model based on the target model loss until the model training stop condition is met, thereby obtaining a trained gastric cancer lesion segmentation model.

[0139] The technical solution of the embodiment of the present invention obtains a three-dimensional computed tomography sample image and a gastric lesion segmentation label corresponding to the three-dimensional computed tomography sample image, and then inputs the three-dimensional computed tomography sample image into the semantic segmentation model to be trained to obtain a gastric lesion segmentation result, wherein the semantic segmentation model includes at least one residual coding module, at least one bottleneck module and at least one decoding module, and the residual coding module and the decoding module are jump-connected, and then the target model loss is determined based on the gastric lesion segmentation result and the gastric lesion segmentation label, and the model parameters of the semantic segmentation model are updated based on the target model loss until the model training stop condition is met, thereby obtaining a trained gastric cancer lesion segmentation model. The above technical solution optimizes the feature extraction capability of the model through the residual coding module, effectively improves the accuracy of the gastric cancer lesion segmentation model, realizes the automatic segmentation of gastric cancer lesions through the gastric cancer lesion segmentation model, and efficiently and accurately segments the gastric cancer lesion area.

[0140] In some optional implementations, the training sample acquisition module 510 may be specifically used to:

[0141] Acquire a three-dimensional computer tomography image, and perform a preprocessing operation on the three-dimensional computer tomography image to obtain a preprocessed three-dimensional computer tomography image, wherein the preprocessing operation includes: spatially redirecting the three-dimensional computer tomography image; resampling the three-dimensional computer tomography image; and performing foreground segmentation on the three-dimensional computer tomography image;

[0142] Inputting the preprocessed three-dimensional computed tomography image into a pre-trained gastric lesion segmentation model to obtain a gastric lesion segmentation label;

[0143] Acquiring a preset CT value range interval, and cropping the preprocessed three-dimensional CT image based on the preset CT value range interval to obtain a cropped three-dimensional CT image;

[0144] The cropped three-dimensional computed tomography image is subjected to voxel intensity normalization to obtain a three-dimensional computed tomography sample image.

[0145] In some optional implementations, the gastric lesion segmentation result prediction module 520 may be specifically used for:

[0146] Inputting the three-dimensional computed tomography sample image into a first residual coding module of a semantic segmentation model to be trained to obtain a first computed tomography coding feature image;

[0147] Inputting the first computed tomography encoding feature image into a second residual coding module of a semantic segmentation model to be trained to obtain a second computed tomography encoding feature image;

[0148] Inputting the second computed tomography encoding feature image into a third residual coding module of the semantic segmentation model to be trained to obtain a third computed tomography encoding feature image;

[0149] Inputting the third computed tomography encoding feature image into a fourth residual coding module of the semantic segmentation model to be trained to obtain a fourth computed tomography encoding feature image;

[0150] Inputting the fourth computed tomography encoding feature image into at least one bottleneck module of the semantic segmentation model to be trained to obtain a computed tomography bottleneck feature image;

[0151] Inputting the computed tomography bottleneck feature image and the fourth computed tomography encoding feature image into a fourth decoding module of the semantic segmentation model to be trained to obtain a fourth computed tomography decoding feature image;

[0152] Inputting the fourth computed tomography decoding feature image and the third computed tomography encoding feature image into a third decoding module of the semantic segmentation model to be trained to obtain a third computed tomography decoding feature image;

[0153] Inputting the third computed tomography decoding feature image and the second computed tomography encoding feature image into a second decoding module of a semantic segmentation model to be trained to obtain a second computed tomography decoding feature image;

[0154] The second computed tomography decoding feature image and the first computed tomography encoding feature image are input into a first decoding module of a semantic segmentation model to be trained to obtain a gastric lesion segmentation result.

[0155] In some optional embodiments, the residual encoding module includes: a first basic residual module and at least one second basic residual module, the first basic residual module and the second basic residual module are spliced ​​and connected, the first basic residual module includes a first convolution layer, a second convolution layer, a third convolution layer and a downsampling layer, and the second basic residual module includes a fourth convolution layer, a fifth convolution layer and a sixth convolution layer.

[0156] In some optional implementations, the model parameter updating module 530 includes:

[0157] a segmentation task loss determination unit, configured to determine the segmentation task loss based on the gastric lesion segmentation result and the gastric lesion segmentation label;

[0158] a classification task loss determination unit, configured to determine the classification task loss based on the fourth computed tomography encoding feature image and the gastric lesion segmentation label;

[0159] a deep supervision loss determination unit, configured to determine a target deep supervision loss based on the computed tomography bottleneck feature image, the fourth computed tomography decoding feature image, the third computed tomography decoding feature image, the second computed tomography decoding feature image, and the gastric lesion segmentation label;

[0160] A target model loss determination unit is used to determine the target model loss based on the segmentation task loss, the classification task loss and the target deep supervision loss.

[0161] In some optional implementations, the deep supervision loss determination unit may be specifically used to:

[0162] Determining a first deep supervision loss based on the computed tomography bottleneck feature image and the gastric lesion segmentation label;

[0163] determining a second deep supervision loss based on the fourth computed tomography decoded feature image and the gastric lesion segmentation label;

[0164] determining a third deep supervision loss based on the third computed tomography decoded feature image and the gastric lesion segmentation label;

[0165] determining a fourth deep supervision loss based on the second computed tomography decoded feature image and the gastric lesion segmentation label;

[0166] A target deep supervision loss is determined based on the first deep supervision loss, the second deep supervision loss, the third deep supervision loss, and the fourth deep supervision loss.

[0167] The training device for the gastric cancer lesion segmentation model provided in the embodiment of the present invention can execute the training method for the gastric cancer lesion segmentation model provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0168] Embodiment 6

[0169] Fig.10 This is a schematic diagram of the structure of a gastric cancer lesion segmentation device provided in Example 6 of the present invention. Fig.10 As shown, the device comprises:

[0170] A computer tomography image acquisition module 610, used to acquire a computer tomography image to be segmented;

[0171] an initial gastric cancer lesion segmentation result prediction module 620, configured to input the computed tomography image to be segmented into a trained gastric cancer lesion segmentation model to obtain an initial gastric cancer lesion segmentation result, wherein the gastric cancer lesion segmentation model is obtained by training according to the gastric cancer lesion segmentation model training method described in any embodiment of the present invention;

[0172] The target gastric cancer lesion segmentation result determination module 630 is used to perform morphological processing and / or maximum connected domain extraction on the initial gastric cancer lesion segmentation result to obtain a target gastric cancer lesion segmentation result.

[0173] The technical solution of the embodiment of the present invention realizes automatic segmentation of gastric cancer lesions through a gastric cancer lesion segmentation model, thereby efficiently and accurately segmenting the gastric cancer lesion area.

[0174] The gastric cancer lesion segmentation device provided in the embodiment of the present invention can execute the gastric cancer lesion segmentation method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0175] Embodiment 7

[0176] Fig.11 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0177] like Fig.11As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The I / O interface 15 is also connected to the bus 14.

[0178] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0179] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 11 executes the various methods and processes described above, such as a training method for a gastric cancer lesion segmentation model, which includes:

[0180] Acquire a three-dimensional computed tomography sample image and a gastric lesion segmentation label corresponding to the three-dimensional computed tomography sample image;

[0181] Inputting the three-dimensional computed tomography sample image into a semantic segmentation model to be trained to obtain a gastric lesion segmentation result, wherein the semantic segmentation model includes at least one residual encoding module, at least one bottleneck module and at least one decoding module, and the residual encoding module is jump-connected with the decoding module;

[0182] The target model loss is determined based on the gastric lesion segmentation result and the gastric lesion segmentation label, and the model parameters of the semantic segmentation model are updated based on the target model loss until the model training stop condition is met, thereby obtaining a trained gastric cancer lesion segmentation model.

[0183] In some embodiments, the training method of the gastric cancer lesion segmentation model or the gastric cancer lesion segmentation method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the training method of the gastric cancer lesion segmentation model or the gastric cancer lesion segmentation method described above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the training method of the gastric cancer lesion segmentation model or the gastric cancer lesion segmentation method by any other appropriate means (for example, by means of firmware).

[0184] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0185] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0186] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, 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 foregoing.

[0187] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0188] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0189] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0190] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0191] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A training method for a gastric cancer lesion segmentation model, characterized in that: include: Acquire a three-dimensional computed tomography sample image and a gastric lesion segmentation label corresponding to the three-dimensional computed tomography sample image; Inputting the three-dimensional computed tomography sample image into a semantic segmentation model to be trained to obtain a gastric lesion segmentation result, wherein the semantic segmentation model includes at least one residual encoding module, at least one bottleneck module and at least one decoding module, and the residual encoding module is jump-connected with the decoding module; The target model loss is determined based on the gastric lesion segmentation result and the gastric lesion segmentation label, and the model parameters of the semantic segmentation model are updated based on the target model loss until the model training stop condition is met, thereby obtaining a trained gastric cancer lesion segmentation model.

2. The method according to claim 1, characterized in that The step of acquiring a three-dimensional computer tomography sample image and a gastric lesion segmentation label corresponding to the three-dimensional computer tomography sample image comprises: Acquire a three-dimensional computer tomography image, and perform a preprocessing operation on the three-dimensional computer tomography image to obtain a preprocessed three-dimensional computer tomography image, wherein the preprocessing operation includes: spatially redirecting the three-dimensional computer tomography image; resampling the three-dimensional computer tomography image; and performing foreground segmentation on the three-dimensional computer tomography image; Inputting the preprocessed three-dimensional computed tomography image into a pre-trained gastric lesion segmentation model to obtain a gastric lesion segmentation label; Acquiring a preset CT value range interval, and cropping the preprocessed three-dimensional CT image based on the preset CT value range interval to obtain a cropped three-dimensional CT image; The cropped three-dimensional computed tomography image is subjected to voxel intensity normalization to obtain a three-dimensional computed tomography sample image.

3. The method according to claim 1, characterized in that The step of inputting the three-dimensional computer tomography sample image into the semantic segmentation model to be trained to obtain a gastric lesion segmentation result includes: Inputting the three-dimensional computed tomography sample image into a first residual coding module of a semantic segmentation model to be trained to obtain a first computed tomography coding feature image; Inputting the first computed tomography encoding feature image into a second residual coding module of a semantic segmentation model to be trained to obtain a second computed tomography encoding feature image; Inputting the second computed tomography encoding feature image into a third residual coding module of the semantic segmentation model to be trained to obtain a third computed tomography encoding feature image; Inputting the third computed tomography encoding feature image into a fourth residual coding module of the semantic segmentation model to be trained to obtain a fourth computed tomography encoding feature image; Inputting the fourth computed tomography encoding feature image into at least one bottleneck module of the semantic segmentation model to be trained to obtain a computed tomography bottleneck feature image; Inputting the computed tomography bottleneck feature image and the fourth computed tomography encoding feature image into a fourth decoding module of the semantic segmentation model to be trained to obtain a fourth computed tomography decoding feature image; Inputting the fourth computed tomography decoding feature image and the third computed tomography encoding feature image into a third decoding module of the semantic segmentation model to be trained to obtain a third computed tomography decoding feature image; Inputting the third computed tomography decoding feature image and the second computed tomography encoding feature image into a second decoding module of a semantic segmentation model to be trained to obtain a second computed tomography decoding feature image; The second computed tomography decoding feature image and the first computed tomography encoding feature image are input into a first decoding module of a semantic segmentation model to be trained to obtain a gastric lesion segmentation result.

4. The method according to claim 1 or 3, characterized in that: The residual coding module includes: a first basic residual module and at least one second basic residual module, the first basic residual module and the second basic residual module are spliced ​​and connected, the first basic residual module includes a first convolution layer, a second convolution layer, a third convolution layer and a downsampling layer, and the second basic residual module includes a fourth convolution layer, a fifth convolution layer and a sixth convolution layer.

5. The method according to claim 3, characterized in that: The determining the target model loss based on the gastric lesion segmentation result and the gastric lesion segmentation label comprises: determining a segmentation task loss based on the gastric lesion segmentation result and the gastric lesion segmentation label; determining a classification task loss based on the fourth computed tomography encoded feature image and the gastric lesion segmentation label; Determining a target deep supervision loss based on the computed tomography bottleneck feature image, the fourth computed tomography decoding feature image, the third computed tomography decoding feature image, the second computed tomography decoding feature image, and the gastric lesion segmentation label; A target model loss is determined based on the segmentation task loss, the classification task loss, and the target deep supervision loss.

6. The method according to claim 5, characterized in that The determining of the target deep supervision loss based on the computed tomography bottleneck feature image, the fourth computed tomography decoding feature image, the third computed tomography decoding feature image, the second computed tomography decoding feature image and the gastric lesion segmentation label comprises: Determining a first deep supervision loss based on the computed tomography bottleneck feature image and the gastric lesion segmentation label; determining a second deep supervision loss based on the fourth computed tomography decoded feature image and the gastric lesion segmentation label; determining a third deep supervision loss based on the third computed tomography decoded feature image and the gastric lesion segmentation label; determining a fourth deep supervision loss based on the second computed tomography decoded feature image and the gastric lesion segmentation label; A target deep supervision loss is determined based on the first deep supervision loss, the second deep supervision loss, the third deep supervision loss, and the fourth deep supervision loss.

7. A gastric cancer lesion segmentation method, characterized in that: include: acquiring a computed tomography image to be segmented; Inputting the computed tomography image to be segmented into a trained gastric cancer lesion segmentation model to obtain an initial gastric cancer lesion segmentation result, wherein the gastric cancer lesion segmentation model is obtained after training according to the training method for the gastric cancer lesion segmentation model according to any one of claims 1 to 6; The initial gastric cancer lesion segmentation result is subjected to morphological processing and / or maximum connected domain extraction to obtain a target gastric cancer lesion segmentation result.

8. A training device for a gastric cancer lesion segmentation model, characterized in that: include: A training sample acquisition module, used to acquire a three-dimensional computer tomography sample image and a gastric lesion segmentation label corresponding to the three-dimensional computer tomography sample image; a gastric lesion segmentation result prediction module, used for inputting the three-dimensional computed tomography sample image into a semantic segmentation model to be trained to obtain a gastric lesion segmentation result, wherein the semantic segmentation model comprises at least one residual encoding module, at least one bottleneck module and at least one decoding module, and the residual encoding module is jump-connected with the decoding module; A model parameter updating module is used to determine the target model loss based on the gastric lesion segmentation result and the gastric lesion segmentation label, and to update the model parameters of the semantic segmentation model based on the target model loss until the model training stop condition is met, thereby obtaining a trained gastric cancer lesion segmentation model.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; Wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the training method for the gastric cancer lesion segmentation model described in any one of claims 1-6, or execute the gastric cancer lesion segmentation method described in claim 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the training method for a gastric cancer lesion segmentation model according to any one of claims 1 to 6, or to implement the gastric cancer lesion segmentation method according to claim 7 when executed.

Images