CN115018768A - An automatic assessment system for stent neointimal coverage based on OCT platform - Google Patents

Abstract

The application is suitable for the technical field of medical image processing, and provides an automatic evaluation system for neointimal coverage rate of a stent based on an OCT platform. The system comprises: the device comprises an acquisition module, a processing module and an evaluation module; the acquisition module is used for acquiring an OCT image of a target lumen, and an embedded bracket is arranged in the target lumen; the processing module is used for inputting the OCT image into a preset image segmentation model for processing and outputting a mask image corresponding to the OCT image, and the mask image is used for marking a lumen and an embedded bracket; the evaluation module is used for calculating the coverage rate of the neointima between the lumen and the embedded bracket according to the mask image. The embedded support in the OCT image is identified by directly adopting the preset image segmentation model, so that the problem that the judgment result is inaccurate when the support and the lumen in the OCT image are judged to be the embedded support or not can be effectively avoided, the detection speed of the coverage degree of the embedded support is accelerated, and the accuracy of the judgment result is improved.

Description

An automatic assessment system for stent neointimal coverage based on OCT platform

technical field

The present application belongs to the technical field of medical image processing, and in particular relates to an automatic assessment system for stent neointimal coverage based on an OCT platform.

Background technique

Generally, stents in blood vessels include embedded stents and non-embedded stents. In the prior art, the stent and lumen in the collected optical coherence tomography (OCT) images need to be identified first. For the recognized stent, the distance between the stent and the lumen is measured, and whether the recognized stent is an embedded stent is determined by comparing the distance between the stent and the lumen with the preset distance threshold, and then the embedded stent is determined according to the judgment result. The status of stent healing, so as to formulate reasonable and effective medical orders.

However, in the above method, if there is an error in the recognition result, the distance between the measured stent and the lumen will be inaccurate, which will affect the judgment result of the embedded stent, reduce the recognition accuracy, and then affect the doctor's evaluation of the healing status of the embedded stent. judgment.

SUMMARY OF THE INVENTION

The present application provides an automatic evaluation system for stent neointimal coverage based on an OCT platform, which can obtain the neointimal coverage between the lumen and the embedded stent by analyzing the collected OCT images.

In a first aspect, the present application provides an automatic evaluation system for stent neointimal coverage based on an OCT platform. The system includes: an acquisition module, a processing module, and an evaluation module. The acquisition module is used to acquire an OCT image of a target lumen, and the target tube There is an embedded stent in the cavity; the processing module is used to input the OCT image into the preset image segmentation model for processing, and output the mask image corresponding to the OCT image, and the mask image is used to identify the lumen and the embedded stent; the evaluation module uses To assess the neointimal coverage between the lumen and the embedded stent based on the mask image.

In a possible implementation manner of the first aspect, the preset segmentation model includes a first convolution layer, a first downsampling module, a second downsampling module, a first upsampling module, a second upsampling module, a third The upsampling module and the first activation layer, the first convolutional layer, the first downsampling module, the second downsampling module, and the first upsampling module are connected in sequence, and the feature map output by the first upsampling module is the same as the first downsampling module. The feature map output by the sampling module is input to the second up-sampling module after being processed by the first splicing operation and the second activation function; the feature map output by the first down-sampling module is input to the feature map and the third up-sampling module. The feature map output by a convolution layer and the feature map processed by the second upsampling module are processed by the second stitching operation and the first activation function and output to obtain a mask image.

In a possible implementation manner of the first aspect, the first downsampling module and the second downsampling module include a downsampling layer and a second convolutional layer, respectively, a first upsampling module, a second upsampling module, and a third upsampling module The sampling module includes an upsampling layer and a third convolutional layer respectively; the first convolutional layer, the second convolutional layer and the third convolutional layer all use the same mode convolution.

In a possible implementation manner of the first aspect, the training method of the image segmentation model includes: building a network model, and the network model includes an initial image segmentation model and a SE-ResNet model;

According to the preset loss function and training set, the network model is trained, and the initial image segmentation model is trained to obtain the trained image segmentation model. The SE-ResNet model is used to extract the predicted edge mask of the OCT image samples in the training set.

In a possible implementation of the first aspect, the training set includes OCT image samples, mask image samples corresponding to each OCT image sample, and edge mask samples corresponding to each OCT image sample; the loss function is used to constrain the OCT image The error between the predicted image corresponding to the sample and the mask image sample, and the error between the predicted edge mask and the edge mask sample corresponding to the OCT image sample, the predicted image is the initial image segmentation model to obtain after processing the OCT image sample The image, the predicted edge mask is the image obtained after processing the OCT image sample by the SE-ResNet model.

In a possible implementation manner of the first aspect, the above SE-ResNet model includes a fourth convolution layer, a compression layer, and an excitation layer that are connected in sequence,

The output of the fourth convolutional layer and the output of the excitation layer are processed by weighting operation, and the output after the weighting operation and the input of the fourth convolutional layer are processed by addition operation, and then the edge is output.

In a possible implementation manner of the first aspect, the scaling factor in the above excitation layer is 16.

In a second aspect, the present application provides a method for detecting neointimal coverage, which is applied to the automatic assessment of the neointimal coverage of a stent based on the OCT platform described in the first aspect or any optional manner of the first aspect. In the system, the method includes:

Obtain an OCT image of the target lumen, and the target lumen is provided with an embedded stent;

Input the OCT image into the preset image segmentation model for processing, and output the mask image corresponding to the OCT image, and the mask image is used to identify the lumen and the embedded stent;

The neointimal coverage between the lumen and the embedded stent was assessed based on the mask image.

In a possible implementation manner of the second aspect, the preset segmentation model includes a first convolution layer, a first downsampling module, a second downsampling module, a first upsampling module, a second upsampling module, a third The upsampling module and the first activation layer, the first convolutional layer, the first downsampling module, the second downsampling module, and the first upsampling module are connected in sequence, and the feature map output by the first upsampling module is the same as the first downsampling module. The feature map output by the sampling module is input to the second up-sampling module after being processed by the first splicing operation and the second activation function; the feature map output by the first down-sampling module is input to the feature map and the third up-sampling module. The feature map output by a convolution layer and the feature map processed by the second upsampling module are processed by the second stitching operation and the first activation function and output to obtain a mask image.

In a possible implementation manner of the second aspect, the first downsampling module and the second downsampling module respectively include a downsampling layer and a second convolutional layer, a first upsampling module, a second upsampling module and a third upsampling module The sampling module includes an upsampling layer and a third convolutional layer respectively; the first convolutional layer, the second convolutional layer and the third convolutional layer all use the same mode convolution.

In a possible implementation manner of the second aspect, the training method of the image segmentation model includes: building a network model, and the network model includes an initial image segmentation model and a SE-ResNet model;

According to the preset loss function and training set, the network model is trained, and the initial image segmentation model is trained to obtain the trained image segmentation model. The SE-ResNet model is used to extract the predicted edge mask of the OCT image samples in the training set.

In a possible implementation manner of the second aspect, the training set includes OCT image samples, mask image samples corresponding to each OCT image sample, and edge mask samples corresponding to each OCT image sample; the loss function is used to constrain the OCT image The error between the predicted image corresponding to the sample and the mask image sample, and the error between the predicted edge mask and the edge mask sample corresponding to the OCT image sample, the predicted image is the initial image segmentation model to obtain after processing the OCT image sample The image, the predicted edge mask is the image obtained after processing the OCT image sample by the SE-ResNet model.

In a possible implementation manner of the second aspect, the above SE-ResNet model includes a fourth convolution layer, a compression layer, and an excitation layer that are connected in sequence,

The output of the fourth convolutional layer and the output of the excitation layer are processed by weighting operation, and the output after the weighting operation and the input of the fourth convolutional layer are processed by addition operation, and then the edge is output.

In a possible implementation manner of the second aspect, the scaling factor in the above excitation layer is 16.

In a third aspect, the present application provides a terminal device, including a memory, a processor, and a computer program stored in the memory and running on the processor, where the processor implements any of the second aspect or the second aspect when the processor executes the computer program. The method described in an optional manner.

In a fourth aspect, a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, implements the method described in the second aspect or any optional manner of the second aspect.

In a fifth aspect, an embodiment of the present application provides a computer program product, which when the computer program product runs on a terminal device, enables the terminal device to execute the method described in the second aspect or any optional manner of the second aspect.

The beneficial effects that the embodiments of the present application have compared with the prior art are:

An automatic evaluation system for stent neointimal coverage based on an OCT platform provided in this application, the system directly uses a preset image segmentation model to process the OCT image of the target lumen to obtain a mask image corresponding to the OCT image. , and then evaluate the neointimal coverage between the lumen and the embedded stent according to the above mask image. Directly using the preset image segmentation model to identify the embedded stent in the OCT image can effectively avoid using the identification results of the stent and lumen in the OCT image to determine whether the stent is an embedded stent, resulting in inaccurate judgment results and affecting the doctor's judgment. The problem of the healing state of the embedded stent speeds up the detection of the coverage of the embedded stent and improves the accuracy of the judgment results.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only for the present application. In some embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without any creative effort.

1 is a schematic structural diagram of an image segmentation model provided by an embodiment of the present application;

FIG. 2 is a schematic flowchart of generating an image segmentation model provided by an embodiment of the present application;

3 is a schematic structural diagram of a SE-ResNet model provided by an embodiment of the present application;

4 is a schematic diagram of a result of processing the same image with different image segmentation models provided by an embodiment of the present application;

5 is a schematic diagram of an edge fitting image provided by an embodiment of the present application;

6 is a schematic diagram of an automatic assessment system for stent neointimal coverage based on an OCT platform provided by an embodiment of the present application;

FIG. 7 is a schematic diagram of a terminal device provided by an embodiment of the present application.

Detailed ways

Aiming at the problem of using the recognition result of the stent in the OCT image to determine whether the stent is an embedded stent, resulting in inaccurate judgment results and affecting the doctor's judgment on the healing status of the embedded stent, the present application provides a stent neointimal coverage rate based on the OCT platform. Automatic evaluation system, the system directly uses the preset image segmentation model to identify the embedded stent in the OCT image, which effectively avoids using the identification results of the stent and lumen in the OCT image to determine whether the stent is an embedded stent, resulting in the judgment result. Inaccurate, the problem that affects the doctor's judgment on the healing state of the embedded stent, speeds up the detection speed of the coverage of the embedded stent, and improves the accuracy of the judgment result.

The technical solutions of the present application will be described in detail below with specific examples. The following specific embodiments may be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

FIG. 1 is a schematic structural diagram of an image segmentation model provided by an embodiment of the present application. The image segmentation model can be deployed in an optical coherence tomography (OCT) image acquisition device, and can also be deployed in other control devices associated with the OCT image acquisition device. For example, the device deploying the image segmentation model can It is a mobile terminal such as a smartphone, a tablet computer, a camera, or other devices that can perform image segmentation, such as a desktop computer, a robot, and a server.

Referring to FIG. 1 , the image segmentation model provided by the embodiment of the present application is a densely connected U-Net model, and the densely connected U-Net model is used to extract a mask image fused with different feature information from an OCT image.

The depth of the densely connected U-Net model is 3 layers, specifically including a first convolution layer, a first downsampling module, a second downsampling module, a first upsampling module, a second activation function, and a second upsampling module , a third upsampling module, and a first activation function. The first convolutional layer, the first down-sampling module, the second down-sampling module and the first up-sampling module are connected in sequence, and the feature map output by the first up-sampling module and the feature map output by the first down-sampling module are subjected to a first splicing operation , the second activation function is processed and input to the second upsampling module; the feature map output by the first downsampling module is input to the feature map obtained after processing by the third upsampling module, the feature map output by the first convolution layer and the second The feature map processed by the up-sampling module is processed by the second stitching operation and the first activation function and then output to obtain the first mask image.

It should be understood that both the first downsampling module and the second downsampling module include a downsampling layer and a second convolutional layer, wherein the downsampling layer is implemented by a maximum pooling layer, and the size of the feature image processed by the downsampling layer is will shrink by half. Correspondingly, the first upsampling module, the second upsampling module and the third upsampling module all include an upsampling layer and a third convolutional layer.

In order to efficiently fuse the feature information between different convolutional layers, the size of the output feature map should be consistent with the size of the input feature map. In the embodiment of the present application, the first convolutional layer, the second convolutional layer in the downsampling module, and the third convolutional layer in the upsampling module all use the same mode convolution.

It should be noted that the image segmentation model provided in this application has generality. It can be applied to the field of medical image segmentation, and can extract organs, tissues, embedded stents, etc. in medical images, and complete the segmentation tasks of different organs. It can also be applied to other tasks where the image segmentation effect is the evaluation index.

It can be understood that, for different image segmentation tasks, the initial image segmentation model can be trained by designing corresponding training sets and loss functions, so as to obtain image segmentation models that can be applied to different image segmentation tasks.

According to actual application requirements, the execution subject for training the image segmentation model and the execution subject for performing the image segmentation task using the image segmentation model may be the same or different.

The following will exemplify the training process and effects of the image segmentation model provided by the present application by taking the segmentation task of lumen and embedded stent segmentation from OCT images as an example.

FIG. 2 shows a schematic flowchart of an image segmentation model generation provided by an embodiment of the present application. Referring to FIG. 2 , a network model is first constructed, and the network model includes an initial image segmentation model (that is, an initial U-Net model of intensive connection) and SE-ResNet model.

3 shows a schematic structural diagram of a SE-ResNet model provided by an embodiment of the present application. Referring to FIG. 3, the SE-ResNet model includes a fourth convolution layer, a compression layer, and an excitation layer. The fourth convolution layer The output of the layer and the output of the excitation layer are processed by a weighting operation, and the output after the weighting operation and the input of the fourth convolution layer are processed by an addition operation to output an edge mask.

It should be understood that the input OCT image is processed by the fourth convolution layer to obtain a feature map X, where X∈R ^C×W×H , C represents the number of channels of the feature map, W represents the width of the feature map, and H represents the feature map height of.

The obtained feature map X is input to the compression layer for processing, and the output is Z, where the calculation formula of Z is as follows (1), the compression layer adopts the global average pooling operation, that is, the feature map X input to the compression layer is subjected to global spatial information. Compression yields output Z.

In the above formula (1), W represents the width of the feature map, H represents the height of the feature map, and (i, j) represents the pixels in the feature map.

The Z obtained after compression processing is input to the excitation layer (including two sets of fully connected layers and activation functions set at intervals) to obtain the excitation value S, where the calculation formula of S is as follows:

S=σ(W ₂ δ(W ₁ Z)) (2)

In the above formula (2), W ₂ δ(W ₁ Z) and W ₁ Z represent two fully connected layers, σ is a sigmoid function, δ can be a ReLU (rectified linear unit) activation function, and W1 and W2 are two The parameter matrix of the fully connected layer.

It is worth noting that, for the lumen and embedded stent segmentation tasks, the scaling parameters of the excitation layer provided in this application are all 16, and the scaling parameters of the excitation layer can be determined according to different practical applications, which are not limited in this application.

Perform a weighting operation (scale operation) on the feature map obtained after processing the excitation value S and the second convolution layer, and add the feature map after the weighting operation to the input OCT image to obtain an edge mask X _out , where the edge The formula for calculating the mask X _out is as follows:

X _out =F _scale (X,S) (3)

It should be noted that in this SE-ResNet model, the compression layer performs feature compression on the feature map input to the global average pooling, converts the feature map of each channel into a real number, and outputs the feature map of the number of dimensions and the input. eigenvectors with equal number of feature channels. The excitation layer evaluates the weights of the feature maps of each channel, and obtains the weights corresponding to each channel. The weighting operation is to multiply the weight value of each channel in the scaled feature map with the two-dimensional matrix of the corresponding channel in the feature map obtained after processing by the fourth convolution layer to obtain the feature map after the weighting operation.

Next, for the segmentation task of the lumen and the embedded stent, a corresponding training set is collected, and the training set includes OCT image samples, mask image samples corresponding to the above OCT image samples, and edge mask samples corresponding to the above OCT image samples. Wherein, the acquisition method of the OCT image sample includes, but is not limited to, directly using the OCT image in the existing OCT image database and acquiring the OCT image from a public website.

In practical applications, the mask image samples corresponding to the above OCT image samples are manually marked image samples, which are used for comparison with the mask images output by the image segmentation model.

In one possible implementation manner, after the OCT image samples are acquired, image preprocessing is first performed on the OCT image samples. Exemplarily, the method of image preprocessing may include grayscale conversion, normalization, histogram equalization, and/or Gamma correction, and the like.

Then, the above-mentioned network model is iteratively trained according to the OCT image samples in the training set and the preset loss function, and the above-mentioned initial image segmentation model is trained to obtain a trained image segmentation model. The preset loss function is used to constrain the error between the predicted image corresponding to the OCT image sample and the mask image sample, and the error between the predicted edge mask corresponding to the OCT image sample and the edge mask sample, where the predicted image is The initial image segmentation model is the image obtained after processing the OCT image sample, and the predicted edge mask is the image obtained after processing the OCT image sample by the SE-ResNet model.

For the segmentation task of segmenting lumen and embedded stents from OCT images, the loss function in this embodiment of the present application adopts a combined loss function. The combined loss function includes a region-based Dice loss and an edge-based Boundary loss. The formula of the combined loss function See formula (4):

L=αL _Dice +βL _Boundary (4)

In the above formula (4), L _Dice represents the region-based Dice loss, L _Boundary represents the edge-based Boundary loss, α and β represent the balance coefficient, which is used to balance the impact of the region-based loss and the edge-based loss on the output results, L The calculation formula of _Dice is shown in the following formula (5), and the calculation formula of L _Boundary is shown in the following formula (6).

表示该像素点分为某一类的概率。In the above formula (5), i represents the pixel in the image, c represents the classification corresponding to the pixel, and g represents whether the classification of the pixel is correct,

Indicates the probability that the pixel is classified into a certain class.

In the above formula (6), φG represents the boundary level set, and S _θ (ξ) represents the probability of the SE-ResNet model output.

It should be understood that for the segmentation task of segmenting lumen and embedded stents from OCT images, the image segmentation model trained using the above-mentioned combined loss function can reduce the loss value of the image area as much as possible while supplementing the edge information in the image, effectively. Solve the problem that the segmentation loss value of the lumen image is large during segmentation, resulting in low image segmentation accuracy.

The densely connected U-Net model in the network model provided by the embodiment of the present application can fuse feature information of different depths. Since the Boundary loss in the above-mentioned combined loss function fully takes into account the edge loss in the image segmentation process, the SE-ResNet model is applied In the image segmentation model, the ability to reuse and express edge features can be enhanced. Therefore, using a network model composed of a densely connected U-Net model and a SE-ResNet model to segment the obtained OCT image can improve the model's ability to image images. improved segmentation accuracy and improved the identification of lumen and embedded stents.

Based on the segmentation task of segmenting the lumen and the embedded stent from the OCT image, as shown in FIG. 4 , a traditional U-Net image segmentation model and the image provided by the present application are respectively used for the same OCT image. The result of segmentation by the segmentation model, as shown in (a) in Figure 4 is the OCT image to be segmented, and (b) in Figure 4 is the result of segmenting the OCT image by using the traditional U-Net image segmentation model, (c) in FIG. 4 is the result of segmenting the OCT image by using the image segmentation model provided by the present application. According to FIG. 4 , not only the feasibility of the image segmentation model provided by the embodiment of the present application is verified, but also from FIG. 4 . It can be seen that the image segmented by using the image segmentation model provided by the embodiment of the present application is more accurate, and compared with the prior art, the accuracy of image segmentation can be obviously improved.

A mask image is obtained by segmenting the OCT image according to the image segmentation model provided in the embodiment of the present application, and the obtained mask image can be applied to the task of detecting the neointimal index. Specifically, the mask image can be used to calculate the neointimal index in the OCT image, and then the degree of coverage of the embedded stent in the OCT image can be determined according to the neointimal index, thereby completing the detection task.

It should be understood that the neointimal index represents the ratio of the area enclosed by the edge of the embedded stent to the area enclosed by the edge of the lumen in the OCT image. The specific calculation formula is shown in the following formula (7):

In the above formula (7), S _stent represents the area covered by the embedded stent in the OCT image; S _lumen represents the cross-sectional area of the lumen in the OCT image. If the ratio in formula (7) is larger, it means that the degree of closeness of the embedded stent to the blood vessel wall (also called adherence) is higher, and the embedded stent is deeply embedded in the blood vessel; on the contrary, if formula (7) The smaller the ratio, the lower the degree of adherence of the embedded stent in the blood vessel, and the shallower the embedded degree of the embedded stent in the blood vessel.

In the actual application process, as shown in FIG. 5, a schematic diagram of an edge fitting image provided by an embodiment of the present application is shown. Referring to FIG. 5, the mask image shown in (c) in FIG. 4 is fitted, corresponding to The closed edge of the embedded stent and the closed edge of the lumen are formed. As shown in Figure 5, the area enclosed by the dotted line is the closed edge of the embedded stent, and the area enclosed is the closed edge of the lumen. Then, by calculating the edge of the embedded stent in the mask image The enclosed area is S _stent , and S _lumen is obtained by calculating the area enclosed by the lumen edge in the mask image; finally, the neointimal index is calculated according to the above formula (7).

The support function of the embedded stent on the inner wall of the blood vessel is measured by calculating the ratio, which avoids the process of calculating the embedded depth of the embedded stent in the vascular lumen, because the manual measurement of the embedded stent embedded depth is inconsistent or overly dependent on the edge single. The problem of inaccurate calculation results caused by dot pixels speeds up the detection speed of the degree of endometrium covering the embedded stent, and improves the accuracy of the calculation results.

Based on the same inventive concept, as an implementation of the above method, an embodiment of the present application provides an OCT platform-based automatic evaluation system for the neointimal coverage of embedded stents. Referring to FIG. 6 , the OCT platform-based embedded stent neointimal coverage The film coverage automatic evaluation system system 600 includes:

an acquisition module 601, configured to acquire an OCT image of a target lumen, where an embedded stent is arranged in the target lumen;

The processing module 602 is used to input the OCT image into a preset image segmentation model for processing, and output a mask image corresponding to the OCT image, and the mask image is used to identify the lumen and the embedded stent;

The evaluation module 603 is used for evaluating the neointimal coverage between the lumen and the embedded stent according to the mask image.

Optionally, the preset segmentation model includes a first convolution layer, a first downsampling module, a second downsampling module, a first upsampling module, a second upsampling module, a third upsampling module, and a first activation layer. , the first convolution layer, the first down-sampling module, the second down-sampling module, and the first up-sampling module are connected in sequence, and the feature map output by the first up-sampling module and the feature map output by the first down-sampling module are processed by the first A splicing operation and the second activation function are processed and input to the second upsampling module; the feature map output by the first downsampling module is input to the feature map processed by the third upsampling module and the feature map output by the first convolution layer And the feature map processed by the second upsampling module is processed by the second stitching operation and the first activation function and output to obtain a mask image.

Optionally, the first downsampling module and the second downsampling module include a downsampling layer and a second convolutional layer, respectively, and the first upsampling module, the second upsampling module, and the third upsampling module include an upsampling layer and The third convolutional layer; the first convolutional layer, the second convolutional layer and the third convolutional layer all use the same mode convolution.

Optionally, the training method of the image segmentation model includes: building a network model, and the network model includes an initial image segmentation model and a SE-ResNet model;

According to the preset loss function and training set, the network model is trained, and the initial image segmentation model is trained to obtain the trained image segmentation model. The SE-ResNet model is used to extract the predicted edge mask of the OCT image samples in the training set.

Optionally, the training set includes an OCT image sample, a mask image sample corresponding to each OCT image sample, and an edge mask sample corresponding to each OCT image sample; the loss function is used to constrain the predicted image and mask corresponding to the OCT image sample. The error between the image samples, and the error between the predicted edge mask corresponding to the OCT image sample and the edge mask sample, the predicted image is the image obtained after processing the OCT image sample by the initial image segmentation model, and the predicted edge mask is The image obtained after the SE-ResNet model processes the OCT image sample.

Optionally, the above SE-ResNet model includes a fourth convolution layer, a compression layer and an excitation layer that are connected in sequence,

The output of the fourth convolutional layer and the output of the excitation layer are processed by weighting operation, and the output after the weighting operation and the input of the fourth convolutional layer are processed by addition operation, and then the edge is output.

Optionally, the scaling factor in the above excitation layer is 16.

The system 600 for automatic assessment of stent neointimal coverage based on the OCT platform provided in this embodiment can implement the above method embodiments, and the implementation principle and technical effect thereof are similar, and are not repeated here.

Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. Module completion, that is, dividing the internal structure of the device into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment may be integrated in one processing unit, or each unit may exist physically alone, or two or more units may be integrated in one unit, and the above-mentioned integrated units may adopt hardware. It can also be realized in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing from each other, and are not used to limit the protection scope of the present application. For the specific working processes of the units and modules in the above-mentioned system, reference may be made to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

Based on the same inventive concept, an embodiment of the present application also provides a terminal device. FIG. 7 is a schematic diagram of a terminal device provided by an embodiment of the present application. As shown in FIG. 7 , the terminal device provided by this embodiment includes: a memory 701 and a processor 702. The memory 701 is used for storing a computer program 703; the processor 702 is used for The methods described in the above method embodiments are executed when the computer program is invoked. Alternatively, when the processor 702 executes the computer program 703, the functions of the modules/units in the foregoing device embodiments, for example, the functions of the units 601 to 603 shown in FIG. 6 are implemented.

Exemplarily, the computer program 703 may be divided into one or more modules/units, and the one or more modules/units are stored in the memory 701 and executed by the processor 702 to complete. this application. The one or more modules/units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

Those skilled in the art can understand that FIG. 7 is only an example of a terminal device, and does not constitute a limitation on the terminal device. It may include more or less components than the one shown in the figure, or combine some components, or different components, such as The terminal device 700 may further include an input and output device, a network access device, a bus, and the like.

The processor 702 may be a central processing unit (Central Processing Unit, CPU), or other general-purpose processors, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory 701 may be an internal storage unit of the terminal device, such as a hard disk or a memory of the terminal device. The memory 701 may also be an external storage device of the terminal device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card equipped on the terminal device, Flash card (Flash Card) and so on. Further, the memory 701 may also include both an internal storage unit of the terminal device and an external storage device. The memory 701 is used to store the computer program and other programs and data required by the terminal device. The memory 701 can also be used to temporarily store data that has been output or will be output.

The terminal device provided in this embodiment may execute the foregoing method embodiments, and the implementation principle and technical effect thereof are similar, and details are not described herein again.

Embodiments of the present application further provide a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method described in the foregoing method embodiment is implemented.

The embodiments of the present application further provide a computer program product, when the computer program product runs on a terminal device, the terminal device executes the method described in the above method embodiments.

An embodiment of the present application further provides a chip system, including a processor, where the processor is coupled to a memory, and the processor executes a computer program stored in the memory to implement the method described in the above method embodiments. Wherein, the chip system may be a single chip or a chip module composed of multiple chips.

If the above-mentioned integrated units are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on this understanding, the present application realizes all or part of the processes in the methods of the above embodiments, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When executed by a processor, the steps of each of the above method embodiments can be implemented. Wherein, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form, and the like. The computer-readable storage medium may include at least: any entity or device capable of carrying computer program codes to the photographing device/terminal device, recording medium, computer memory, read-only memory (ROM), random access Memory (Random Access Memory, RAM), electrical carrier signal, telecommunication signal and software distribution medium. For example, U disk, mobile hard disk, disk or CD, etc. In some jurisdictions, under legislation and patent practice, computer readable media may not be electrical carrier signals and telecommunications signals.

In the foregoing embodiments, the description of each embodiment has its own emphasis. For parts that are not described or described in detail in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of this application.

In the embodiments provided in this application, it should be understood that the disclosed apparatus/device and method may be implemented in other manners. For example, the apparatus/equipment embodiments described above are only illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or Components may be combined or may be integrated into another system, or some features may be omitted, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.

It is to be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described feature, integer, step, operation, element and/or component, but does not exclude one or more other The presence or addition of features, integers, steps, operations, elements, components and/or sets thereof.

It will also be understood that, as used in this specification and the appended claims, the term "and/or" refers to and including any and all possible combinations of one or more of the associated listed items.

As used in the specification of this application and the appended claims, the term "if" may be contextually interpreted as "when" or "once" or "in response to determining" or "in response to detecting ". Similarly, the phrases "if it is determined" or "if the [described condition or event] is detected" may be interpreted, depending on the context, to mean "once it is determined" or "in response to the determination" or "once the [described condition or event] is detected. ]" or "in response to detection of the [described condition or event]".

In addition, in the description of the specification of the present application and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the description, and should not be construed as indicating or implying relative importance.

References in this specification to "one embodiment" or "some embodiments" and the like mean that a particular feature, structure or characteristic described in connection with the embodiment is included in one or more embodiments of the present application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in other embodiments," etc. in various places in this specification are not necessarily All refer to the same embodiment, but mean "one or more but not all embodiments" unless specifically emphasized otherwise. The terms "including", "including", "having" and their variants mean "including but not limited to" unless specifically emphasized otherwise.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: The technical solutions described in the foregoing embodiments can still be modified, or some or all of the technical features thereof can be equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the technical solutions of the embodiments of the present application. scope.

Claims (10)

1. an automatic assessment system for stent neointimal coverage based on OCT platform, is characterized in that, described system comprises: acquisition module, processing module and assessment module; The acquisition module is used to acquire the OCT image of the target lumen, and the target lumen is provided with an embedded stent; The processing module is configured to input the OCT image into a preset image segmentation model for processing, and output a mask image corresponding to the OCT image, where the mask image is used to identify the lumen and the embedded stent ; The evaluation module is used for evaluating the neointimal coverage between the lumen and the embedded stent according to the mask image. 2. The system according to claim 1, wherein the preset image segmentation model comprises a first convolutional layer, a first downsampling module, a second downsampling module, a first upsampling module, a second An upsampling module, a third upsampling module and a first activation function, the first convolution layer, the first downsampling module, the second downsampling module, and the first upsampling module are connected in sequence, so The feature map output by the first upsampling module and the feature map output by the first downsampling module are input to the second upsampling module after being processed by the first splicing operation and the second activation function; the first downsampling The feature map output by the module is input to the feature map obtained after processing by the third upsampling module, the feature map output by the first convolutional layer, and the feature map processed by the second upsampling module. After the second splicing operation and outputting the mask image after the first activation function is processed. 3. The system according to claim 2, wherein the first downsampling module and the second downsampling module respectively comprise a downsampling layer and a second convolutional layer, the first upsampling module, The second upsampling module and the third upsampling module respectively comprise an upsampling layer and a third convolutional layer; The first convolutional layer, the second convolutional layer and the third convolutional layer all use the same mode convolution. 4. The system according to claim 2, wherein the training method of the image segmentation model comprises: Building a network model, the network model includes an initial image segmentation model and a SE-ResNet model; According to the preset loss function and training set, the network model is trained, the initial image segmentation model is trained to obtain the trained image segmentation model, and the SE-ResNet model is used to extract the OCT in the training set. Predicted edge mask for image samples. 5. The system according to claim 4, wherein the training set comprises the OCT image samples, a mask image sample corresponding to each of the OCT image samples, and an edge corresponding to each of the OCT image samples mask sample; The loss function is used to constrain the error between the predicted image corresponding to the OCT image sample and the mask image sample, and the difference between the predicted edge mask corresponding to the OCT image sample and the edge mask sample. error, the predicted image is an image obtained by processing the OCT image sample by the initial image segmentation model, and the predicted edge mask is obtained by processing the OCT image sample by the SE-ResNet model image. 6. The system according to claim 4, wherein the SE-ResNet model comprises a fourth convolution layer, a compression layer and an excitation layer connected in sequence, The output of the fourth convolution layer and the output of the excitation layer are subjected to a weighting operation, and the output after the weighting operation and the input of the fourth convolution layer are subjected to an addition operation to output the edge. 7 . The system of claim 6 , wherein the scaling factor in the excitation layer is 16. 8 . 8. a detection method of neoendometrial coverage, is characterized in that, described method is applied in the stent neoendome coverage automatic assessment system based on OCT platform as described in any one of claim 1-6, The methods described include: obtaining an OCT image of a target lumen, where an embedded stent is arranged in the target lumen; inputting the OCT image into a preset image segmentation model for processing, and outputting a mask image corresponding to the OCT image, where the mask image is used to identify the lumen and the embedded stent; The neointimal coverage between the lumen and the embedded stent is assessed from the mask image. 9. A terminal device, comprising a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor implements the computer program as claimed in the claims when executing the computer program 8 the method described. 10. A computer-readable storage medium storing a computer program, wherein the computer program implements the method according to claim 8 when executed by a processor.

Images