Multi-scale detection method and device for tumor in digestive tract endoscope image

Abstract

The invention relates to the technical field of medical detection, in particular to a multi-scale detection method and device for tumors in digestive tract endoscopic images, and the method comprises the steps: obtaining an initial segment image, a middle segment image and a tail segment image of an endoscopic tumor in a target region of a digestive tract, and preprocessing the images to obtain a first image set, a second image set and a third image set; performing stripping analysis on the image set to obtain a profile image set; performing gray processing on the contour image set to obtain a gray image set; inputting the gray-scale image sets into a plane coordinate system, performing edge cutting on each gray-scale image set, and obtaining coordinates of each image set after cutting; and inputting each image set coordinate into three-dimensional modeling software to execute reconstruction, obtaining a reconstructed target area endoscopic tumor image, and grading the image according to a preset tumor grade. According to the method, through the image maps of the initial segment, the middle segment and the tail segment of the endoscopic tumor in the target area of the digestive tract, the image set coordinates and the three-dimensional software are combined to obtain a visual tumor three-dimensional pattern, the tumor grade is detected and determined, and multi-scale detection is achieved.

Description

Technical Field

[0001] The present application relates to the field of medical detection technology, and in particular to a multi-scale detection method for tumors in gastrointestinal endoscopic images, and also to a device. Background Art

[0002] At present, computer technology has been widely used in the medical field. A digestive endoscopy image often contains a lot of information. At present, most of the digestive endoscopy image analysis algorithms based on graph theory take each pixel of the digestive endoscopy image as a node of a graph, and connect any adjacent pair of nodes to establish the edge of the graph. Therefore, this edge expresses the adjacency relationship between the image and the pixel in the digestive endoscopy image. After the adjacency relationship of the graph is established, the weight of each edge can be established by calculating the information of the pixels represented by the two adjacent nodes. The weight expresses the similarity between the pixels in the image. Existing improvements for digestive endoscopy image recognition are usually based on graph theory to extract the feature points of digestive endoscopy images and establish the corresponding internal structure of the image.

[0003] In the prior art, there is a Chinese invention patent application with a publication date of June 21, 2019, a publication number of CN109919212A, and a name of multi-scale detection method and device for tumors in gastrointestinal endoscopic images. It discloses marking tumor images for each gastrointestinal endoscopic image sample, and marking the tumor nature level of the gastrointestinal endoscopic image samples marked as tumor images according to a preset tumor nature level; training a gastrointestinal endoscopic image detection model based on the gastrointestinal endoscopic image samples after the tumor nature level is marked; obtaining the endoscopic image to be detected; and obtaining the detection result of the endoscopic image to be detected according to the gastrointestinal endoscopic image detection model, wherein the detection result includes one or more information of the first classification result, the second classification result, and the multi-scale tumor area detection result.

[0004] The above technical solution can reduce the labeling cost and realize accurate positioning of the tumor area in the gastrointestinal endoscopic image, but it cannot realize multi-scale detection of gastrointestinal tumors. Summary of the invention

[0005] The inventors discovered through research that the detection of tumors in the digestive tract needs to be performed through a gastroenteroscopy, but the location of the tumor can only be determined through imaging. The tumor cannot be visually evaluated, and the grade of the tumor cannot be determined, which cannot pave the way for the next step of medical development.

[0006] The purpose of the present application is to provide a multi-scale detection method and device for tumors in gastrointestinal endoscopic images, which solves the technical problem that the prior art cannot provide a method for multi-scale detection of tumors and accurately determine the tumor grade by combining endoscopic tumor initial, middle and terminal images.

[0007] According to one aspect of the present application, a multi-scale detection method for tumors in digestive tract endoscopic images is provided, the method being executed by a controller and comprising:

[0008] Obtaining images of the initial, middle and terminal segments of the endoscopic tumor in the target area of ​​the digestive tract, preprocessing the images of the initial, middle and terminal segments of the endoscopic tumor to obtain a first image atlas, a second image atlas and a third image atlas;

[0009] The first image atlas, the second image atlas and the third image atlas are stripped and analyzed to obtain a first contour image atlas, a second contour image atlas and a third contour image atlas;

[0010] Performing grayscale processing on the first contour image atlas, the second contour image atlas, and the third contour image atlas to obtain a first grayscale atlas, a second grayscale atlas, and a third grayscale atlas;

[0011] The first grayscale atlas, the second grayscale atlas, and the third grayscale atlas are input into the plane coordinate system, and the edge clipping is performed on each grayscale atlas to obtain the coordinates (x i ,y i ), where i represents the i-th grayscale atlas;

[0012] The coordinates of each atlas (x i ,y i ) Input the three-dimensional modeling software to perform reconstruction, obtain the reconstructed microscopic tumor image in the target area, and grade the image according to the preset tumor grade.

[0013] In some embodiments, the acquisition of initial, middle and terminal image images of the microscopic tumor in the target area of ​​the digestive tract is specifically as follows: the initial image image of the microscopic tumor in the target area is the image presented when the endoscope approaches the digestive tract tumor, the middle image image of the microscopic tumor in the target area is the image presented when the endoscope is located at the digestive tract tumor, and the terminal image image of the microscopic tumor in the target area is the image presented when the endoscope is far away from the digestive tract tumor, wherein the presentation positions of the initial image image and the terminal image image are respectively located on opposite sides of the digestive tract tumor position.

[0014] In some embodiments, the preprocessing of the initial, middle and final endoscopic tumor images is specifically: performing at least normalization, grayscale and pixelation processing on the initial, middle and final endoscopic tumor images.

[0015] In some embodiments, the stripping analysis of the first image atlas, the second image atlas, and the third image atlas is specifically:

[0016] Obtain each pixel point in the first image atlas, the second image atlas, and the third image atlas, and record the pixel point as an anchor point;

[0017] Get the anchor pixel values ​​between adjacent anchor points. When the difference between the anchor pixel values ​​is less than the standard deviation value, the anchor points corresponding to the corresponding related differences are merged and recorded as the first anchor point group. When the difference between the anchor pixel values ​​is greater than the standard deviation value, the anchor points corresponding to the corresponding related differences are merged and recorded as the second anchor point group. When the difference between the anchor pixel values ​​is equal to the standard deviation value, the anchor points corresponding to the corresponding related differences are merged and recorded as the third anchor point group.

[0018] The image atlas corresponding to the first anchor point group is regarded as the first contour image atlas, the image atlas corresponding to the second anchor point group is regarded as the second contour image atlas, and the image atlas corresponding to the third anchor point group is regarded as the third contour image atlas.

[0019] In some embodiments, the edge clipping is performed on each grayscale atlas to obtain the coordinates (x i ,y i ), specifically:

[0020] Select a standard rectangular view frame, perform edge standardization cropping on the first grayscale atlas, the second grayscale atlas, and the third grayscale atlas, to obtain a first rectangular grayscale atlas, a second rectangular grayscale atlas, and a third rectangular grayscale atlas;

[0021] Put the first rectangular grayscale atlas, the second rectangular grayscale atlas and the third rectangular grayscale atlas into the plane coordinate system and obtain the corresponding coordinates (x i ,y i ).

[0022] In some embodiments, the coordinates (x i ,y i ) Input the three-dimensional modeling software to perform reconstruction, obtain the reconstructed microscopic tumor image in the target area, and grade the image according to the preset tumor grade, specifically:

[0023] The coordinates (x i ,y i ) are respectively input into the 3D modeling software, and restored by spline sweeping and surface modeling to obtain the microscopic tumor pattern in the target area;

[0024] By comparing the microscopic tumor pattern in the target area with the preset tumor grade, the tumor grade of the target area is determined and marking is performed.

[0025] According to another aspect of the present application, a multi-scale detection device for tumors in digestive tract endoscopic images is provided, including a device body, wherein the device body includes functional components for performing detection, and the functional components include:

[0026] An acquisition module, the acquisition module is used to acquire the images of the initial, middle and terminal segments of the endoscopic tumor in the target area of ​​the digestive tract, pre-process the images of the initial, middle and terminal segments of the endoscopic tumor, and obtain a first image atlas, a second image atlas and a third image atlas;

[0027] A first preprocessing module, the first preprocessing module is used for stripping and analyzing the first image atlas, the second image atlas and the third image atlas to obtain a first contour image atlas, a second contour image atlas and a third contour image atlas;

[0028] A second preprocessing module, the second preprocessing module is used to perform grayscale processing on the first contour image atlas, the second contour image atlas and the third contour image atlas to obtain a first grayscale atlas, a second grayscale atlas and a third grayscale atlas;

[0029] The edge processing module is used to input the first grayscale atlas, the second grayscale atlas and the third grayscale atlas into a plane coordinate system, perform edge clipping on each grayscale atlas, and obtain the coordinates (x i ,y i ), where i represents the i-th grayscale atlas;

[0030] The reconstruction module is used to transform each atlas coordinate (x i ,y i ) Input the three-dimensional modeling software to perform reconstruction, obtain the reconstructed microscopic tumor image in the target area, and grade the image according to the preset tumor grade.

[0031] In some embodiments, the acquisition module, the first preprocessing module, the second preprocessing module, the edge processing module, and the reconstruction module are all connected to the device body data.

[0032] Compared with the prior art, the present application has the following beneficial effects: the present application obtains the images of the initial, middle and terminal segments of the microscopic tumor in the target area of ​​the digestive tract, combines the coordinates of each atlas, and restores the digestive tract tumor state through three-dimensional software and three-dimensional restoration, obtains a visualized three-dimensional tumor state, quickly and accurately detects and determines the tumor grade, and realizes multi-scale detection of tumors in digestive tract endoscopic images. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0034] Figure 1 is a flow chart of the multi-scale detection method of the present application;

[0035] Figure 2 It is a schematic diagram of the structure of the multi-scale detection device of the present application. DETAILED DESCRIPTION

[0036] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0037] This embodiment provides a multi-scale detection method and device for tumors in digestive tract endoscopic images. The multi-scale detection method and device are already in the testing stage. Figure 1-2 Describe them together.

[0038] Application Overview

[0039] Computer technology has been widely used in the medical field. Digestive endoscopy images often contain a large amount of information. The detection of tumors in the digestive tract requires gastroenteroscopy, but the location of the tumor can only be determined by imaging, and the tumor cannot be visually evaluated, and the grade of the tumor cannot be quickly determined, which cannot pave the way for the next step of medical treatment. As a result, only biopsy resection can be performed. For benign tumors, this method is highly feasible. However, for malignant tumors, whether they can be directly removed without causing secondary damage depends on the doctor's accumulated experience. Therefore, for inexperienced doctors, direct resection is often the only option, which increases the risk of secondary damage.

[0040] Based on the above considerations, the present application pre-processes the endoscopic tumor images of the initial, middle and terminal segments of the target area of ​​the digestive tract through the images of the initial, middle and terminal segments of the endoscopic tumor, and restores them through three-dimensional modeling using three-dimensional software to obtain a visualized tumor pattern, thereby achieving rapid determination of the tumor grade. After the three-dimensional restoration, more diverse evaluation and analysis of the tumor can also be performed.

[0041] Example

[0042] Exemplary Methods

[0043] In order to solve the aforementioned problems and achieve the corresponding effects, the present embodiment provides a multi-scale detection method for tumors in digestive tract endoscopic images, which is executed by a controller, including obtaining images of the initial, middle and terminal segments of endoscopic tumors in a target area of ​​the digestive tract, preprocessing the initial, middle and terminal segment endoscopic tumor images to obtain a first image atlas, a second image atlas and a third image atlas; performing stripping analysis on the first image atlas, the second image atlas and the third image atlas to obtain a first contour image atlas, a second contour image atlas and a third contour image atlas; performing grayscale processing on the first contour image atlas, the second contour image atlas and the third contour image atlas to obtain a first grayscale atlas, a second grayscale atlas and a third grayscale atlas; inputting the first grayscale atlas, the second grayscale atlas and the third grayscale atlas into a plane coordinate system, performing edge clipping on each grayscale atlas, and obtaining the coordinates (x i ,y i ), where i represents the i-th grayscale atlas; the coordinates of each atlas (x i ,y i ) Input the three-dimensional modeling software to perform reconstruction, obtain the reconstructed microscopic tumor image in the target area, and grade the image according to the preset tumor grade.

[0044] The multi-scale detection method of tumors in gastrointestinal endoscopic images combines endoscopic tumor images of the initial, middle and terminal segments to solve the problem that the existing technology only uses a single gastrointestinal endoscope to detect gastrointestinal tumors, which cannot effectively obtain the three-dimensional state of the entire tumor and thus cannot accurately locate the tumor grade.

[0045] Below, each step will be described in detail.

[0046] In some embodiments, the initial image of the microscopic tumor in the target area is the image presented when the endoscope approaches the digestive tract tumor, the middle image of the microscopic tumor in the target area is the image presented when the endoscope is located at the digestive tract tumor, and the final image of the microscopic tumor in the target area is the image presented when the endoscope is far away from the digestive tract tumor, wherein the presentation positions of the initial image and the final image are located on opposite sides of the digestive tract tumor. It is understandable that in this implementation, by acquiring images of three positions through the built-in camera of the gastroscope, the three-angle comprehensive restoration of the internal tumor of the digestive tract can be achieved, and multi-angle detection can be achieved.

[0047] In some embodiments, at least normalization, grayscale and pixelation are performed on the initial, middle and final endoscopic tumor images. It is understandable that the preprocessing in this implementation is a common method, and the purpose is also to clean the image. In other words, in other equivalent embodiments, other processing methods other than the above can also be used, such as image flipping, image size adjustment, etc.

[0048] In some embodiments, the first image atlas, the second image atlas, and the third image atlas are stripped and analyzed. The stripping method in this embodiment is to place the pixelated and grayscaled first, second, and third image atlases in a plane rectangular coordinate system, and then use the stripping analysis method on the first, second, and third image atlases to obtain multiple graphic contours.

[0049] In detail, each pixel point in the first image atlas, the second image atlas and the third image atlas is obtained, and the pixel point is recorded as an anchor point; the anchor pixel value between adjacent anchor points is obtained, when the difference between the anchor pixel values ​​is less than the standard deviation value, the anchor points corresponding to the corresponding related difference values ​​are merged and recorded as the first anchor point group, when the difference between the anchor pixel values ​​is greater than the standard deviation value, the anchor points corresponding to the corresponding related difference values ​​are merged and recorded as the second anchor point group; when the difference between the anchor pixel values ​​is equal to the standard deviation value, the anchor points corresponding to the corresponding related difference values ​​are merged and recorded as the third anchor point group; the image atlas corresponding to the first anchor point group is regarded as the first contour image atlas, the image atlas corresponding to the second anchor point group is regarded as the second contour image atlas, and the image atlas corresponding to the third anchor point group is regarded as the third contour image atlas.

[0050] In some embodiments, a standard rectangular view frame is selected, and the first grayscale atlas, the second grayscale atlas, and the third grayscale atlas are edge-standardized and cropped to obtain a first rectangular grayscale atlas, a second rectangular grayscale atlas, and a third rectangular grayscale atlas; the first rectangular grayscale atlas, the second rectangular grayscale atlas, and the third rectangular grayscale atlas are placed in a plane coordinate system to obtain corresponding coordinates (x i ,y i ). It can be understood that in this embodiment, selecting the standard rectangular view frame can be understood as marking a window. After the image passes through the window, the window clipping will be performed, and the image area after clipping is determined by the window shape. After obtaining the clipped image, put it into the coordinate system to determine the horizontal and vertical coordinate values, that is, obtain the coordinates (x i ,y i )’s specific value.

[0051] Further, the coordinates (x i ,y i ) are respectively input into the 3D modeling software, restored by spline sweeping and surface modeling, and the microscopic tumor pattern in the target area is obtained. The tumor grade of the target area is determined by comparing the microscopic tumor pattern in the target area with the preset tumor grade, and marking is performed. It can be understood that after the coordinate system is placed in the 3D modeling software, the positions of the three images can be determined according to the coordinate values. After combining them into a closed dimension chain, the three-dimensional approximate pattern can be restored by related operations such as stretching, spline sweeping, and surface modeling, and then subsequent tumor grade determination and possible pathological predictive analysis can be performed.

[0052] In order to better understand the method of the present application, this embodiment further provides a device, including a device body, the device body includes a functional component for performing detection, and the functional component includes:

[0053] An acquisition module, which is used to acquire images of the initial, middle and terminal segments of endoscopic tumors in the target area of ​​the digestive tract, and preprocess the images of the initial, middle and terminal segments of endoscopic tumors to obtain a first image atlas, a second image atlas and a third image atlas;

[0054] A first preprocessing module, the first preprocessing module is used for stripping and analyzing the first image atlas, the second image atlas and the third image atlas to obtain a first contour image atlas, a second contour image atlas and a third contour image atlas;

[0055] A second preprocessing module, the second preprocessing module is used to perform grayscale processing on the first contour image atlas, the second contour image atlas and the third contour image atlas to obtain a first grayscale atlas, a second grayscale atlas and a third grayscale atlas;

[0056] The edge processing module is used to input the first grayscale atlas, the second grayscale atlas and the third grayscale atlas into the plane coordinate system, perform edge clipping on each grayscale atlas, and obtain the coordinates (x i ,y i ), where i represents the i-th grayscale atlas;

[0057] The reconstruction module is used to transform the coordinates (x i ,y i ) Input the three-dimensional modeling software to perform reconstruction, obtain the reconstructed microscopic tumor image in the target area, and grade the image according to the preset tumor grade.

[0058] It should be noted that, in some embodiments, the acquisition module, the first preprocessing module, the second preprocessing module, the edge processing module and the reconstruction module are all connected to the device body data to form a functional closed loop to perform multi-scale detection.

[0059] The schematic diagram of the device provided in the embodiment of the present application includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a multi-scale detection program. When the processor executes the computer program, each step in the above multi-scale detection method embodiment is implemented. Alternatively, when the processor executes the computer program, the functions of each module in the above device embodiments are implemented.

[0060] Exemplarily, the computer program can be divided into one or more modules, one or more modules are stored in the memory and executed by the processor to complete the present application. One or more modules can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the multi-scale detection device. For example, the computer program can be divided into an acquisition module, a first preprocessing module, a second preprocessing module, an edge processing module, and a reconstruction module, and the specific functions of each module are as described above.

[0061] In addition, the aforementioned processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the multi-scale detection device, and uses various interfaces and lines to connect various parts of the entire multi-scale detection device.

[0062] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A multi-scale detection method for tumors in digestive tract endoscopic images, the method being executed by a controller, characterized in that: include: Obtaining images of the initial, middle and terminal segments of the endoscopic tumor in the target area of ​​the digestive tract, preprocessing the images of the initial, middle and terminal segments of the endoscopic tumor to obtain a first image atlas, a second image atlas and a third image atlas; The first image atlas, the second image atlas and the third image atlas are stripped and analyzed to obtain a first contour image atlas, a second contour image atlas and a third contour image atlas; Performing grayscale processing on the first contour image atlas, the second contour image atlas, and the third contour image atlas to obtain a first grayscale atlas, a second grayscale atlas, and a third grayscale atlas; Input the first grayscale atlas, the second grayscale atlas, and the third grayscale atlas into the plane coordinate system, perform edge clipping on each grayscale atlas, and obtain the coordinates (x i ,y i ), where i represents the i-th grayscale atlas; The coordinates of each atlas (x i ,y i ) Input the three-dimensional modeling software to perform reconstruction, obtain the reconstructed microscopic tumor image in the target area, and grade the image according to the preset tumor grade.

2. The method according to claim 1, characterized in that The obtaining of initial, middle and terminal image maps of the microscopic tumor in the target area of ​​the digestive tract is specifically as follows: the initial image map of the microscopic tumor in the target area is the image presented when the endoscope approaches the digestive tract tumor, the middle image map of the microscopic tumor in the target area is the image presented when the endoscope is located at the digestive tract tumor, and the terminal image map of the microscopic tumor in the target area is the image presented when the endoscope is far away from the digestive tract tumor, wherein the presentation positions of the initial image map and the terminal image map are respectively located on opposite sides of the digestive tract tumor position.

3. The method according to claim 1, characterized in that The preprocessing of the initial, middle and final endoscopic tumor images specifically includes: performing at least normalization, grayscale and pixelation processing on the initial, middle and final endoscopic tumor images.

4. The method according to claim 1, characterized in that: The stripping analysis of the first image atlas, the second image atlas and the third image atlas is specifically as follows: Obtain each pixel point in the first image atlas, the second image atlas, and the third image atlas, and record the pixel point as an anchor point; Get the anchor pixel values ​​between adjacent anchor points. When the difference between the anchor pixel values ​​is less than the standard deviation value, the anchor points corresponding to the corresponding related difference values ​​are merged and recorded as the first anchor point group. When the difference between the anchor pixel values ​​is greater than the standard deviation value, the anchor points corresponding to the corresponding related difference values ​​are merged and recorded as the second anchor point group. When the difference between the anchor pixel values ​​is equal to the standard deviation value, the anchor points corresponding to the corresponding related difference values ​​are merged and recorded as the third anchor point group; The image atlas corresponding to the first anchor point group is regarded as the first contour image atlas, the image atlas corresponding to the second anchor point group is regarded as the second contour image atlas, and the image atlas corresponding to the third anchor point group is regarded as the third contour image atlas.

5. The method according to claim 1, characterized in that: The above method performs edge clipping on each grayscale atlas to obtain the coordinates of each atlas after clipping (x i ,y i ), specifically: Select a standard rectangular view frame, perform edge standardization cropping on the first grayscale atlas, the second grayscale atlas, and the third grayscale atlas, to obtain a first rectangular grayscale atlas, a second rectangular grayscale atlas, and a third rectangular grayscale atlas; Put the first rectangular grayscale atlas, the second rectangular grayscale atlas and the third rectangular grayscale atlas into the plane coordinate system and obtain the corresponding coordinates (x i ,y i ).

6. The method according to claim 1, characterized in that The coordinates of each atlas (x i ,y i ) Input the three-dimensional modeling software to perform reconstruction, obtain the reconstructed microscopic tumor image in the target area, and grade the image according to the preset tumor grade, specifically: The coordinates (x i ,y i ) are respectively input into the 3D modeling software, and restored by spline sweeping and surface modeling to obtain the microscopic tumor pattern in the target area; By comparing the microscopic tumor pattern in the target area with the preset tumor grade, the tumor grade of the target area is determined and marking is performed.

7. A multi-scale detection device for tumors in digestive tract endoscopic images, comprising a device body, wherein the device body includes functional components for performing detection, characterized in that: The functional components include: An acquisition module, the acquisition module is used to acquire the images of the initial, middle and terminal segments of the endoscopic tumor in the target area of ​​the digestive tract, pre-process the images of the initial, middle and terminal segments of the endoscopic tumor, and obtain a first image atlas, a second image atlas and a third image atlas; A first preprocessing module, the first preprocessing module is used for stripping and analyzing the first image atlas, the second image atlas and the third image atlas to obtain a first contour image atlas, a second contour image atlas and a third contour image atlas; A second preprocessing module, the second preprocessing module is used to perform grayscale processing on the first contour image atlas, the second contour image atlas and the third contour image atlas to obtain a first grayscale atlas, a second grayscale atlas and a third grayscale atlas; The edge processing module is used to input the first grayscale atlas, the second grayscale atlas and the third grayscale atlas into a plane coordinate system, perform edge clipping on each grayscale atlas, and obtain the coordinates (x i ,y i ), where i represents the i-th grayscale atlas; The reconstruction module is used to transform each atlas coordinate (x i ,y i ) Input the three-dimensional modeling software to perform reconstruction, obtain the reconstructed microscopic tumor image in the target area, and grade the image according to the preset tumor grade.

8. The device according to claim 7, characterized in that The acquisition module, the first preprocessing module, the second preprocessing module, the edge processing module and the reconstruction module are all data-connected to the device body.

Images