A Method and System for Calculating Traumatic Brain Injury Volume in Stroke Based on OCT Images

Abstract

This invention discloses a method and system for calculating the volume of traumatic brain injury in stroke patients based on OCT images. The method includes: removing the glass slide and the portion above it from the OCT image to obtain the target image; preprocessing the target image by cropping the effective region, threshold filtering, and image closing operation; extracting the brain boundary from a single OCT image; extracting the brain boundary from a set of OCT images; calculating the volume of the traumatic brain injury; and judging the stroke rehabilitation effect based on changes in the volume of the traumatic brain injury. This invention provides a novel OCT image processing method that allows for real-time in vivo observation of the stroke rehabilitation process and provides a quantitative evaluation standard to assist in judging the stroke rehabilitation effect, offering technical support and convenience for finding effective stroke rehabilitation methods.

Description

Technical Field

[0001] This invention belongs to the field of medical image processing and relates to a method and system for calculating the volume of craniocerebral injury in stroke based on OCT images. Background Technology

[0002] Stroke, also known as cerebral apoplexy, is a group of diseases caused by brain tissue damage due to ruptured blood vessels in the brain (hemorrhagic stroke) or blockage of blood vessels in the brain (ischemic stroke). The main clinical symptoms are sudden death, fainting, facial paralysis, speech impairment, and hemiplegia. Due to its high incidence, disability and mortality rates, it has become one of the major diseases affecting human life and health.

[0003] With the development of medical technology, the mortality rate of stroke has been decreasing year by year, but it often leaves varying degrees of functional impairment, especially hemiplegia. This symptom severely restricts patients' limb movement, affects their daily living abilities, and reduces their quality of life. Exploring new diagnostic and treatment methods and effective interventions is currently a hot topic and a challenge in the field of neuroscience. In order to explore new diagnostic and treatment methods, it is necessary to evaluate the rehabilitation effects of these methods.

[0004] Previous stroke research has relied on in vitro imaging or 3D imaging to indirectly reflect changes in brain volume. However, with the advent of optical coherence tomography (OCT), in vivo real-time experiments can now be conducted. Nevertheless, current research on OCT image processing is still limited, and there is a need to develop and improve OCT image processing techniques. Summary of the Invention

[0005] This invention addresses the shortcomings of existing technologies by providing a method, system, device, and storage medium for calculating the volume of craniocerebral injury in stroke based on OCT images.

[0006] In a first aspect, the present invention provides a method for calculating the volume of traumatic brain injury in stroke based on OCT images, the method comprising the following steps:

[0007] Step 1: Obtain the target image;

[0008] Step 2: Preprocess the target image;

[0009] Step 3: Extract the brain boundary from a single OCT image;

[0010] Step 4: Extract the brain boundary from the entire set of OCT images;

[0011] Step 5: Calculate the volume of the brain injury.

[0012] Secondly, the present invention provides a system for calculating the volume of traumatic brain injury in stroke based on OCT images, comprising:

[0013] The image acquisition module is used to acquire the target image;

[0014] The preprocessing module is used to preprocess the target image;

[0015] The first brain boundary extraction module is used to extract the brain boundary of a single OCT image;

[0016] The second brain boundary extraction module is used to extract the brain boundary of the entire set of OCT images;

[0017] The volume calculation module is used to calculate the volume of craniocerebral injury.

[0018] Thirdly, the present invention provides a computer device, including a memory, a processor, and computer-executable instructions stored in the memory and executable on the processor, wherein the processor executes the instructions to implement the method for calculating the volume of craniocerebral injury in stroke based on OCT images.

[0019] Fourthly, the present invention provides a computer storage medium storing computer-executable instructions, which, when executed, implement the method for calculating the volume of craniocerebral injury in stroke based on OCT images.

[0020] This invention can obtain real-time in vivo intracranial injury volume for quantitative stroke rehabilitation assessment, increasing the timeliness and accuracy of rehabilitation assessment, providing a basis for judging the effectiveness of subsequent stroke rehabilitation methods, and reducing the time and energy costs for researchers. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating the specific implementation steps of the method for calculating the volume of traumatic brain injury in stroke based on OCT images and the rehabilitation assessment in this application.

[0022] Figure 2 This is an example of the method for calculating the volume of traumatic brain injury in stroke based on OCT images and the effect diagrams at each stage of the rehabilitation assessment process according to the embodiments of this application.

[0023] Figure 3 This is a module architecture diagram of the post-stroke rehabilitation assessment system based on OCT images, according to an embodiment of this application. Detailed Implementation

[0024] The specific embodiments of the present invention are described below with reference to the accompanying drawings.

[0025] like Figure 1 As shown, this application provides a method for calculating the volume of traumatic brain injury in stroke based on OCT images, using rat OCT images as an example:

[0026] Step 1: Using manually labeled points and line fitting, obtain the expression for the lower glass plate in the image, thereby removing the lower glass plate and the pixels above it from the OCT image to facilitate subsequent image processing; specifically:

[0027] 1-1 The regression coefficients of the lower glass slide curve were obtained by linear fitting of manually marked points. The fitting expression is as follows:

[0028]

[0029] Where k0, k1, k2, k3, k4, b are the regression coefficients of the fitted straight line, and x, y, z are the three-dimensional coordinates of the manually labeled points.

[0030] 1-2 The expression for the lower glass slide curve is obtained based on the regression coefficients, as follows:

[0031] z=k0*x+k1*y+k2*x*y+k3*x*x+k4*y*y+b (2)

[0032] The above formula is used to remove the pixels above and below the upper and lower glass plates of the original OCT image, and the function F of the lower glass plate in the three-dimensional coordinate system is obtained. glass (x,y,z), such as Figure 2 The lower glass slide is shown in the image of craniocerebral injury in a rat.

[0033] Step 2: Preprocess the OCT image; specifically:

[0034] 2-1. Extract the effective region to reduce the impact of noise and impurities on the OCT image. The expression is as follows:

[0035] When z1 ≤ z ≤ z2

[0036] Where z1 to z2 is the effective image range in a set of OCT images, z represents the vertical axis of the three-dimensional coordinate system, that is, the axis in the front-back direction, and x1 and x2 are the left and right boundaries of the OCT image confirmed after observation of one of the OCT images.

[0037] 2-2 Set a threshold to reduce the overall contrast of the OCT image and highlight edge contrast. The expression is:

[0038]

[0039] Where f(x,y) is a function of the OCT image, TH is the calculated threshold, and N is the number of non-zero pixels in the OCT image.

[0040] 2-3. Perform a closing operation on the OCT image to reduce the gap and decrease the likelihood of intermediate blood vessels being identified as boundaries. The expression is:

[0041]

[0042] Where F is the OCT image matrix and B is the filter kernel, the size of which is selected according to the filtering effect. It is an image dilation operation. It is an image erosion operation.

[0043] Step 3: Extract the brain boundary from a single rat OCT image to obtain the function for extracting the rat cerebral cortex, denoted as L, specifically:

[0044] 3-1 Combine the Sobel and Canny operators to filter OCT images and normalize each operator separately. The expression is:

[0045]

[0046] Where ε is a number between 0 and 1. c (x, y) is the result of filtering the OCT image using the Canny operator, f s (x,y) is the result of filtering the OCT image using the Sobel operator.

[0047] 3-2 Global normalization. The expression is:

[0048]

[0049] 3-3 The minimum path, i.e., the boundary of the rat cerebral cortex, is found using dynamic programming. The expression is:

[0050] f cost (i,0)=f NF (i,0),0￡i￡m⑻

[0051]

[0052] Among them, f cost (i, j) represents the minimum path sum from the leftmost point of the OCT image to point (i, j), m is the number of rows in the OCT image, and q represents the number of rows that are allowed to differ between adjacent points on the shortest path. The sum is calculated iteratively using the above formula.

[0053] The specific process of the dynamic programming method is to start from the leftmost column of points in the OCT image and search for the next pixel point to the right. During this process, it is necessary to ensure that the formula (9) is satisfied. Finally, the rat cerebral cortex is deduced based on the minimum value of the rightmost column of the image.

[0054] Combining formulas (6), (7), (8), and (9) yields the function L for extracting the rat cerebral cortex.

[0055] Step 4: Extract the boundaries of the rat cerebral cortex from the entire set of OCT images. Specifically:

[0056] 4-1 Extract OCT images at certain intervals. The expression is:

[0057]

[0058] Among them, f 3D (x, y, z) is the three-dimensional matrix of the entire OCT, where z is the vertical axis of the three-dimensional coordinate system, just like in step 2. 3D_J (x,y,z) represents the three-dimensional map of the rat cerebral cortex found, and g represents the sample interval.

[0059] 4-2 Determine the middle as the starting OCT image, and search for the OCT images to be extracted on both sides.

[0060] The expression is:

[0061] f3 D _ J(x, y,z0± g ) = L ( f 3D ( x, y,z0 ±g ) ), y 3 max (0, min (L(f 3D ( x,y,z0)) - l * g) (11)

[0062] Where L represents the function for extracting the rat cerebral cortex, represents the cerebral cortex of the plane indicated in parentheses, and the range of y is determined by the previously extracted OCT image. λ is a coefficient used to adjust the smoothness and accuracy of the cerebral cortex.

[0063] 4-3 Based on the existing boundaries of the rat cerebral cortex, the remaining rat cerebral cortex is obtained. The expression is:

[0064]

[0065] Where I represents the floor function and S represents the modulo function.

[0066] Step 4 yields the function F of the entire rat cortex in a three-dimensional coordinate system. brain (x,y,z), such as Figure 2 The cerebral cortex is shown in the images of craniocerebral injury in rats, where F brain y in f corresponds to 3D_J z in (x,y,z), F brain z in f corresponds to f 3D_J y in (x,y,z).

[0067] Step 5: Calculate the volume of the rat's intracranial injury by integrating the equations.

[0068]

[0069] Where V harm It is the volume of the head injury, z glass It is F glass z in z brain It is F brain The z in the middle.

[0070] Step 6: Based on the volume of the brain injury over a period of time, draw a graph showing the change in the volume of the brain injury over time to assess the stroke rehabilitation effect.

[0071] like Figure 3 As shown, this application embodiment also provides a computer device, including a memory, a processor, and computer-executable instructions stored in the memory and executable on the processor. When the processor executes the instructions, it implements the above-described method for calculating the volume of craniocerebral injury in stroke based on OCT images.

[0072] This application embodiment also provides a computer storage medium storing computer-executable instructions, which, when executed, implement the above-described method for calculating the volume of brain injury in stroke based on OCT images.

[0073] The embodiments of this application can help doctors process OCT images and assist in evaluating the effectiveness of stroke rehabilitation, thereby determining the merits of stroke rehabilitation methods.

[0074] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for calculating the volume of traumatic brain injury in stroke based on OCT images, characterized in that... The method includes the following steps: Step 1: Obtain the target image; Step 2: Preprocess the target image; Step 3: Extract the brain boundary from a single OCT image; Step 4: Extract the brain boundary from the entire set of OCT images; Step 5: Calculate the volume of the traumatic brain injury; Step 2 specifically involves: 2-1. Extract the effective region to reduce noise and impurities. The expression is as follows: (1) in, arrive It is the effective image range in a set of OCT images. , The left and right boundaries of one of the OCT images confirmed after observation; 2-2. Calculate the threshold to reduce the overall contrast of the OCT image and highlight the edge contrast. The expression is: ， (2) Where f(x,y) is a function of the OCT image, TH is the calculated threshold, and N is the number of non-zero pixels in the OCT image; 2-3. Perform a closing operation on the image to reduce the gaps in the central blood vessels. The expression is: (3) Where F is the OCT image matrix, B is the filter kernel, and its size is selected according to the filtering effect; '⊕' is the image dilation operation, ' 'This is an image erosion operation; Step 3 specifically involves: 3-1. Combine the Sobel and Canny operators to filter the OCT image and normalize them individually. The expression is: (4) Where ε is a number between 0 and 1, f c (x, y) is the result of filtering the OCT image using the Canny operator, f s (x,y) is the result of filtering the OCT image using the Sobel operator; 3-2. Overall normalization, the expression is: （5） Among them, f NF (x,y) is the OCT image obtained in the previous step after filtering by the Sobel and Canny operators; 3-3. Find the boundary of the cerebral cortex using dynamic programming, expressed as: （6） （7） Among them, f cost (i, j) represents the minimum path sum from the leftmost point of the OCT image to point (i, j), m represents the number of rows in the OCT image, and q represents the number of rows that are allowed to differ between adjacent points of the shortest path. Step 4 specifically involves: 4-1. Extract the target image at intervals, the expression is: （8） Among them, f 3D (x,y,z) represents the three-dimensional matrix of the entire OCT, f 3D_J (x,y,z) represents the three-dimensional map of the cerebral cortex found, and g represents the sample interval. 4-2. Determine the starting OCT image, and then sequentially search for the remaining images; the expression is: （9） Where L is the function for extracting the cerebral cortex, and λ is a coefficient used to adjust the smoothness and accuracy of the cerebral cortex; 4-3. Based on the identified boundaries of the cerebral cortex, deduce the remaining cerebral cortex; the expression is: （10） Where I is the floor function and S is the modulo function; Step 5 specifically involves: The integral formula for calculating the volume of a traumatic brain injury is as follows: （11） Where V harm It is the volume of the head injury, z glass This is the z-coordinate value of the lower glass slide in a three-dimensional coordinate system. brain It is the z-coordinate value of the entire cerebral cortex in a three-dimensional coordinate system.

2. The method for calculating the volume of traumatic brain injury in stroke based on OCT images according to claim 1, characterized in that, It also includes step 6, judging the effect of stroke rehabilitation based on changes in the volume of the traumatic brain injury.

3. A system for calculating the volume of traumatic brain injury in stroke based on OCT images, used to implement the method described in claim 1, characterized in that, include: The image acquisition module is used to acquire the target image; The preprocessing module is used to preprocess the target image; The first brain boundary extraction module is used to extract the brain boundary of a single OCT image; The second brain boundary extraction module is used to extract the brain boundary of the entire set of OCT images; The volume calculation module is used to calculate the volume of craniocerebral injury.

4. The system for calculating the volume of traumatic brain injury in stroke based on OCT images according to claim 3, characterized in that, Also includes: The rehabilitation effect assessment module is used to determine the rehabilitation effect of stroke based on changes in the volume of traumatic brain injury.

5. A computer device, comprising a memory, a processor, and computer-executable instructions stored in the memory and executable on the processor, characterized in that, When the processor executes the instructions, it implements the method for calculating the volume of craniocerebral injury in stroke based on OCT images as described in claim 1 or 2.

6. A computer storage medium storing computer-executable instructions, characterized in that, When the computer-executable instructions are executed, they implement the method for calculating the volume of craniocerebral injury in stroke based on OCT images as described in claim 1 or 2.

Images