A dynamic monitoring system for rectal cancer radiotherapy effects combined with imaging analysis

Abstract

The present invention relates to the technical field of image motion analysis, and more particularly to a dynamic monitoring system for rectal cancer radiotherapy effects combined with imaging analysis. The system determines a reference recovery index representing the recovery effect from the dimensions of edge change and area change based on the characteristics that radiotherapy inhibits the spread of cancer cells, causing the edges to converge in the same direction toward a central location with a more severe infection, and that the lesion area becomes smaller as the treatment effect becomes better. The system then determines a more accurate shape recovery degree representing the reference recovery index based on the reference recovery index and the characteristics that the quasi-circularity of rectal tissue causes the lesion area to gradually shrink inward and the edges to become smoother. The system then more accurately corrects a modified radiotherapy recovery curve based on the shape recovery degree, making dynamic monitoring of the rectal cancer radiotherapy effect more intuitive and accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of image motion analysis, and in particular to a dynamic monitoring system for rectal cancer radiotherapy effects combined with imaging analysis. Background Art

[0002] Rectal cancer is a common malignant tumor of the digestive system, and radiotherapy is one of its main treatments. Experienced doctors usually make subjective assessments of radiotherapy effects based on their experience; however, for inexperienced medical students, learning how to evaluate radiotherapy effects is more difficult. Therefore, in order to enable medical students to more intuitively learn and understand the effects of radiotherapy for rectal cancer, a system that can dynamically understand the effects of radiotherapy for rectal cancer is needed to assist medical students in their learning and understanding.

[0003] Existing technologies, typically based on machine learning, utilize the characteristic that the grayscale of rectal cancer lesions in CT images changes over the course of radiotherapy to intuitively assess the effectiveness of radiotherapy for rectal cancer based on the grayscale changes in the lesions. However, the degree of change in CT images for patients at different recovery stages is typically small, making the corresponding grayscale changes difficult to capture. Furthermore, the impact of shape and area changes during the recovery process on the effectiveness of radiotherapy for rectal cancer is not considered, resulting in less intuitive and less effective dynamic monitoring of the effects of radiotherapy for rectal cancer. Summary of the Invention

[0004] In order to solve the technical problems that the method of intuitively evaluating the effect of rectal cancer radiotherapy based on the grayscale change value of the lesion area is difficult to capture the corresponding grayscale changes, and does not take into account the impact of the shape and area changes during the recovery process of rectal cancer radiotherapy on the radiotherapy effect, resulting in the dynamic monitoring of the rectal cancer radiotherapy effect being not intuitive and the monitoring effect being poor, the purpose of this application is to provide a dynamic monitoring system for rectal cancer radiotherapy effect combined with imaging analysis. The technical solutions adopted are as follows:

[0005] This application proposes a dynamic monitoring system for rectal cancer radiotherapy effects combined with imaging analysis, the system comprising:

[0006] A data acquisition and preprocessing module is configured to acquire an initial CT image of rectal cancer in an initial state and a restored CT image of rectal cancer at each sampling moment during the course of rectal cancer radiotherapy; and determine an initial lesion region in the initial CT image of rectal cancer and a restored lesion region in the restored CT image of rectal cancer based on an image segmentation method;

[0007] A first determination module is configured to determine a matching recovery region of the initial lesion region based on an intersection between the initial lesion region and the lesion recovery regions corresponding to each other sampling moment; and determine an initial effect change curve of the initial lesion region based on an overall grayscale change of all matching recovery regions;

[0008] The second determination module is configured to determine the degree of diffusion recovery at each sampling moment based on the overall trend of edge changes between the edge of the initial lesion area and each corresponding matched recovery area; and determine a reference recovery index based on the area reduction of each matched recovery area before each sampling moment and the degree of diffusion recovery;

[0009] The radiotherapy effect monitoring module is used to determine the degree of shape recovery based on the edge smoothness of the matching recovery area, the reference recovery index and the circular convergence in shape; correct the initial effect change curve according to the shape recovery degree to determine the corrected radiotherapy recovery curve; and dynamically monitor the radiotherapy effect of rectal cancer based on the corrected radiotherapy recovery curve.

[0010] Furthermore, the process of determining the initial lesion area in the rectal cancer CT initial image and the restored lesion area in the rectal cancer CT restored image based on the image segmentation method includes:

[0011] The CA significance algorithm is used to calculate the average significance value of each segmented region in the rectal cancer CT initial image at all segmentation scales; the segmented region whose normalized value of the average significance value is greater than a preset significance threshold is used as the lesion initial region; based on the acquisition process of the lesion initial region, the lesion recovery region in the rectal cancer CT recovery image is determined.

[0012] Furthermore, the process of obtaining the matching recovery area includes:

[0013] After mapping the rectal cancer CT restored image at each sampling moment to the rectal cancer CT initial image, among all the lesion restored areas corresponding to each sampling moment, the lesion restored area with the largest overlapping area with the lesion initial area is used as the matching restored area of ​​the lesion initial area in each rectal cancer CT restored image.

[0014] Furthermore, the process of obtaining the initial effect change curve includes:

[0015] The average grayscale value of all pixels in the matching recovery area at the previous sampling moment corresponding to each matching recovery area is used as the contrast grayscale value of each matching recovery area; the average grayscale value of all pixels in each matching recovery area is used as the overall grayscale value; the grayscale change value of each matching recovery area is determined according to the difference between the contrast grayscale value and the overall grayscale value; the grayscale change values ​​of all matching recovery areas in the initial lesion area are arranged in chronological order and curve fitting is performed to determine the initial effect change curve of the initial lesion area.

[0016] Furthermore, the process of obtaining the diffusion recovery degree includes:

[0017] The area of ​​each corresponding matching recovery region is proportionally enlarged to the area size of the corresponding initial lesion region with the corresponding centroid as the center to obtain the corresponding reference recovery region;

[0018] After mapping the reference restored area to the image of the corresponding initial lesion area, connecting each boundary pixel point of the initial lesion area with the nearest boundary pixel point on the reference restored area to obtain all first reference connecting lines; and extending all the first reference connecting lines to obtain all the intersection points as corresponding reference convergence points;

[0019] The average of all distances between all reference convergence points corresponding to each matching recovery area is taken as the corresponding convergence distance;

[0020] The diffusion recovery degree of each matching recovery area is determined according to the convergence distance and the number of the reference convergence points.

[0021] Furthermore, the process of determining the diffusion recovery degree of each matching recovery area according to the convergence distance and the number of reference convergence points includes:

[0022] A negative correlation mapping is performed on the product of the number of reference convergence points and the convergence distance to determine the diffusion recovery degree of each matching recovery area.

[0023] Furthermore, the process of obtaining the reference recovery index includes:

[0024] The difference between the area of ​​each matching recovery area corresponding to the initial area of ​​the lesion and the area of ​​the corresponding matching recovery area at the previous sampling moment is negatively correlated to determine the area reduction index of each matching recovery area; the product of the area reduction index and the diffusion recovery degree is normalized to determine the reference recovery index of each matching recovery area.

[0025] Furthermore, the process of obtaining the shape recovery degree includes:

[0026] Detecting the number of corner points in each matching restoration area using a corner detection algorithm; multiplying the negative correlation mapping value of the number of corner points by the reference restoration index as the edge restoration index;

[0027] The distance between each boundary pixel point and the corresponding centroid in the matched restoration area is used as the centroid distance of each boundary pixel point; the centroid distance difference between each boundary pixel point and each other boundary pixel point is calculated; the mean of all centroid distance differences corresponding to all boundary pixels is negatively correlated and mapped to determine the shape restoration index of the matched restoration area;

[0028] The product of the shape recovery index and the edge recovery index is normalized to determine the shape recovery degree of the matching recovery area.

[0029] Furthermore, the process of obtaining the modified radiotherapy recovery curve includes:

[0030] Correcting the grayscale change value corresponding to the matching restoration area on the initial effect change curve according to the shape restoration degree, and determining the corrected change value of the matching restoration area;

[0031] The corrected change values ​​of all matching recovery areas of the lesion initial area are arranged in time sequence and then curve fitting is performed to determine a corrected radiotherapy recovery curve of the lesion initial area.

[0032] Furthermore, the process of obtaining the modified change value includes:

[0033] The product of the positive correlation mapping value of the shape restoration degree and the grayscale change value of the corresponding matching restoration area on the initial effect change curve is used as the corrected change value of the matching restoration area.

[0034] In a second aspect, the present application provides a computer device comprising a memory and a processor. The memory is used to store computer program code, and the processor is used to call and execute the computer program code from the memory to execute the system of the first aspect or any embodiment of the first aspect of the present application.

[0035] In a third aspect, the present application provides a computer program product, comprising computer program code, which, when executed, executes the system of the first aspect or any embodiment of the first aspect of the present application.

[0036] In a fourth aspect, the present application provides a computer-readable storage medium, which stores computer program code. When the computer program code is executed, it executes the system of the first aspect of the present application or any embodiment of the first aspect.

[0037] This application has the following beneficial effects:

[0038] This application first determines the lesion area at each sampling moment based on significance, then determines the matching recovery area for each lesion area at each sampling moment based on the intersection of the lesion areas. Based on the characteristic that the body's recovery speed increases with the continuous improvement of the disease, the grayscale change of the matching recovery area is used to determine the initial effect change curve in the initial state. Furthermore, based on the characteristics that radiotherapy inhibits the spread of cancer cells, causing the edges to converge in the same direction toward the center of the more severely infected area, and the characteristics that the lesion area decreases as the treatment effect improves, a reference recovery index representing the recovery effect is determined from the dimensions of edge change and area change. Then, based on the reference recovery index and the characteristic that the quasi-circularity of rectal tissue causes the lesion area to gradually shrink inward and the edges to become smoother, a more accurate shape recovery degree representing the reference recovery index is determined. Thus, the modified radiotherapy recovery curve is more accurately corrected based on the shape recovery degree, making dynamic monitoring of the radiotherapy effect of rectal cancer more intuitive and accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0040] Figure 1 This is a structural block diagram of a system for dynamically monitoring the radiotherapy effect of rectal cancer combined with imaging analysis provided by one embodiment of the present invention;

[0041] Figure 2 The present invention provides a schematic diagram of a computer device structure according to an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to further illustrate the technical means and efficacy adopted by the present invention to achieve the predetermined purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation method, structure, characteristics and efficacy of a dynamic monitoring system for rectal cancer radiotherapy effects combined with imaging analysis proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment, and the specific features, structures or characteristics in one or more embodiments may be combined in any suitable form. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0043] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0044] The specific scheme of the dynamic monitoring system for rectal cancer radiotherapy effect combined with imaging analysis provided by the present invention is described in detail below with reference to the accompanying drawings.

[0045] The present application provides a system for dynamic monitoring of the radiotherapy effect of rectal cancer combined with imaging analysis. Figure 1 , which shows a structural block diagram of a dynamic monitoring system for rectal cancer radiotherapy effects combined with imaging analysis provided by an embodiment of the present invention. The system includes: a data acquisition and preprocessing module 101, a first determination module 102, a second determination module 103 and a radiotherapy effect monitoring module 104.

[0046] The data acquisition and preprocessing module 101 is used to acquire the initial CT image of rectal cancer in the initial state and the restored CT image of rectal cancer at each sampling moment during the radiotherapy treatment of rectal cancer; and determine the initial lesion area in the initial CT image of rectal cancer and the restored lesion area in the restored CT image of rectal cancer based on significance.

[0047] First, the rectal cancer CT images obtained by the same patient during the course of rectal cancer radiotherapy and each time a CT scan is performed before the treatment are collected, wherein each sampling moment corresponds to the moment of each CT scan, that is, the sampling frequency is determined by the frequency of the CT scan. It should be noted that the rectal cancer CT images used for medical education in this application have been authorized for use by the patient and will not be further described later. In a specific implementation method of an embodiment of the present invention, the rectal cancer CT image obtained when the patient undergoes the last CT scan before the start of the rectal cancer radiotherapy treatment process is used as the rectal cancer CT initial image; the rectal cancer CT image obtained when the patient undergoes a CT scan during the course of rectal cancer radiotherapy is used as the rectal cancer CT recovery image; by collecting the rectal cancer CT initial image and the rectal cancer CT recovery image for comparative analysis, it is convenient to observe the treatment effect more intuitively.

[0048] The rectal cancer lesion area has a larger grayscale value than other normal areas, is whiter, and has better integrity. Therefore, the lesion area, i.e., the high-saliency area, can be divided by saliency detection. Preferably, in some possible implementations of the embodiments of the present invention, the process of determining the lesion initial area in the rectal cancer CT initial image and the lesion recovery area in the rectal cancer CT recovery image based on the image segmentation method includes:

[0049] The CA significance algorithm is used to calculate the average significance value of each segmented region in the rectal cancer CT initial image at all segmentation scales; the segmented region whose normalized value of the average significance value is greater than the preset significance threshold is used as the lesion initial region; based on the acquisition process of the lesion initial region, the lesion recovery region in the rectal cancer CT restored image is determined. Among them, the lesion recovery region is the lesion region in the rectal cancer CT restored image, and the lesion initial region is the lesion region in the rectal cancer CT initial image. In a specific implementation of an embodiment of the present invention, the preset significance threshold is set to 0.7, which can be adjusted according to the specific implementation environment, and the normalization method adopts linear normalization. It should be noted that the calculation of average significance belongs to the technical term in the CA significance algorithm. The CA significance algorithm is a technical means well known to those skilled in the art and will not be further defined or elaborated here.

[0050] In other specific implementations of the embodiments of the present invention, based on the target detection method, the rectal cancer CT initial image and the rectal cancer CT restored image are input into a trained target detection model, the corresponding lesion area is output, and the lesion area segmented from the rectal cancer CT initial image is used as the lesion initial area, and the lesion area segmented from the rectal cancer CT restored image is used as the lesion restored area. In a specific implementation of the embodiments of the present invention, the target detection model selects the R-CNN model, which can be adjusted according to the specific implementation environment. The R-CNN model is a commonly used model in target detection and will not be further described here.

[0051] The first determination module 102 is used to determine the matching recovery area of ​​the initial lesion area according to the intersection between the initial lesion area and the lesion recovery areas corresponding to other sampling moments; and to determine the initial effect change curve of the initial lesion area according to the overall grayscale change of all matching recovery areas.

[0052] Before analyzing changes in the lesion area, it is first necessary to determine the corresponding lesion recovery areas after each change in the initial lesion area in different rectal cancer CT restoration images. Since the spread of rectal cancer lesions is restricted during the recovery process of radiotherapy, the lesion area generally does not undergo significant positional changes. Therefore, in some possible implementations of the present invention, the process of obtaining the matching restoration area includes:

[0053] After mapping the rectal cancer CT restored image at each sampling time onto the initial rectal cancer CT image, the lesion restored region with the largest overlap with the initial lesion region at each sampling time is used as the matching restored region of the initial lesion region in each rectal cancer CT restored image. In other words, the matching restored region is the lesion region at the corresponding position in the rectal cancer CT restored image at each sampling time. The changes in the matching restored region can be used to analyze the treatment effect or recovery effect of the corresponding initial lesion region.

[0054] When the condition improves after radiotherapy, the grayscale of the diseased area will return to normal, that is, the grayscale will become smaller and smaller, and as the treatment continues, the body's function recovery speed will also continue to increase, thereby affecting the decreasing trend of the grayscale value; that is, the better the treatment effect, the faster the grayscale value of the diseased area decreases, and the corresponding grayscale change value will continue to increase, and the larger the grayscale change value, the better the treatment and recovery effect.

[0055] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the initial effect change curve includes:

[0056] The grayscale value average of all pixels in the matching recovery area at the previous sampling moment corresponding to each matching recovery area is used as the contrast grayscale value of each matching recovery area; the grayscale value average of all pixels in each matching recovery area is used as the overall grayscale value; the grayscale change value of each matching recovery area is determined based on the difference between the contrast grayscale value and the overall grayscale value; the grayscale change values ​​of all matching recovery areas in the initial lesion area are arranged in chronological order and then curve fitting is performed to determine the initial effect change curve of the initial lesion area. For the initial effect change curve, since its vertical axis is the grayscale change value, the grayscale change value reflects the degree of grayscale reduction in the lesion area between adjacent sampling moments, and the degree of grayscale reduction reflects the treatment effect, the initial effect change curve can intuitively reflect the effect of rectal cancer radiotherapy in the dimension of the overall change of grayscale value. It should be noted that curve fitting is a technical means well known to those skilled in the art and will not be further defined or elaborated here.

[0057] In a specific implementation of the embodiment of the present invention, the grayscale change value acquisition process is expressed as follows: ;in, For the The initial lesion area is in Grayscale change value of the matching recovery area corresponding to the sampling moment; For the The initial lesion area is in The overall grayscale value of the matching recovery area corresponding to the sampling moment, that is, The contrast gray value of the matching restoration area corresponding to the sampling moment; For the The initial lesion area is in It should be noted that the grayscale change value of the matching recovery area at the first sampling moment is calculated with the grayscale change value of the initial lesion area; in addition, for patients, there is usually only one initial lesion area, so in order to avoid the influence of chance, this application uses the first sampling moment as the grayscale change value. The calculation and analysis are performed on the initial lesion area, thereby preventing the situation where calculation cannot be performed when there are multiple initial lesion areas.

[0058] The second determination module 103 is used to determine the degree of diffusion recovery at each sampling moment based on the overall trend of edge changes between the edge of the initial lesion area and each corresponding matching recovery area; and to determine the reference recovery index based on the area reduction of each lesion matching recovery area and the degree of diffusion recovery before each sampling moment.

[0059] Since the degree of change in CT images of patients in different recovery stages is usually small, the corresponding grayscale changes are not easy to capture. Therefore, the effect of rectal cancer radiotherapy based solely on the initial effect change curve is poor. Therefore, it is necessary to further combine other aspects to conduct a more intuitive analysis of the effect of rectal cancer radiotherapy. After rectal cancer cells become cancerous, the cancerous area of ​​the rectum spreads to the surrounding area with the initial cancerous cells as the origin. During the improvement process after radiotherapy, the spread of cancerous cells will be suppressed and the lesion area will gradually shrink, which is manifested as the lesion area gradually shrinking inward, and the direction of convergence is relatively consistent.

[0060] Therefore, the present application first analyzes the treatment effect based on the contraction characteristics during the recovery process. Preferably, in some possible implementations of the present invention, the process of obtaining the degree of diffusion recovery includes:

[0061] The area of ​​each corresponding matching recovery area is proportionally enlarged to the area size of the corresponding initial lesion area with the corresponding centroid as the center to obtain the corresponding reference recovery area; after mapping the reference recovery area to the image where the corresponding initial lesion area is located, each boundary pixel point of the initial lesion area is connected with the nearest boundary pixel point on the reference recovery area to obtain all first reference lines; all intersection points obtained after extending all first reference lines are used as corresponding reference convergence points; the average of all distances between all reference convergence points corresponding to each matching recovery area is used as the corresponding convergence distance; according to the convergence distance and the number of reference convergence points, the diffusion recovery degree of each matching recovery area is determined.

[0062] Since the direction of convergence during the recovery process conforms to the consistency characteristic, ideally, when the area of ​​the matched recovery region is proportionally enlarged with the corresponding centroid as the center, the corresponding reference recovery region should be consistent with the corresponding initial lesion area. However, there are usually certain deviations during implementation. Considering the consistency of the convergence direction, the intersection points of the lines corresponding to the convergence directions of each edge point, namely the first reference lines, will be closer. The closer the corresponding intersection points are, the higher the consistency of the convergence direction and the better the consistency of the recovery degree. Therefore, the smaller the convergence distance, the greater the degree of diffusion recovery representing the recovery effect. The fewer the number of reference convergence points, the more overlapped the intersection points are, the better the consistency of the convergence direction, and the greater the corresponding degree of diffusion recovery representing the recovery effect. Therefore, the process of further determining the degree of diffusion recovery of each matched recovery region based on the convergence distance and the number of reference convergence points includes: performing a negative correlation mapping on the product of the number of reference convergence points and the convergence distance to determine the degree of diffusion recovery of each matched recovery region.

[0063] In a specific implementation of the embodiment of the present invention, the process of obtaining the diffusion recovery degree is expressed by the formula: ;in, For the The initial lesion area is in The diffusion recovery degree of the matching recovery area corresponding to the sampling moment; For the The initial lesion area is in The average of all distances between all reference convergence points corresponding to the matching recovery area at the sampling moment, that is, the Euclidean distance between each reference convergence point and each other reference convergence point is counted, and the average of all Euclidean distances corresponding to all reference convergence points is taken as the corresponding convergence distance; For the The initial lesion area is in The number of reference convergence points corresponding to the matching recovery area at each sampling moment. is an exponential function with a natural constant as the base. Implementers can adopt other negative correlation mapping methods according to the specific implementation environment, such as 、 and the reciprocal, where is the hyperbolic tangent function, which will not be further described here.

[0064] As the treatment effect gets better and better, the cancerous cells of rectal cancer are controlled, the immune system begins to gradually return to normal, and the lesion area will gradually decrease. As the treatment progresses, the body gradually improves, and the area of ​​the lesion area will decrease faster and faster. Therefore, based on the degree of diffusion recovery and the change in the area of ​​the lesion recovery area, the treatment effect is further analyzed. Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the reference recovery index includes:

[0065] The difference between the area of ​​each matched recovery region corresponding to the initial lesion region and the area of ​​the corresponding matched recovery region at the previous sampling moment is negatively correlated to determine the area reduction index of each matched recovery region. The product of the area reduction index and the degree of diffusion recovery is normalized to determine the reference recovery index of each matched recovery region. Similar to the principle of grayscale change value, the greater the reduction in the area of ​​each matched recovery region compared to the area of ​​the matched recovery region at the previous sampling moment, the more consistent it is with the characteristic that the area of ​​the lesion region decreases faster and faster when the treatment effect is good. Here, the area of ​​each matched recovery region is subtracted from the area of ​​the matched recovery region at the previous sampling moment. Therefore, the corresponding difference needs to be negatively correlated and combined with the degree of diffusion recovery to obtain a reference recovery index representing the recovery effect of each matched recovery region.

[0066] In a specific implementation of the embodiment of the present invention, the process of obtaining the reference recovery index is expressed as follows: ;in, For the The initial lesion area is in The reference recovery index of the matching recovery area corresponding to the sampling time; For the The initial lesion area is in The diffusion recovery degree of the matching recovery area corresponding to the sampling moment; No. The initial lesion area is in The area of ​​the matching recovery region corresponding to the sampling moment; For the The initial lesion area is in The area of ​​the matching recovery region corresponding to the sampling moment; For the The initial lesion area is in The area reduction index of the matching recovery area corresponding to the sampling moment is as follows: It is usually a negative value. After the inverse of the exponential function with the natural constant as the base is negatively mapped, the area reduction index obtained will take a positive value and is usually greater than 1. is a linear normalization function.

[0067] The radiotherapy effect monitoring module 104 is used to determine the degree of shape recovery based on the edge smoothness of the matching recovery area, the reference recovery index, and the circular convergence in shape; to correct the initial effect change curve based on the degree of shape recovery to determine the corrected radiotherapy recovery curve; and to dynamically monitor the radiotherapy effect of rectal cancer based on the corrected radiotherapy recovery curve.

[0068] Because of the quasi-circular nature of the rectal tissue, as the treatment continues, the lesion area gradually shrinks inward and tends to disappear, and as the treatment progresses, the corresponding edge will gradually become smoother and the shape will approach a circle until it disappears; that is, for the matching recovery area at each sampling moment, the smoother and more circular the corresponding edge is, the greater the degree of regression of the lesion area and the better the treatment effect.

[0069] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the degree of shape recovery includes:

[0070] The number of corner points in each matched recovery area is detected using a corner detection algorithm; the product of the negative correlation mapping value of the number of corner points and the reference recovery index is used as the edge recovery index. The more corner points there are, the less smooth the edge is. Therefore, the number of corner points is negatively correlated and mapped, and then multiplied by the reference recovery index, so that the resulting edge recovery index represents the radiotherapy treatment effect. In a specific implementation of an embodiment of the present invention, the corner detection algorithm uses the Harris corner detection algorithm, which can be adjusted according to the specific implementation environment. The Harris corner detection algorithm is a technical means well known to those skilled in the art and will not be further defined or elaborated here.

[0071] The distance between each boundary pixel point and the corresponding centroid in the matching restoration area is used as the centroid distance of each boundary pixel point; the centroid distance difference between each boundary pixel point and each other boundary pixel point is calculated; the mean of all centroid distance differences corresponding to all boundary pixels are negatively correlated to determine the shape restoration index of the matching restoration area. For a circle, the length from the point on the boundary to the centroid point is the same, and the corresponding centroid distance difference is 0. Therefore, the smaller the mean of all centroid distance differences corresponding to all pixels, the better the corresponding restoration effect, that is, the larger the shape restoration index, the better the restoration effect. Finally, based on the relationship between the shape restoration index, the edge restoration index and the restoration effect, the product of the shape restoration index and the edge restoration index is normalized to determine the shape restoration degree of the matching restoration area. It should be noted that the boundary pixel point is the pixel point located on the boundary of the region.

[0072] In a specific implementation of the embodiment of the present invention, the process of obtaining the degree of shape recovery is expressed by the formula: ;in, For the The initial lesion area is in The shape recovery degree of the matching recovery area corresponding to the sampling moment; For the The initial lesion area is in The number of corner points in the matching recovery area corresponding to the sampling time; For the The initial lesion area is in The reference recovery index of the matching recovery area corresponding to the sampling time; For the The initial lesion area is in The edge recovery index of the matching recovery area corresponding to the sampling moment; For the The initial lesion area is in The mean of all centroid distance differences corresponding to all boundary pixels in the matching recovery area at the sampling moment; For the The initial lesion area is in The shape recovery index of the matching recovery area corresponding to the sampling moment. It should be noted that, in order to ensure that the calculation results are meaningful, when performing fractional operations in the embodiment of the present invention, when encountering a denominator of 0, it is necessary to add a parameter adjustment factor greater than 0 to the denominator to prevent the denominator from being 0. The value of the parameter adjustment factor is set by the implementer according to the actual situation. In this application, it is set to 0.1.

[0073] The degree of shape recovery combines the characterization of the radiotherapy effect of rectal cancer by boundary convergence direction, area change, edge smoothness and regional quasi-circularity. Therefore, the initial effect change curve can be corrected by the obtained shape recovery degree, so that the corrected radiotherapy recovery curve reflects the treatment effect more intuitively, thereby performing more intuitive and accurate dynamic monitoring of the radiotherapy effect of rectal cancer based on the corrected radiotherapy recovery curve.

[0074] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the modified radiotherapy recovery curve includes:

[0075] The grayscale change values ​​corresponding to the matched recovery region on the initial effect change curve are corrected based on the degree of shape recovery to determine the corrected change value of the matched recovery region. The corrected change values ​​of all matched recovery regions in the initial lesion region are arranged in chronological order and then curve-fitted to determine the corrected radiotherapy recovery curve for the initial lesion region. Because the initial effect change curve is obtained by fitting the grayscale change values ​​of each matched recovery region, the grayscale change values ​​are corrected based on the degree of shape recovery obtained for each matched recovery region, which represents the recovery effect or treatment effect, to achieve the purpose of correcting the curve.

[0076] Preferably, in a specific implementation of an embodiment of the present invention, the process of obtaining the corrected change value includes: taking the product of the positive correlation mapping value of the shape recovery degree and the grayscale change value of the corresponding matching recovery area on the initial effect change curve as the corrected change value of the matching recovery area. The reason for performing positive correlation mapping on the shape recovery degree is that the value range of the shape recovery degree is 0 to 1. If it is directly multiplied by the grayscale change value, it will only make the obtained corrected radiotherapy recovery curve less obvious. Therefore, it is necessary to perform positive correlation mapping on it so that the value of the shape recovery degree is not less than 1. In a specific implementation of an embodiment of the present invention, the process of obtaining the corrected change value is expressed by the formula: ;in, For the The initial lesion area is in Corrected change value of the matching recovery area corresponding to the sampling moment; For the The initial lesion area is in Grayscale change value of the matching recovery area corresponding to the sampling moment; For the The initial lesion area is in The shape recovery degree of the matching recovery area corresponding to each sampling moment is calculated; a positive correlation mapping is performed by adding the real number 1 to the shape recovery degree, so that the correction change value is always magnified compared to the grayscale change value, thereby improving the intuitiveness of the treatment effect observation.

[0077] Finally, the radiotherapy effect of rectal cancer is dynamically monitored based on the obtained modified radiotherapy recovery curve. The curve can be used to monitor the radiotherapy effect of rectal cancer more intuitively, and the presentation of the curve can enable medical students to learn and understand the radiotherapy effect of rectal cancer more intuitively.

[0078] In summary, this application determines a reference recovery index that characterizes the recovery effect from the dimensions of edge change and area change based on the characteristics that radiotherapy inhibits the spread of cancer cells so that the edges converge in the same direction toward the center where the infection is more severe, and the characteristics that the lesion area becomes smaller as the treatment effect becomes better. Then, based on the reference recovery index and the characteristics that the quasi-circularity of rectal tissue causes the lesion area to gradually shrink inward and the edges become smoother, a more accurate shape recovery degree that characterizes the reference recovery index is determined. Thus, the corrected radiotherapy recovery curve is more accurately corrected according to the shape recovery degree, making the dynamic monitoring of the radiotherapy effect of rectal cancer more intuitive and accurate.

[0079] The present application also provides a computer device. Figure 2 , which shows a schematic diagram of the structure of a computer device provided by an embodiment of the present invention, the computer device includes a memory 201, a processor 202, and a computer program 203 stored in the memory 201 and running on the processor 202, wherein when the processor 202 executes the computer program 203, the computer device can execute any one of the above-mentioned combined imaging analysis dynamic monitoring systems for rectal cancer radiotherapy.

[0080] An embodiment of the present application also provides a computer program product. When the computer program product is run on a computer device, the computer device can execute any one of the above-mentioned combined imaging analysis dynamic monitoring systems for rectal cancer radiotherapy.

[0081] An embodiment of the present application also provides a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer device, the computer device can execute any one of the aforementioned combined imaging analysis dynamic monitoring systems for rectal cancer radiotherapy.

[0082] In the embodiments provided in the present application, it should be understood that the provided computer devices, computer program products, and computer-readable storage media are all used to execute the corresponding systems provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the systems provided above and will not be repeated here.

[0083] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0084] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A dynamic monitoring system for rectal cancer radiotherapy effects combined with imaging analysis, characterized in that: The system comprises: A data acquisition and preprocessing module is configured to acquire an initial CT image of rectal cancer in an initial state and a restored CT image of rectal cancer at each sampling moment during the course of rectal cancer radiotherapy; and determine an initial lesion region in the initial CT image of rectal cancer and a restored lesion region in the restored CT image of rectal cancer based on an image segmentation method; A first determination module is configured to determine a matching recovery region of the initial lesion region based on an intersection between the initial lesion region and the lesion recovery regions corresponding to each other sampling moment; and determine an initial effect change curve of the initial lesion region based on an overall grayscale change of all matching recovery regions; The second determination module is used to determine the degree of diffusion recovery at each sampling moment according to the overall trend of edge changes between the edge of the initial lesion area and each corresponding matching recovery area; determine the reference recovery index according to the area reduction of each lesion matching recovery area before each sampling moment and the degree of diffusion recovery; the degree of diffusion recovery is determined by negatively correlating the product of the number of corresponding reference convergence points and the corresponding convergence distance, the corresponding reference convergence points are all intersection points obtained by extending all first reference lines, all first reference lines are obtained by mapping the reference recovery area to the image where the corresponding initial lesion area is located, and then connecting each boundary pixel point of the initial lesion area with the nearest boundary pixel point on the reference recovery area, the reference recovery area is obtained by proportionally enlarging the area of ​​each corresponding matching recovery area with the corresponding centroid as the center to the area size of the corresponding initial lesion area, and the corresponding convergence distance is the average of all distances between all reference convergence points corresponding to each matching recovery area; The radiotherapy effect monitoring module is used to determine the degree of shape recovery based on the edge smoothness of the matching recovery area, the reference recovery index and the circular convergence in shape; correct the initial effect change curve according to the shape recovery degree to determine the corrected radiotherapy recovery curve; and dynamically monitor the radiotherapy effect of rectal cancer based on the corrected radiotherapy recovery curve.

2. A dynamic monitoring system for rectal cancer radiotherapy effect combined with imaging analysis according to claim 1, characterized in that: The process of determining the initial lesion area in the rectal cancer CT initial image and the restored lesion area in the rectal cancer CT restored image based on the image segmentation method includes: The CA significance algorithm is used to calculate the average significance value of each segmented region in the rectal cancer CT initial image at all segmentation scales; the segmented region whose normalized value of the average significance value is greater than a preset significance threshold is used as the lesion initial region; based on the acquisition process of the lesion initial region, the lesion recovery region in the rectal cancer CT recovery image is determined.

3. The system for dynamic monitoring of rectal cancer radiotherapy effects combined with imaging analysis according to claim 1, characterized in that: The process of obtaining the matching recovery area includes: After mapping the rectal cancer CT restored image at each sampling moment to the rectal cancer CT initial image, among all the lesion restored areas corresponding to each sampling moment, the lesion restored area with the largest overlapping area with the lesion initial area is used as the matching restored area of ​​the lesion initial area in each rectal cancer CT restored image.

4. The system for dynamic monitoring of rectal cancer radiotherapy effects combined with imaging analysis according to claim 1, characterized in that: The process of obtaining the initial effect change curve includes: The average grayscale value of all pixels in the matching recovery area at the previous sampling moment corresponding to each matching recovery area is used as the contrast grayscale value of each matching recovery area; the average grayscale value of all pixels in each matching recovery area is used as the overall grayscale value; the grayscale change value of each matching recovery area is determined according to the difference between the contrast grayscale value and the overall grayscale value; the grayscale change values ​​of all matching recovery areas in the initial lesion area are arranged in chronological order and curve fitting is performed to determine the initial effect change curve of the initial lesion area.

5. The system for dynamic monitoring of rectal cancer radiotherapy effects combined with imaging analysis according to claim 1, characterized in that: The process of obtaining the reference recovery index includes: The difference between the area of ​​each matching recovery area corresponding to the initial area of ​​the lesion and the area of ​​the corresponding matching recovery area at the previous sampling moment is negatively correlated to determine the area reduction index of each matching recovery area; the product of the area reduction index and the diffusion recovery degree is normalized to determine the reference recovery index of each matching recovery area.

6. The system for dynamic monitoring of rectal cancer radiotherapy effects combined with imaging analysis according to claim 1, characterized in that: The process of obtaining the shape recovery degree includes: Detecting the number of corner points in each matching restoration area using a corner detection algorithm; multiplying the negative correlation mapping value of the number of corner points by the reference restoration index as the edge restoration index; The distance between each boundary pixel point and the corresponding centroid in the matched restoration area is used as the centroid distance of each boundary pixel point; the centroid distance difference between each boundary pixel point and each other boundary pixel point is calculated; the mean of all centroid distance differences corresponding to all boundary pixels is negatively correlated and mapped to determine the shape restoration index of the matched restoration area; The product of the shape recovery index and the edge recovery index is normalized to determine the shape recovery degree of the matching recovery area.

7. The system for dynamic monitoring of rectal cancer radiotherapy effects combined with imaging analysis according to claim 4, characterized in that: The process of obtaining the modified radiotherapy recovery curve includes: Correcting the grayscale change value corresponding to the matching restoration area on the initial effect change curve according to the shape restoration degree, and determining the corrected change value of the matching restoration area; The corrected change values ​​of all matching recovery areas of the lesion initial area are arranged in time sequence and then curve fitting is performed to determine a corrected radiotherapy recovery curve of the lesion initial area.

8. The system for dynamic monitoring of rectal cancer radiotherapy effects combined with imaging analysis according to claim 7, characterized in that: The process of obtaining the modified change value includes: The product of the positive correlation mapping value of the shape restoration degree and the grayscale change value of the corresponding matching restoration area on the initial effect change curve is used as the corrected change value of the matching restoration area.

Images