A method and system for discerning the degree of cancer risk
By constructing a graph structure and using graph convolutional networks and generative adversarial networks to generate simulated prostate cancer growth videos, combined with a long short-term neural network model, the accuracy problem of prostate cancer risk assessment in existing technologies is solved, achieving more accurate and consistent prostate cancer risk assessment and dynamic analysis.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 简勇
- Filing Date
- 2025-09-08
- Publication Date
- 2026-05-12
AI Technical Summary
Existing prostate cancer screening methods rely on PSA levels, which are not accurate enough. PI-RADS scores rely on subjective judgment, leading to inconsistent diagnostic grading and making it difficult to accurately assess prostate cancer risk.
By acquiring the personal information and prostate-specific antigen (PSA) test results of the screening subjects, a graph structure is constructed and graph convolutional networks and generative adversarial networks are used to generate a simulated prostate cancer growth video. The risk of prostate cancer is then assessed by combining a long short-term neural network model with a short-term neural network model.
It enables accurate assessment of prostate cancer risk, improves the objectivity and consistency of diagnosis, and provides a dynamic view of prostate cancer development.
Smart Images
Figure CN121101470B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of prostate cancer technology, specifically to a method and system for identifying the degree of cancer risk. Background Technology
[0002] Prostate cancer is one of the most common malignant tumors in men, seriously threatening their lives and health. The main methods for prostate cancer screening include prostate-specific antigen (PSA) testing and digital rectal examination (DRE). However, in current clinical practice, prostate biopsy is often determined solely by PSA levels. While convenient, this approach has limitations in accuracy, as elevated PSA levels do not always indicate prostate cancer and can be caused by other non-malignant factors, such as benign prostatic hyperplasia (BPH) or inflammation. In recent years, multiparametric magnetic resonance imaging (mpMRI) has gradually become an important tool for prostate cancer diagnosis. mpMRI provides high-resolution images of the internal structures of the prostate, helping to identify potential malignant lesions. The Prostate Imaging Reporting and Data System (PI-RADS) based on mpMRI significantly improves the accuracy of prostate cancer screening by grading clinically significant prostate cancers using a standardized scoring system. However, PI-RADS scoring relies heavily on the subjective judgment of radiologists, which can lead to inconsistencies and errors in prostate cancer diagnostic grading.
[0003] Therefore, how to accurately assess the risk of developing prostate cancer is an urgent problem to be solved. Summary of the Invention
[0004] The main technical problem addressed by this invention is how to accurately assess the risk of developing prostate cancer.
[0005] According to a first aspect, the present invention provides a method for identifying cancer risk levels, comprising: acquiring personal information of a screening subject and the prostate-specific antigen (PSA) test results of the screening subject; determining a risk value of the antigen test of the screening subject based on the personal information of the screening subject and the PSA test results of the screening subject; if the risk value of the antigen test of the screening subject is greater than a preset threshold, acquiring a prostate magnetic resonance imaging (MRI) scan image of the screening subject; constructing a graph structure based on the PSA test results of the screening subject and the prostate MRI scan image of the screening subject, the graph structure including two nodes and an edge between the two nodes, the two nodes including an antigen test node and an MRI test node, the node features of the antigen test node including the personal information of the screening subject and the PSA test results of the screening subject, the node features of the MRI test node including the prostate MRI scan image of the screening subject, and the features of the edge between the nodes including the consistency between the antigen test and the MRI test; processing the graph structure based on a graph convolutional network to determine prostate cancer analysis information; generating a prostate cancer simulated growth video using a generative adversarial network based on the prostate MRI scan image of the screening subject and the prostate cancer analysis information; and determining the prostate cancer risk level based on the prostate cancer simulated growth video.
[0006] In one possible implementation, determining the prostate cancer risk level based on the simulated prostate cancer growth video includes: processing the simulated prostate cancer growth video based on a risk level determination model to determine the prostate cancer risk level, wherein the risk level determination model is a long short-term neural network model.
[0007] In one possible implementation, acquiring multiple prostate cancer images arranged in chronological order further includes: acquiring prostate MRI scan images of the screening target further includes: preprocessing the prostate MRI scan images of the screening target, the preprocessing including denoising, standardization, and contrast enhancement.
[0008] In one possible implementation, the graph convolutional network takes the graph structure as input and outputs prostate cancer analysis information.
[0009] According to a second aspect, the present invention provides a cancer risk assessment system, comprising:
[0010] The first acquisition module is used to acquire the personal information of the screening subjects and the prostate-specific antigen test results of the screening subjects.
[0011] An antigen screening module is used to determine the risk value of the antigen test for the screening subject based on the personal information of the screening subject and the prostate-specific antigen test results of the screening subject.
[0012] The second acquisition module is used to acquire the prostate magnetic resonance scan image of the screening object if the risk value of the antigen test of the screening object is greater than a preset threshold.
[0013] The graph structure module is used to construct a graph structure based on the prostate-specific antigen (PSA) test results and the prostate magnetic resonance imaging (MRI) scan images of the screening subjects. The graph structure includes two nodes and an edge between the two nodes. The two nodes include an antigen test node and an MRI test node. The node features of the antigen test node include the personal information of the screening subjects and the PSA test results of the screening subjects. The node features of the MRI test node include the prostate MRI scan images of the screening subjects. The features of the edge between the nodes include the consistency between the antigen test and the MRI test.
[0014] An analysis information determination module is used to process the graph structure based on a graph convolutional network to determine prostate cancer analysis information;
[0015] A generative adversarial module is used to generate a simulated prostate cancer growth video using a generative adversarial network based on the prostate MRI scan images of the screened subjects and the prostate cancer analysis information.
[0016] A risk level determination module is used to determine the risk level of prostate cancer based on the simulated prostate cancer growth video.
[0017] In one possible implementation, the danger level determination module is further configured to:
[0018] The risk level of prostate cancer is determined by processing the simulated growth video of prostate cancer based on a risk level determination model, wherein the risk level determination model is a long short-term neural network model.
[0019] In one possible implementation, the first acquisition module is further configured to:
[0020] The prostate MRI scan images of the screening subjects are preprocessed, including noise reduction, standardization, and contrast enhancement.
[0021] In one possible implementation, the graph convolutional network takes the graph structure as input and outputs prostate cancer analysis information.
[0022] According to a third aspect, embodiments of the present invention provide an electronic device, including: a processor; a memory; and a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method as described above, the method including: acquiring personal information of a screening subject and prostate-specific antigen (PSA) test results of the screening subject; determining a risk value for the PSA test of the screening subject based on the personal information of the screening subject and the PSA test results of the screening subject; if the risk value for the PSA test of the screening subject is greater than a preset threshold, acquiring a prostate magnetic resonance imaging (MRI) image of the screening subject; and constructing a graph based on the PSA test results of the screening subject and the prostate MRI image of the screening subject. The graph structure comprises two nodes and an edge between them. The two nodes are an antigen testing node and an MRI testing node. The node features of the antigen testing node include the screening subject's personal information and the screening subject's prostate-specific antigen (PSA) test results. The node features of the MRI testing node include the screening subject's prostate MRI scan image. The edge features between the nodes include the consistency between the antigen test and the MRI scan. The graph structure is processed using a graph convolutional network to determine prostate cancer analysis information. Based on the screening subject's prostate MRI scan image and the prostate cancer analysis information, a generative adversarial network (GAN) is used to generate a simulated prostate cancer growth video. The prostate cancer risk level is determined based on the simulated prostate cancer growth video.
[0023] According to the fourth aspect, this embodiment provides a computer-readable storage medium storing a computer program thereon. When executed by a processor, the program implements the aforementioned cancer risk identification method. The method includes: acquiring personal information of a screening subject and the prostate-specific antigen (PSA) test results of the screening subject; determining a risk value for the PSA test based on the personal information of the screening subject and the PSA test results; if the risk value for the PSA test is greater than a preset threshold, acquiring a prostate magnetic resonance imaging (MRI) image of the screening subject; and constructing a graph structure based on the PSA test results and the prostate MRI image of the screening subject, the graph structure including two... An edge exists between two nodes, including an antigen testing node and an MRI testing node. The node features of the antigen testing node include the screening subject's personal information and the screening subject's prostate-specific antigen test results. The node features of the MRI testing node include the screening subject's prostate MRI scan image. The edge features between the nodes include the consistency between the antigen test and the MRI scan. A graph structure is processed using a graph convolutional network to determine prostate cancer analysis information. Based on the screening subject's prostate MRI scan image and the prostate cancer analysis information, a generative adversarial network is used to generate a prostate cancer simulated growth video. The prostate cancer risk level is determined based on the simulated growth video.
[0024] This invention provides a method and system for identifying cancer risk. The method includes: acquiring personal information of the screening subject and the prostate-specific antigen (PSA) test results of the screening subject; determining the risk value of the antigen test of the screening subject based on the personal information and the PSA test results; if the risk value of the antigen test of the screening subject is greater than a preset threshold, acquiring a prostate magnetic resonance imaging (MRI) scan image of the screening subject; constructing a graph structure based on the PSA test results and the prostate MRI scan image of the screening subject, the graph structure including two nodes and an edge between the two nodes, the two nodes including an antigen test node and an MRI scan node, the node features of the antigen test node including the personal information of the screening subject and the PSA test results of the screening subject, the node features of the MRI scan node including the prostate MRI scan image of the screening subject, and the features of the edge between the nodes including the consistency between the antigen test and the MRI scan; processing the graph structure based on a graph convolutional network to determine prostate cancer analysis information; generating a prostate cancer simulated growth video using a generative adversarial network based on the prostate MRI scan image of the screening subject and the prostate cancer analysis information; and determining the prostate cancer risk level based on the prostate cancer simulated growth video. This method can accurately assess the risk of developing prostate cancer. Attached Figure Description
[0025] Figure 1 This is a schematic diagram illustrating an application scenario of a cancer risk assessment method provided in an embodiment of the present invention.
[0026] Figure 2 A flowchart illustrating a method for identifying the degree of cancer risk provided in an embodiment of the present invention;
[0027] Figure 3 A schematic diagram of a graph structure provided in an embodiment of the present invention;
[0028] Figure 4 This is a schematic diagram of a cancer risk identification system provided in an embodiment of the present invention;
[0029] Figure 5 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation
[0030] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of the invention. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to the present invention are not shown or described in the specification. This is to avoid obscuring the core parts of the invention with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.
[0031] Figure 1 This is a schematic diagram illustrating an application scenario of a cancer risk assessment method provided in an embodiment of the present invention. Figure 1 The application scenarios for the cancer risk identification method can include servers 11, networks 12, terminals 13, and storage devices 14.
[0032] In some embodiments, server 11 may be a single server or a group of servers. Server 11 can access information and / or data stored in terminal 13 or storage device 14 via network 12. In some embodiments, server 11 may be used to perform... Figure 2 The method for identifying the level of cancer risk is shown in the figure.
[0033] Network 12 can facilitate the exchange of information and / or data. In some embodiments, network 12 can be any form of wired or wireless network, or any combination thereof.
[0034] Terminal 13 may refer to one or more terminal devices used by a user. In some embodiments, terminal 13 may include one or more combinations of mobile devices, tablet computers, laptop computers, etc.
[0035] Storage device 14 can store data and / or instructions, for example, storage device 14 can store data instructions for cancer risk assessment methods.
[0036] In this embodiment of the invention, the following are provided: Figure 2 The method for identifying the degree of cancer risk shown includes steps S1 to S7:
[0037] Step S1: Obtain the personal information of the screening subject and the prostate-specific antigen test results of the screening subject.
[0038] The screening is for individuals who undergo prostate cancer screening.
[0039] The personal information of those being screened includes, but is not limited to, age, gender, family medical history, genetic information, and past medical history.
[0040] The prostate-specific antigen (PSA) test is a blood test used to measure the level of prostate-specific antigen in the blood. PSA is a protein produced by prostate cells, and normally only a small amount enters the bloodstream. PSA test results are usually expressed in ng / mL.
[0041] PSA levels can serve as a preliminary risk assessment indicator, helping to determine whether further investigation is needed.
[0042] Step S2: Determine the risk value of the antigen test for the screening subject based on the personal information of the screening subject and the prostate-specific antigen test results of the screening subject.
[0043] In some embodiments, the risk value of the antigen test for the screening subject can be determined based on the screening subject's personal information, the screening subject's prostate-specific antigen (PSA) test results, and a preset table of risk values for the antigen test. The preset table includes the screening subject's personal information, the screening subject's PSA test results, and the corresponding risk values for the antigen test. The preset table can be manually constructed based on historical data.
[0044] The risk value of the screening subject antigen test represents the risk value for prostate cancer determined based on the information from the screening subject antigen test.
[0045] Step S3: If the risk value of the antigen test of the screening object is greater than a preset threshold, then obtain the prostate magnetic resonance scan image of the screening object.
[0046] The preset threshold is a pre-set threshold. When the risk value of the antigen test of the screening object is greater than the preset threshold, the MRI scan image of the screening object's prostate is obtained.
[0047] A prostate magnetic resonance imaging (MRI) image is an image obtained by scanning the prostate region using magnetic resonance imaging (MRI) technology. In some embodiments, obtaining the prostate MRI image of the screening subject further includes preprocessing the prostate MRI image of the screening subject, the preprocessing including noise reduction, normalization, and contrast enhancement.
[0048] Step S4: Construct a graph structure based on the prostate-specific antigen (PSA) test results and the prostate MRI scan image of the screening subject. The graph structure includes two nodes and an edge between the two nodes. The two nodes include an antigen test node and an MRI test node. The node features of the antigen test node include the screening subject's personal information and the PSA test results. The node features of the MRI test node include the prostate MRI scan image of the screening subject. The features of the edge between the nodes include the consistency between the antigen test and the MRI test.
[0049] A graph structure is a data structure used to represent entities (nodes) and their relationships (edges). Two nodes include an antigen testing node and an MRI testing node. The node characteristics of the antigen testing node include the screening subject's personal information and the screening subject's prostate-specific antigen (PSA) test results. The node characteristics of the MRI testing node include the screening subject's prostate MRI scan image. The edge characteristics between the nodes include the degree of consistency between the antigen test and the MRI results. In some embodiments, Figure 3 This is a schematic diagram of a graph structure provided in an embodiment of the present invention. As shown in the figure, node A represents an antigen testing node, and node B represents a magnetic resonance imaging (MRI) testing node. The characteristics of the edges between nodes include the degree of consistency between antigen testing and MRI testing.
[0050] The diagnosis of prostate cancer often relies on multiple data types, such as PSA test results and MRI images. Complex relationships can exist between these data modalities. Graph structures can effectively represent the relationships between these data, enabling graph convolutional networks to learn cross-modal feature representations.
[0051] In some embodiments, a deep neural network model can be used to process the prostate MRI scan image and the prostate-specific antigen (PSA) test results of the screening subject to determine the degree of consistency between the antigen test and the MRI scan. The input to the deep neural network model is the prostate MRI scan image and the PSA test results of the screening subject, and the output of the deep neural network model is the degree of consistency between the antigen test and the MRI scan. The deep neural network model includes deep neural networks (DNNs).
[0052] Step S5: Process the graph structure based on a graph convolutional network to determine prostate cancer analysis information.
[0053] In some embodiments, prostate cancer analysis information includes the presence or absence of prostate cancer, tumor location, size, shape, density, margin clarity, and whether there are signs of invasion of surrounding tissues or organs, as well as the reliability of antigen testing. The graph convolutional network takes the graph structure as input and outputs the prostate cancer analysis information.
[0054] Graph convolutional networks (GCNNs) can capture the complex relationships between nodes in a graph, including node connection patterns and node features. In prostate cancer analysis, this means that GCNNs can simultaneously consider PSA test results and MRI image information, as well as the consistency between them. GCNNs can integrate data from different sources, such as antigen test nodes and MRI test nodes, into a unified framework. This allows the model to consider the individual profile of the screening subject, PSA levels, and MRI image features simultaneously. GCNNs can learn low-dimensional feature representations of nodes, which can capture the intrinsic properties of nodes and the relationships between them. In prostate cancer analysis, this helps in identifying tumor characteristics and behavior.
[0055] Step S6: Based on the prostate MRI scan images of the screened subjects and the prostate cancer analysis information, a generative adversarial network is used to generate a prostate cancer simulated growth video.
[0056] Generative Adversarial Networks (GANs) are deep learning models consisting of a generator and a discriminator. The generator is responsible for generating new images, while the discriminator is responsible for distinguishing between generated and real images. Through adversarial training, the generator gradually learns to generate realistic images.
[0057] The prostate cancer simulated growth video is a sequence of images generated by a generative adversarial network, simulating the changes of prostate cancer over time, and presented in video format.
[0058] By generating simulated growth videos, the changes in prostate cancer over a preset time period can be visually displayed. The input to the generative adversarial network (GAN) consists of multiple prostate cancer images arranged in chronological order and the degree of prostate cancer development; the output of the GAN is a simulated prostate cancer growth video.
[0059] Generative Adversarial Networks (GANs) can synthesize a series of images that appear to depict the actual growth process of prostate cancer. These synthesized images can fill in gaps in the time series, providing a continuous view of how the cancer progresses over time. These images can reveal subtle changes in the cancer's development that might not be apparent in real images due to limitations in sampling frequency or image quality. The input to the GAN is a prostate MRI scan image of the screening subject and the prostate cancer analysis information; the output of the GAN is a simulated prostate cancer growth video.
[0060] Step S7: Determine the risk level of prostate cancer based on the simulated prostate cancer growth video.
[0061] The prostate cancer risk level is a quantitative indicator derived from analyzing simulated prostate cancer growth videos, reflecting a patient's risk of developing prostate cancer. The prostate cancer risk level can be a value between 0 and 1; the higher the value, the greater the risk of developing prostate cancer.
[0062] In some embodiments, the prostate cancer simulated growth video can be processed based on a risk determination model to determine the risk level of prostate cancer. This risk determination model is a long short-term neural network (LSN) model. The LSN model can process sequence data of arbitrary length, capture sequence information, and output results based on the correlation between preceding and following data within the sequence. Processing the prostate cancer simulated growth video at consecutive time points using an LSN model can output features that consider the correlation between the simulated growth videos at different time points, making the output features more accurate and comprehensive.
[0063] In some embodiments, the risk determination model includes an information change sequence determination layer, a risk information determination layer, and a prostate cancer risk determination layer. The inputs to the information change sequence determination layer are time-series data of tumor size, tumor shape, tumor location, tumor density, and tumor edge clarity. The outputs of the information change sequence determination layer are the same time-series data of tumor size, tumor shape, tumor location, tumor density, and tumor edge clarity. The inputs to the risk information determination layer are the same time-series data of tumor size, tumor shape, tumor location, tumor density, and tumor edge clarity. The outputs of the risk information determination layer are the risk levels of tumor size, tumor shape, tumor location, tumor density, and tumor edge clarity. The inputs to the risk determination layer are the same time-series data of tumor size, tumor shape, tumor location, tumor density, and tumor edge clarity. The output of the risk determination layer is the risk level of prostate cancer.
[0064] The Information Change Sequence Determination Layer focuses on extracting time-series data of various features of prostate cancer at different time points, including tumor size, shape, location, density, and margin clarity. The Risk Information Determination Layer receives the time-series data output from the previous layer and evaluates each feature (tumor size, shape, location, density, and margin clarity) to determine its individual risk level. The Prostate Cancer Risk Determination Layer receives the risk levels of each feature output from the previous layer and synthesizes them to ultimately determine the overall risk level of prostate cancer. The main purpose of this layering is to progressively refine and comprehensively assess the risk level of prostate cancer. By decomposing the entire model into multiple layers, each focusing on processing a specific type of information, it ensures that the model can fully capture key features at each level and ultimately synthesize these features to determine the overall risk level of prostate cancer.
[0065] The development of prostate cancer is a dynamic process. Using long short-term neural network models to analyze simulated videos can better capture time-dependent relationships and thus determine the overall risk of prostate cancer.
[0066] Based on the same inventive concept Figure 4 This is a schematic diagram of a cancer risk assessment system provided in an embodiment of the present invention. The cancer risk assessment system includes:
[0067] The first acquisition module 41 is used to acquire the personal information of the screening subject and the prostate-specific antigen test results of the screening subject.
[0068] Antigen screening module 42 is used to determine the risk value of antigen test for the screening subject based on the personal information of the screening subject and the prostate-specific antigen test results of the screening subject;
[0069] The second acquisition module 43 is used to acquire the prostate magnetic resonance scan image of the screening object if the risk value of the antigen test of the screening object is greater than a preset threshold.
[0070] Graph structure module 44 is used to construct a graph structure based on the prostate-specific antigen test results and the prostate magnetic resonance scan image of the screening object. The graph structure includes two nodes and an edge between the two nodes. The two nodes include an antigen test node and a magnetic resonance test node. The node features of the antigen test node include the personal information of the screening object and the prostate-specific antigen test results of the screening object. The node features of the magnetic resonance test node include the prostate magnetic resonance scan image of the screening object. The features of the edge between the nodes include the consistency between the antigen test and the magnetic resonance test.
[0071] The analysis information determination module 45 is used to process the graph structure based on the graph convolutional network to determine prostate cancer analysis information;
[0072] Generative adversarial module 46 is used to generate a prostate cancer simulated growth video using a generative adversarial network based on the prostate magnetic resonance scan image of the screened object and the prostate cancer analysis information.
[0073] Risk level determination module 47 is used to determine the risk level of prostate cancer based on the prostate cancer simulated growth video.
[0074] Based on the same inventive concept, embodiments of the present invention provide an electronic device, such as... Figure 5 As shown, it includes:
[0075] Includes: a processor 51; a memory 52; and a computer program; wherein the computer program is stored in the memory 52 and configured to be executed by the processor 51 to implement the cancer risk identification method provided above, the method including: acquiring personal information of the screening subject and the prostate-specific antigen (PSA) test results of the screening subject; determining the risk value of the antigen test of the screening subject based on the personal information of the screening subject and the PSA test results of the screening subject; if the risk value of the antigen test of the screening subject is greater than a preset threshold, acquiring a prostate magnetic resonance imaging (MRI) image of the screening subject; constructing a graph structure based on the PSA test results of the screening subject and the prostate MRI image of the screening subject, The graph structure comprises two nodes and an edge between them. The two nodes are an antigen testing node and an MRI testing node. The node features of the antigen testing node include the screening subject's personal information and the screening subject's prostate-specific antigen (PSA) test results. The node features of the MRI testing node include the screening subject's prostate MRI scan image. The edge between the nodes includes the consistency between the antigen test and the MRI results. The graph structure is processed using a graph convolutional network to determine prostate cancer analysis information. Based on the screening subject's prostate MRI scan image and the prostate cancer analysis information, a generative adversarial network is used to generate a simulated prostate cancer growth video. The prostate cancer risk level is determined based on the simulated prostate cancer growth video.
[0076] Based on the same inventive concept, this embodiment provides a computer-readable storage medium storing a computer program thereon. When executed by processor 51, the program implements the aforementioned cancer risk identification method. The method includes: acquiring personal information of the screening subject and the prostate-specific antigen (PSA) test results of the screening subject; determining the risk value of the PSA test based on the personal information of the screening subject and the PSA test results of the screening subject; if the risk value of the PSA test is greater than a preset threshold, acquiring a prostate magnetic resonance imaging (MRI) image of the screening subject; and constructing a graph structure based on the PSA test results and the prostate MRI image of the screening subject, wherein the graph structure includes... The process involves two nodes and an edge between them. The two nodes are an antigen testing node and an MRI testing node. The node features of the antigen testing node include the screening subject's personal information and the screening subject's prostate-specific antigen (PSA) test results. The node features of the MRI testing node include the screening subject's prostate MRI scan image. The edge between the nodes includes the consistency between the antigen test and the MRI results. The graph structure is processed using a graph convolutional network to determine prostate cancer analysis information. Based on the screening subject's prostate MRI scan image and the prostate cancer analysis information, a generative adversarial network (GAN) is used to generate a simulated prostate cancer growth video. The prostate cancer risk level is determined based on the simulated prostate cancer growth video.
[0077] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.
[0078] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.
[0079] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.
[0080] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.
[0081] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.
Claims
1. A cancer risk assessment system, characterized in that, include: The first acquisition module is used to acquire the personal information of the screening subject and the prostate-specific antigen test results of the screening subject. The first acquisition module is also used for: The prostate magnetic resonance scan images of the screening subjects are preprocessed, including noise reduction, standardization, and contrast enhancement. An antigen screening module is used to determine the risk value of the antigen test for the screening subject based on the personal information of the screening subject and the prostate-specific antigen test results of the screening subject. The second acquisition module is used to acquire the prostate magnetic resonance scan image of the screening object if the risk value of the antigen test of the screening object is greater than a preset threshold. The graph structure module is used to construct a graph structure based on the prostate-specific antigen (PSA) test results and the prostate magnetic resonance imaging (MRI) scan images of the screening subjects. The graph structure includes two nodes and an edge between the two nodes. The two nodes include an antigen test node and an MRI test node. The node features of the antigen test node include the personal information of the screening subjects and the PSA test results of the screening subjects. The node features of the MRI test node include the prostate MRI scan images of the screening subjects. The features of the edge between the nodes include the consistency between the antigen test and the MRI test. The analysis information determination module is used to process the graph structure based on a graph convolutional network to determine prostate cancer analysis information. The input of the graph convolutional network is the graph structure, and the output of the graph convolutional network is the prostate cancer analysis information. The prostate cancer analysis information includes the presence or absence of prostate cancer, the location, size, shape, density, edge clarity, and whether there are signs of invasion of surrounding tissues or organs, as well as the reliability of antigen testing. A generative adversarial module is used to generate a simulated prostate cancer growth video using a generative adversarial network based on the prostate MRI scan images of the screened subjects and the prostate cancer analysis information. The risk level determination module is used to determine the risk level of prostate cancer based on the prostate cancer simulated growth video, and the risk level determination module is further used for: The risk level of prostate cancer is determined by processing the simulated growth video based on a risk assessment model. This model is a long short-term neural network model, comprising an information change sequence determination layer, a risk information determination layer, and a prostate cancer risk level determination layer. The input to the information change sequence determination layer is time-series data on tumor size, tumor shape, tumor location, tumor density, and tumor edge clarity. The output of the information change sequence determination layer is the time-series data on tumor size, tumor shape, tumor location, and tumor density. The time series data for tumor margin clarity is used as input to the hazard information determination layer. The time series data for tumor size, shape, location, density, and margin clarity are used as inputs. The output of the hazard information determination layer is the hazard level of tumor size, shape, location, density, and margin clarity. The input of the hazard level determination layer is the hazard level of tumor size, shape, location, density, and margin clarity. The output of the hazard level determination layer is the hazard level of prostate cancer.
2. An electronic device, characterized in that, include: processor; Memory; And a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the cancer risk identification system as claimed in claim 1.
3. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the cancer risk assessment system as described in claim 1.