Ovarian cancer risk prediction method and system fusing visual images and clinical features

Abstract

The invention relates to the technical field of ovarian cancer risk prediction, and discloses a visual image and clinical feature fused ovarian cancer risk prediction method and system. The method comprises the steps that visual image data and clinical feature data of a patient are obtained, morphological feature data, textural feature data and hemodynamic feature data are extracted according to the visual image data, and the clinical feature data comprise basic data of the patient and biological information detection data; processing to obtain a morphological anomaly evaluation index, a textural feature anomaly evaluation index, a hemodynamic anomaly evaluation index and the like, and processing to obtain an image anomaly evaluation index according to the hemodynamic anomaly evaluation index, the morphological anomaly evaluation index and the textural feature anomaly evaluation index; according to the biological information detection data, the susceptibility evaluation index and the image anomaly evaluation index, processing to obtain an ovarian cancer risk evaluation index, and processing to obtain an ovarian cancer risk evaluation grade; therefore, the ovarian cancer risk prediction technology fusing visual images and clinical features is realized.

Description

Technical Field

[0001] The present application relates to the technical field of ovarian cancer risk prediction, and in particular, to an ovarian cancer risk prediction method and system that integrates visual images and clinical characteristics. Background Art

[0002] Ovarian cancer is a common malignant tumor of the female reproductive system, with a high mortality rate and hidden early symptoms. Many patients are already in the advanced stage when diagnosed, which seriously threatens women's health. Traditional diagnosis based on a single clinical indicator or image interpretation has limitations. Individual differences and the complexity of the disease lead to frequent misdiagnosis and missed diagnosis. At present, although some studies have tried to use visual imaging data or clinical feature data to predict the risk of ovarian cancer, there are not many technologies that effectively integrate the two. The patent with publication number CN111862079A discloses a high-level serous ovarian cancer recurrence risk prediction system based on imaging omics, and predicts the recurrence risk of ovarian cancer through image processing. The patent with publication number CN118824527A discloses a system for obtaining an ovarian cancer risk index, a sample analysis system and a construction method, and inputs HE4 level, CA125 level and age into a machine learning model to calculate an ovarian cancer risk index used to characterize the probability of a subject suffering from ovarian cancer. When predicting the risk of ovarian cancer, the above two patents only focus on images or clinical features, and lack a method for simultaneously integrating visual images and clinical feature multi-dimensional data. In addition, when processing visual image data, the existing methods often only focus on a single feature, such as morphological features or texture features, and ignore other important information. At the same time, the processing of clinical feature data is not comprehensive enough, and factors such as the patient's susceptibility are not fully considered. There is an urgent need for an ovarian cancer risk prediction method that can simultaneously integrate visual images and clinical feature multi-dimensional data. Summary of the invention

[0003] The purpose of this application is to provide an ovarian cancer risk prediction method and system that integrates visual images and clinical characteristics. By collecting patient examination image data and clinical characteristics such as patient age, genetic medical history, and serum marker levels, a multimodal data set is constructed, and multi-dimensional fusion analysis is performed to achieve the purpose of improving the prediction accuracy of early ovarian cancer risk.

[0004] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0005] The ovarian cancer risk prediction method integrating visual imaging and clinical characteristics includes the following steps:

[0006] Obtaining visual image data and clinical characteristic data of the patient, extracting morphological characteristic data, texture characteristic data and hemodynamic characteristic data based on the visual image data, the clinical characteristic data including basic patient data and bioinformatics detection data;

[0007] Obtain the susceptibility assessment index based on the patient's basic data and bioinformatics test data;

[0008] Obtaining a morphological abnormality assessment index according to the morphological feature data processing, and obtaining a texture feature abnormality assessment index according to the texture feature data processing;

[0009] Obtaining a hemodynamic abnormality assessment index according to the hemodynamic characteristic data processing, and obtaining an image abnormality assessment index according to the hemodynamic abnormality assessment index and the morphological abnormality assessment index and the texture characteristic abnormality assessment index processing;

[0010] The bioinformation detection data, the susceptibility assessment index and the image abnormality assessment index are processed to obtain an ovarian cancer risk assessment index, and the ovarian cancer risk assessment grade is processed to obtain an ovarian cancer risk assessment grade.

[0011] Optionally, in the ovarian cancer risk prediction method integrating visual images and clinical features described in the present application, the visual image data and clinical feature data of the patient are obtained, and morphological feature data, texture feature data and hemodynamic feature data are extracted according to the visual image data, and the clinical feature data includes basic patient data and bioinformation detection data, including:

[0012] The morphological feature data include ovarian volume, circumference and major-minor axis ratio, the texture feature data include contrast, correlation value and entropy value, and the hemodynamic feature data include blood flow velocity, resistance index and pulsatility index;

[0013] The basic patient data include the patient's age and number of births;

[0014] The biological information detection data includes CA125 data, HE4 data and the number of mutated genes. More preferably, it also includes data of one or more tumor markers such as CA199, AFP, CEA, etc.

[0015] Optionally, in the ovarian cancer risk prediction method integrating visual images and clinical features described in the present application, the susceptibility assessment index is obtained by processing the basic data of the patient and the biological information detection data, including:

[0016] The susceptibility assessment index is obtained according to the patient's age, number of births and the number of mutated genes.

[0017] Optionally, in the ovarian cancer risk prediction method integrating visual images and clinical features described in the present application, obtaining a morphological abnormality assessment index according to the morphological feature data processing and obtaining a texture feature abnormality assessment index according to the texture feature data processing include:

[0018] The morphological abnormality assessment index is obtained according to the ovarian volume, circumference and the ratio of the major axis to the minor axis;

[0019] The texture feature abnormality assessment index is obtained according to the contrast, correlation value and entropy value processing.

[0020] Optionally, in the ovarian cancer risk prediction method integrating visual images and clinical features described in the present application, the hemodynamic abnormality assessment index is obtained according to the hemodynamic characteristic data processing, and the image abnormality assessment index is obtained according to the hemodynamic abnormality assessment index and the morphological abnormality assessment index and the texture characteristic abnormality assessment index processing, including:

[0021] Obtaining a hemodynamic abnormality assessment index according to the blood flow velocity, resistance index and pulsatility index;

[0022] The image abnormality assessment index is obtained according to the hemodynamic abnormality assessment index, the morphological abnormality assessment index and the texture feature abnormality assessment index.

[0023] Optionally, in the ovarian cancer risk prediction method integrating visual images and clinical features described in the present application, the step of obtaining an ovarian cancer risk assessment index according to the biological information detection data, the susceptibility assessment index and the image abnormality assessment index, and obtaining an ovarian cancer risk assessment grade, comprises:

[0024] Inputting the CA125 data, HE4 data, susceptibility assessment index and abnormal imaging assessment index into a preset ovarian cancer risk assessment model for processing to obtain an ovarian cancer risk assessment index;

[0025] The ovarian cancer risk assessment index is compared with a preset ovarian cancer risk assessment index threshold, and the range level to which the threshold comparison result belongs is used as the ovarian cancer risk assessment level.

[0026] Optionally, the ovarian cancer risk prediction method integrating visual images and clinical features described in the present application further includes:

[0027] Obtain the test data of ovarian cancer patients before and after treatment, including tumor area before treatment, CA125 data before treatment, HE4 data before treatment, residual tumor area, CA125 data after treatment, and HE4 data after treatment;

[0028] The treatment effect evaluation index is obtained according to the tumor area before treatment, CA125 data before treatment, HE4 data before treatment, residual tumor area, CA125 data after treatment and HE4 data after treatment.

[0029] A second aspect of the present invention provides an ovarian cancer risk prediction system that integrates visual images and clinical characteristics, the system comprising: a memory and a processor, wherein the memory stores a program of an ovarian cancer risk prediction method that integrates visual images and clinical characteristics, and when the program of the ovarian cancer risk prediction method that integrates visual images and clinical characteristics is executed by the processor, the following steps are implemented:

[0030] Obtaining visual image data and clinical characteristic data of the patient, extracting morphological characteristic data, texture characteristic data and hemodynamic characteristic data based on the visual image data, the clinical characteristic data including basic patient data and bioinformatics detection data;

[0031] Obtain the susceptibility assessment index based on the patient's basic data and bioinformatics test data;

[0032] Obtaining a morphological abnormality assessment index according to the morphological feature data processing, and obtaining a texture feature abnormality assessment index according to the texture feature data processing;

[0033] Obtaining a hemodynamic abnormality assessment index according to the hemodynamic characteristic data processing, and obtaining an image abnormality assessment index according to the hemodynamic abnormality assessment index and the morphological abnormality assessment index and the texture characteristic abnormality assessment index processing;

[0034] The bioinformation detection data, the susceptibility assessment index and the image abnormality assessment index are processed to obtain an ovarian cancer risk assessment index, and the ovarian cancer risk assessment grade is processed to obtain an ovarian cancer risk assessment grade.

[0035] Optionally, in the ovarian cancer risk prediction system integrating visual images and clinical features described in the present application, the visual image data and clinical feature data of the patient are obtained, and morphological feature data, texture feature data and hemodynamic feature data are extracted according to the visual image data, and the clinical feature data includes basic patient data and bioinformation detection data, including:

[0036] The morphological feature data include ovarian volume, circumference and major-minor axis ratio, the texture feature data include contrast, correlation value and entropy value, and the hemodynamic feature data include blood flow velocity, resistance index and pulsatility index;

[0037] The basic patient data include the patient's age and number of births;

[0038] The bioinformatics detection data includes CA125 data, HE4 data and the number of mutated genes.

[0039] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0040] The ovarian cancer risk prediction method and system provided in the present application integrates visual images and clinical characteristics. By collecting patient examination image data and clinical characteristics such as patient age, genetic medical history, and serum marker levels, a multimodal data set is constructed, and multi-dimensional fusion analysis is performed to achieve the purpose of improving the prediction accuracy of early ovarian cancer risk.

[0041] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or understood by practicing the embodiments of the present application. The purpose and other advantages of the present application can be realized and obtained by the structures specifically pointed out in the written description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 A flow chart of a method for predicting ovarian cancer risk by integrating visual images and clinical features provided in an embodiment of the present application;

[0044] Figure 2 A flow chart of obtaining a morphological abnormality assessment index and a texture feature abnormality assessment index in a method for predicting ovarian cancer risk by integrating visual images and clinical features provided in an embodiment of the present application;

[0045] Figure 3 A flow chart of obtaining a hemodynamic abnormality assessment index and an imaging abnormality assessment index in a method for predicting ovarian cancer risk by integrating visual images and clinical characteristics provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0047] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0048] Example 1

[0049] like Figure 1 As shown, this embodiment discloses a flow chart of a method for predicting the risk of ovarian cancer by integrating visual images and clinical characteristics. The method for predicting the risk of ovarian cancer by integrating visual images and clinical characteristics is used in a terminal device, such as a computer, a mobile phone terminal, etc. The method for predicting the risk of ovarian cancer by integrating visual images and clinical characteristics comprises the following steps:

[0050] S11. Obtain visual image data and clinical characteristic data of the patient, and extract morphological characteristic data, texture characteristic data and hemodynamic characteristic data according to the visual image data. The clinical characteristic data includes basic data of the patient and bioinformatics detection data;

[0051] S12. Obtaining a susceptibility assessment index based on the patient's basic data and bioinformatics detection data;

[0052] S13, obtaining a morphological abnormality assessment index according to the morphological feature data processing, and obtaining a texture feature abnormality assessment index according to the texture feature data processing;

[0053] S14, obtaining a hemodynamic abnormality assessment index according to the hemodynamic characteristic data, and obtaining an image abnormality assessment index according to the hemodynamic abnormality assessment index and the morphological abnormality assessment index and the texture characteristic abnormality assessment index;

[0054] S15. Obtain an ovarian cancer risk assessment index based on the biological information detection data, the susceptibility assessment index and the image abnormality assessment index, and obtain an ovarian cancer risk assessment grade.

[0055] It should be noted that the present application achieves the purpose of improving the prediction accuracy of early ovarian cancer risk by performing multi-dimensional fusion analysis on the patient's visual image and clinical characteristic data, obtains the patient's visual image data and clinical characteristic data, extracts morphological feature data, texture feature data and hemodynamic feature data based on the visual image data, and the clinical characteristic data includes the patient's basic data and bioinformation detection data, and obtains the susceptibility assessment index, morphological abnormality assessment index, texture feature abnormality assessment index and hemodynamic abnormality assessment index based on the visual image data and clinical characteristic data, so as to realize the assessment of the patient's susceptibility, ovarian morphology, ovarian internal texture and ovarian blood flow, and then obtains the image abnormality assessment index based on the hemodynamic abnormality assessment index, the morphological abnormality assessment index and the texture feature abnormality assessment index, obtains the ovarian cancer risk assessment index based on the bioinformation detection data, the susceptibility assessment index and the image abnormality assessment index, and obtains the ovarian cancer risk assessment level through processing; thereby realizing the technology of ovarian cancer risk prediction by integrating visual images and clinical characteristics.

[0056] Example 2

[0057] The method for predicting ovarian cancer risk by integrating visual images and clinical features comprises the following steps:

[0058] Obtaining visual image data and clinical characteristic data of the patient, extracting morphological characteristic data, texture characteristic data and hemodynamic characteristic data based on the visual image data, the clinical characteristic data including basic patient data and bioinformatics detection data;

[0059] Obtain the susceptibility assessment index based on the patient's basic data and bioinformatics test data;

[0060] Obtaining a morphological abnormality assessment index according to the morphological feature data processing, and obtaining a texture feature abnormality assessment index according to the texture feature data processing;

[0061] Obtaining a hemodynamic abnormality assessment index according to the hemodynamic characteristic data processing, and obtaining an image abnormality assessment index according to the hemodynamic abnormality assessment index and the morphological abnormality assessment index and the texture characteristic abnormality assessment index processing;

[0062] The bioinformation detection data, the susceptibility assessment index and the image abnormality assessment index are processed to obtain an ovarian cancer risk assessment index, and the ovarian cancer risk assessment grade is processed to obtain an ovarian cancer risk assessment grade.

[0063] It should be noted that the present application achieves the purpose of improving the prediction accuracy of early ovarian cancer risk by performing multi-dimensional fusion analysis on the patient's visual image and clinical characteristic data, obtains the patient's visual image data and clinical characteristic data, extracts morphological feature data, texture feature data and hemodynamic feature data based on the visual image data, and the clinical characteristic data includes the patient's basic data and bioinformation detection data, and obtains the susceptibility assessment index, morphological abnormality assessment index, texture feature abnormality assessment index and hemodynamic abnormality assessment index based on the visual image data and clinical characteristic data, so as to realize the assessment of the patient's susceptibility, ovarian morphology, ovarian internal texture and ovarian blood flow, and then obtains the image abnormality assessment index based on the hemodynamic abnormality assessment index, the morphological abnormality assessment index and the texture feature abnormality assessment index, obtains the ovarian cancer risk assessment index based on the bioinformation detection data, the susceptibility assessment index and the image abnormality assessment index, and obtains the ovarian cancer risk assessment level through processing; thereby realizing the technology of ovarian cancer risk prediction by integrating visual images and clinical characteristics.

[0064] According to an embodiment of the present invention, the visual image data and clinical characteristic data of the patient are obtained, and morphological characteristic data, texture characteristic data and hemodynamic characteristic data are extracted according to the visual image data, and the clinical characteristic data includes basic data of the patient and biological information detection data, including:

[0065] The morphological feature data include ovarian volume, circumference and major-minor axis ratio, the texture feature data include contrast, correlation value and entropy value, and the hemodynamic feature data include blood flow velocity, resistance index and pulsatility index;

[0066] The basic patient data include the patient's age and number of births;

[0067] The biological information detection data includes CA125 data, HE4 data and the number of mutated genes. As a specific implementation, it can also include data of one or more tumor markers such as CA199, AFP, CEA, etc.

[0068] It should be noted that the morphological characteristics include ovarian volume, circumference, and ratio of the major axis to the minor axis. Normal ovarian volume, circumference, and ratio of the major axis to the minor axis generally have a certain range. Ovarian cancer often leads to an increase in ovarian volume, circumference, and ratio of the major axis to the minor axis. Therefore, the ovarian morphology can be evaluated based on the ovarian volume, circumference, and ratio of the major axis to the minor axis.

[0069] Texture feature data include contrast, correlation value and entropy value. Contrast is a texture feature parameter. In ovarian ultrasound, CT or MRI images, contrast reflects the grayscale difference between different regions in ovarian tissue. The texture contrast of ultrasound or other images of normal ovarian tissue is relatively low, while the texture contrast of ovarian cancer tissue will increase due to the disorder of cell structure and the complexity of tumor internal components (such as the presence of necrotic foci, calcification, etc.). The correlation value is a texture feature index that measures the linear relationship between pixels in the image. For ovarian images, it reflects the interdependence of the grayscale values ​​of pixels in the ovarian tissue in spatial position. The correlation between pixels in normal ovarian tissue is strong because its cell arrangement and tissue structure are relatively regular, while the correlation between pixels in ovarian cancer tissue is reduced due to the disordered proliferation of cancer cells and the destruction of intercellular connections. The entropy value is an indicator reflecting the randomness of image texture. The entropy value of normal ovarian tissue is low because its texture is relatively regular, while the entropy value of ovarian cancer tissue is usually high due to the diversity and irregularity of its internal structure. Therefore, the texture characteristics of the ovary can be evaluated based on contrast, correlation value and entropy value.

[0070] Hemodynamic characteristic data include blood flow velocity, resistance index and pulsatility index. Resistance index and pulsatility index are commonly used indicators in ultrasound Doppler examination, which are mainly used to evaluate the hemodynamic state of blood vessels. The blood flow velocity of normal ovarian tissue is relatively stable, while the tumor blood vessels of ovarian cancer tissue are rich and the blood flow velocity will increase. The resistance index and pulsatility index of normal ovarian blood vessels are also within a certain range, while the resistance index and pulsatility index of ovarian tumor blood vessels in ovarian cancer patients will be reduced. Therefore, the hemodynamic state of the ovary can be evaluated based on blood flow velocity, resistance index and pulsatility index.

[0071] Bioinformatics detection data includes CA125 data, HE4 data and the number of mutated genes. CA125 (cancer antigen 125) and HE4 (human epididymis protein 4) are obtained through serum marker testing. If both CA125 and HE4 are elevated at the same time, the possibility of ovarian cancer is relatively higher. Mutated genes refer to mutated BRCA1 and BRCA2 genes. Women who carry these gene mutations have a significantly higher risk of ovarian cancer than the general population.

[0072] Example 3

[0073] The method for predicting ovarian cancer risk by integrating visual images and clinical features comprises the following steps:

[0074] Obtaining visual image data and clinical characteristic data of the patient, extracting morphological characteristic data, texture characteristic data and hemodynamic characteristic data based on the visual image data, the clinical characteristic data including basic patient data and bioinformatics detection data;

[0075] Obtain the susceptibility assessment index based on the patient's basic data and bioinformatics test data;

[0076] Obtaining a morphological abnormality assessment index according to the morphological feature data processing, and obtaining a texture feature abnormality assessment index according to the texture feature data processing;

[0077] Obtaining a hemodynamic abnormality assessment index according to the hemodynamic characteristic data processing, and obtaining an image abnormality assessment index according to the hemodynamic abnormality assessment index and the morphological abnormality assessment index and the texture characteristic abnormality assessment index processing;

[0078] The bioinformation detection data, the susceptibility assessment index and the image abnormality assessment index are processed to obtain an ovarian cancer risk assessment index, and the ovarian cancer risk assessment grade is processed to obtain an ovarian cancer risk assessment grade.

[0079] It should be noted that the present application achieves the purpose of improving the prediction accuracy of early ovarian cancer risk by performing multi-dimensional fusion analysis on the patient's visual image and clinical characteristic data, obtains the patient's visual image data and clinical characteristic data, extracts morphological feature data, texture feature data and hemodynamic feature data based on the visual image data, and the clinical characteristic data includes the patient's basic data and bioinformation detection data, and obtains the susceptibility assessment index, morphological abnormality assessment index, texture feature abnormality assessment index and hemodynamic abnormality assessment index based on the visual image data and clinical characteristic data, so as to realize the assessment of the patient's susceptibility, ovarian morphology, ovarian internal texture and ovarian blood flow, and then obtains the image abnormality assessment index based on the hemodynamic abnormality assessment index, the morphological abnormality assessment index and the texture feature abnormality assessment index, obtains the ovarian cancer risk assessment index based on the bioinformation detection data, the susceptibility assessment index and the image abnormality assessment index, and obtains the ovarian cancer risk assessment level through processing; thereby realizing the technology of ovarian cancer risk prediction by integrating visual images and clinical characteristics.

[0080] According to an embodiment of the present invention, the visual image data and clinical characteristic data of the patient are obtained, and morphological characteristic data, texture characteristic data and hemodynamic characteristic data are extracted according to the visual image data, and the clinical characteristic data includes basic data of the patient and biological information detection data, including:

[0081] The morphological feature data include ovarian volume, circumference and major-minor axis ratio, the texture feature data include contrast, correlation value and entropy value, and the hemodynamic feature data include blood flow velocity, resistance index and pulsatility index;

[0082] The basic data of the patient include the patient's age and number of births;

[0083] The biological information detection data includes CA125 data, HE4 data and the number of mutated genes. As a specific implementation, it can also include data of one or more tumor markers such as CA199, AFP, CEA, etc.

[0084] It should be noted that the morphological characteristics include ovarian volume, circumference, and ratio of the major axis to the minor axis. Normal ovarian volume, circumference, and ratio of the major axis to the minor axis generally have a certain range. Ovarian cancer often leads to an increase in ovarian volume, circumference, and ratio of the major axis to the minor axis. Therefore, the ovarian morphology can be evaluated based on the ovarian volume, circumference, and ratio of the major axis to the minor axis.

[0085] Texture feature data include contrast, correlation value and entropy value. Contrast is a texture feature parameter. In ovarian ultrasound, CT or MRI images, contrast reflects the grayscale difference between different regions in ovarian tissue. The texture contrast of ultrasound or other images of normal ovarian tissue is relatively low, while the texture contrast of ovarian cancer tissue will increase due to the disorder of cell structure and the complexity of tumor internal components (such as the presence of necrotic foci, calcification, etc.). The correlation value is a texture feature index that measures the linear relationship between pixels in the image. For ovarian images, it reflects the interdependence of the grayscale values ​​of pixels in the ovarian tissue in spatial position. The correlation between pixels in normal ovarian tissue is strong because its cell arrangement and tissue structure are relatively regular, while the correlation between pixels in ovarian cancer tissue is reduced due to the disordered proliferation of cancer cells and the destruction of intercellular connections. The entropy value is an indicator reflecting the randomness of image texture. The entropy value of normal ovarian tissue is low because its texture is relatively regular, while the entropy value of ovarian cancer tissue is usually high due to the diversity and irregularity of its internal structure. Therefore, the texture characteristics of the ovary can be evaluated based on contrast, correlation value and entropy value.

[0086] Hemodynamic characteristic data include blood flow velocity, resistance index and pulsatility index. Resistance index and pulsatility index are commonly used indicators in ultrasound Doppler examination, which are mainly used to evaluate the hemodynamic state of blood vessels. The blood flow velocity of normal ovarian tissue is relatively stable, while the tumor blood vessels of ovarian cancer tissue are rich and the blood flow velocity will increase. The resistance index and pulsatility index of normal ovarian blood vessels are also within a certain range, while the resistance index and pulsatility index of ovarian tumor blood vessels in ovarian cancer patients will be reduced. Therefore, the hemodynamic state of the ovary can be evaluated based on blood flow velocity, resistance index and pulsatility index.

[0087] Bioinformatics detection data includes CA125 data, HE4 data and the number of mutated genes. CA125 (cancer antigen 125) and HE4 (human epididymis protein 4) are obtained through serum marker testing. If both CA125 and HE4 are elevated at the same time, the possibility of ovarian cancer is relatively higher. Mutated genes refer to mutated BRCA1 and BRCA2 genes. Women who carry these gene mutations have a significantly higher risk of ovarian cancer than the general population.

[0088] According to an embodiment of the present invention, the step of obtaining a susceptibility assessment index based on the patient's basic data and biological information detection data includes:

[0089] The susceptibility assessment index is obtained according to the patient's age, number of births and the number of mutated genes.

[0090] It should be noted that women over 50 who have not given birth have a slightly higher risk of ovarian cancer than young women who have given birth. If a direct relative in the family has ovarian cancer, breast cancer or other related cancers, the possibility of carrying BRCA1 and BRCA2 gene mutations increases, significantly increasing the individual's risk of disease.

[0091] The calculation formula of the susceptibility assessment index is:

[0092] r c =α1L s -α2L y +βY e ;

[0093] Among them, r c is the susceptibility assessment index, L s , L y and Y e are the patient's age, number of births and number of mutated genes respectively; α1, α2 and β are preset characteristic coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database).

[0094] like Figure 2 As shown, this embodiment discloses a flow chart of obtaining a morphological abnormality assessment index and a texture feature abnormality assessment index of an ovarian cancer risk prediction method integrating visual images and clinical features. According to an embodiment of the present invention, obtaining a morphological abnormality assessment index according to the morphological feature data processing and obtaining a texture feature abnormality assessment index according to the texture feature data processing include:

[0095] S21, obtaining a morphological abnormality assessment index according to the ovarian volume, circumference and major-minor axis ratio;

[0096] S22. Obtain a texture feature abnormality assessment index according to the contrast, correlation value and entropy value.

[0097] It should be noted that the morphological abnormality assessment index is obtained by processing the morphological feature data, and the texture feature abnormality assessment index is obtained by processing the texture feature data;

[0098] The calculation formula of the morphological abnormality assessment index is:

[0099]

[0100] in, is the morphological abnormality assessment index, V a , L e and B l are ovarian volume, circumference and major-minor axis ratio, respectively, and χ1, χ2 and χ3 are preset characteristic coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database);

[0101] The calculation formula of the texture feature abnormality evaluation index is:

[0102]

[0103] in, is the texture feature abnormality assessment index, C n , C r and E p are contrast, correlation value and entropy value respectively, and δ1, δ2 and δ3 are preset characteristic coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database).

[0104] like Figure 3 As shown, this embodiment discloses a flow chart of obtaining a hemodynamic abnormality assessment index and an image abnormality assessment index of an ovarian cancer risk prediction method integrating visual images and clinical characteristics. According to an embodiment of the present invention, the hemodynamic abnormality assessment index is obtained according to the hemodynamic feature data processing, and the image abnormality assessment index is obtained according to the hemodynamic abnormality assessment index and the morphological abnormality assessment index and the texture feature abnormality assessment index processing, including:

[0105] S31, obtaining a hemodynamic abnormality assessment index according to the blood flow velocity, resistance index and pulsatility index;

[0106] S32, obtaining an image abnormality assessment index according to the hemodynamic abnormality assessment index, the morphological abnormality assessment index, and the texture feature abnormality assessment index.

[0107] It should be noted that the hemodynamic abnormality assessment index is obtained by processing the hemodynamic characteristic data, and the imaging abnormality assessment index is obtained by processing the hemodynamic abnormality assessment index as well as the morphological abnormality assessment index and the texture characteristic abnormality assessment index;

[0108] The calculation formula of the hemodynamic abnormality assessment index is:

[0109]

[0110] in, is the hemodynamic abnormality assessment index, v b , R i and P i are blood flow velocity, resistance index and pulsatility index respectively, and η1, η2 and η3 are preset weight coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database);

[0111] The calculation formula of the image abnormality assessment index is:

[0112]

[0113] in, is the imaging abnormality assessment index, and are the hemodynamic abnormality assessment index, morphological abnormality assessment index and texture feature abnormality assessment index respectively, and γ1, γ2 and γ3 are preset weight coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database).

[0114] According to an embodiment of the present invention, the step of obtaining an ovarian cancer risk assessment index based on the biological information detection data, the susceptibility assessment index and the image abnormality assessment index, and obtaining an ovarian cancer risk assessment level, includes:

[0115] Inputting the CA125 data, HE4 data, susceptibility assessment index and abnormal imaging assessment index into a preset ovarian cancer risk assessment model for processing to obtain an ovarian cancer risk assessment index;

[0116] The ovarian cancer risk assessment index is compared with a preset ovarian cancer risk assessment index threshold, and the range level to which the threshold comparison result belongs is used as the ovarian cancer risk assessment level.

[0117] It should be noted that CA125 data, HE4 data, susceptibility assessment index and imaging abnormality assessment index are input into a preset ovarian cancer risk assessment model for processing to obtain an ovarian cancer risk assessment index;

[0118] The calculation formula of the ovarian cancer risk assessment model is:

[0119]

[0120] Among them, W OV is the ovarian cancer risk assessment index, C a and H e They are CA125 data and HE4 data, rc and are the susceptibility assessment index and the imaging abnormality assessment index, respectively; ε1 and ε2 are the preset weight coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database).

[0121] According to an embodiment of the present invention, it also includes:

[0122] Obtain the test data of ovarian cancer patients before and after treatment, including tumor area before treatment, CA125 data before treatment, HE4 data before treatment, residual tumor area, CA125 data after treatment, and HE4 data after treatment;

[0123] The treatment effect evaluation index is obtained according to the tumor area before treatment, CA125 data before treatment, HE4 data before treatment, residual tumor area, CA125 data after treatment and HE4 data after treatment.

[0124] It should be noted that the treatment effect evaluation was performed based on the tumor area and serum marker test data of ovarian cancer patients before and after treatment;

[0125] The calculation formula of the treatment effect evaluation index is:

[0126]

[0127] Among them, r d is the treatment effect evaluation index, S z , S z ′、C a , C a ′、H e and H e ′ are respectively the tumor area before treatment, the residual tumor area, the CA125 data before treatment, the CA125 data after treatment, the HE4 data before treatment and the HE4 data after treatment, λ1, λ2 and λ3 are the preset characteristic coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database).

[0128] The present invention also discloses an ovarian cancer risk prediction system integrating visual images and clinical characteristics, comprising a memory and a processor, wherein the memory stores an ovarian cancer risk prediction method program integrating visual images and clinical characteristics, and when the ovarian cancer risk prediction method program integrating visual images and clinical characteristics is executed by the processor, the following steps are implemented:

[0129] Obtaining visual image data and clinical characteristic data of the patient, extracting morphological characteristic data, texture characteristic data and hemodynamic characteristic data based on the visual image data, the clinical characteristic data including basic patient data and bioinformatics detection data;

[0130] Obtain the susceptibility assessment index based on the patient's basic data and bioinformatics test data;

[0131] Obtaining a morphological abnormality assessment index according to the morphological feature data processing, and obtaining a texture feature abnormality assessment index according to the texture feature data processing;

[0132] Obtaining a hemodynamic abnormality assessment index according to the hemodynamic characteristic data processing, and obtaining an image abnormality assessment index according to the hemodynamic abnormality assessment index and the morphological abnormality assessment index and the texture characteristic abnormality assessment index processing;

[0133] The bioinformation detection data, the susceptibility assessment index and the image abnormality assessment index are processed to obtain an ovarian cancer risk assessment index, and the ovarian cancer risk assessment grade is processed to obtain an ovarian cancer risk assessment grade.

[0134] It should be noted that the present application achieves the purpose of improving the prediction accuracy of early ovarian cancer risk by performing multi-dimensional fusion analysis on the patient's visual image and clinical characteristic data, obtains the patient's visual image data and clinical characteristic data, extracts morphological feature data, texture feature data and hemodynamic feature data based on the visual image data, and the clinical characteristic data includes the patient's basic data and bioinformation detection data, and obtains the susceptibility assessment index, morphological abnormality assessment index, texture feature abnormality assessment index and hemodynamic abnormality assessment index based on the visual image data and clinical characteristic data, so as to realize the assessment of the patient's susceptibility, ovarian morphology, ovarian internal texture and ovarian blood flow, and then obtains the image abnormality assessment index based on the hemodynamic abnormality assessment index, the morphological abnormality assessment index and the texture feature abnormality assessment index, obtains the ovarian cancer risk assessment index based on the bioinformation detection data, the susceptibility assessment index and the image abnormality assessment index, and obtains the ovarian cancer risk assessment level through processing; thereby realizing the technology of ovarian cancer risk prediction by integrating visual images and clinical characteristics.

[0135] According to an embodiment of the present invention, the visual image data and clinical characteristic data of the patient are obtained, and morphological characteristic data, texture characteristic data and hemodynamic characteristic data are extracted according to the visual image data, and the clinical characteristic data includes basic data of the patient and biological information detection data, including:

[0136] The morphological feature data include ovarian volume, circumference and major-minor axis ratio, the texture feature data include contrast, correlation value and entropy value, and the hemodynamic feature data include blood flow velocity, resistance index and pulsatility index;

[0137] The basic patient data include the patient's age and number of births;

[0138] The bioinformatics detection data includes CA125 data, HE4 data and the number of mutated genes.

[0139] It should be noted that the morphological characteristics include ovarian volume, circumference, and ratio of the major axis to the minor axis. Normal ovarian volume, circumference, and ratio of the major axis to the minor axis generally have a certain range. Ovarian cancer often leads to an increase in ovarian volume, circumference, and ratio of the major axis to the minor axis. Therefore, the ovarian morphology can be evaluated based on the ovarian volume, circumference, and ratio of the major axis to the minor axis.

[0140] Texture feature data include contrast, correlation value and entropy value. Contrast is a texture feature parameter. In ovarian ultrasound, CT or MRI images, contrast reflects the grayscale difference between different regions in ovarian tissue. The texture contrast of ultrasound or other images of normal ovarian tissue is relatively low, while the texture contrast of ovarian cancer tissue will increase due to the disorder of cell structure and the complexity of tumor internal components (such as the presence of necrotic foci, calcification, etc.). The correlation value is a texture feature index that measures the linear relationship between pixels in the image. For ovarian images, it reflects the interdependence of the grayscale values ​​of pixels in the ovarian tissue in spatial position. The correlation between pixels in normal ovarian tissue is strong because its cell arrangement and tissue structure are relatively regular, while the correlation between pixels in ovarian cancer tissue is reduced due to the disordered proliferation of cancer cells and the destruction of intercellular connections. The entropy value is an indicator reflecting the randomness of image texture. The entropy value of normal ovarian tissue is low because its texture is relatively regular, while the entropy value of ovarian cancer tissue is usually high due to the diversity and irregularity of its internal structure. Therefore, the texture characteristics of the ovary can be evaluated based on contrast, correlation value and entropy value.

[0141] Hemodynamic characteristic data include blood flow velocity, resistance index and pulsatility index. Resistance index and pulsatility index are commonly used indicators in ultrasound Doppler examination, which are mainly used to evaluate the hemodynamic state of blood vessels. The blood flow velocity of normal ovarian tissue is relatively stable, while the tumor blood vessels of ovarian cancer tissue are rich and the blood flow velocity will increase. The resistance index and pulsatility index of normal ovarian blood vessels are also within a certain range, while the resistance index and pulsatility index of ovarian tumor blood vessels in ovarian cancer patients will be reduced. Therefore, the hemodynamic state of the ovary can be evaluated based on blood flow velocity, resistance index and pulsatility index.

[0142] Bioinformatics detection data includes CA125 data, HE4 data and the number of mutated genes. CA125 (cancer antigen 125) and HE4 (human epididymis protein 4) are obtained through serum marker testing. If both CA125 and HE4 are elevated at the same time, the possibility of ovarian cancer is relatively higher. Mutated genes refer to mutated BRCA1 and BRCA2 genes. Women who carry these gene mutations have a significantly higher risk of ovarian cancer than the general population.

[0143] According to an embodiment of the present invention, the step of obtaining a susceptibility assessment index based on the patient's basic data and biological information detection data includes:

[0144] The susceptibility assessment index is obtained according to the patient's age, number of births and the number of mutated genes.

[0145] It should be noted that women over 50 who have not given birth have a slightly higher risk of ovarian cancer than young women who have given birth. If a direct relative in the family has ovarian cancer, breast cancer or other related cancers, the possibility of carrying BRCA1 and BRCA2 gene mutations increases, significantly increasing the individual's risk of disease.

[0146] The calculation formula of the susceptibility assessment index is:

[0147] r c =α1L s -α2L y +βY e ;

[0148] Among them, r c is the susceptibility assessment index, L s , L y and Y e are the patient's age, number of births and number of mutated genes respectively; α1, α2 and β are preset characteristic coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database).

[0149] According to an embodiment of the present invention, the step of obtaining a morphological abnormality assessment index according to the morphological feature data processing and obtaining a texture feature abnormality assessment index according to the texture feature data processing includes:

[0150] The morphological abnormality assessment index is obtained according to the ovarian volume, circumference and the ratio of the major axis to the minor axis;

[0151] The texture feature abnormality assessment index is obtained according to the contrast, correlation value and entropy value processing.

[0152] It should be noted that the morphological abnormality assessment index is obtained by processing the morphological feature data, and the texture feature abnormality assessment index is obtained by processing the texture feature data;

[0153] The calculation formula of the morphological abnormality assessment index is:

[0154]

[0155] in, is the morphological abnormality assessment index, V a , L e and B lare ovarian volume, circumference and major-minor axis ratio, respectively, and χ1, χ2 and χ3 are preset characteristic coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database);

[0156] The calculation formula of the texture feature abnormality evaluation index is:

[0157]

[0158] in, is the texture feature abnormality assessment index, C n , C r and E p are contrast, correlation value and entropy value respectively, and δ1, δ2 and δ3 are preset characteristic coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database).

[0159] According to an embodiment of the present invention, the method of obtaining a hemodynamic abnormality assessment index according to the hemodynamic characteristic data processing, and obtaining an image abnormality assessment index according to the hemodynamic abnormality assessment index and the morphological abnormality assessment index and the texture characteristic abnormality assessment index processing includes:

[0160] Obtaining a hemodynamic abnormality assessment index according to the blood flow velocity, resistance index and pulsatility index;

[0161] The image abnormality assessment index is obtained according to the hemodynamic abnormality assessment index, the morphological abnormality assessment index and the texture feature abnormality assessment index.

[0162] It should be noted that the hemodynamic abnormality assessment index is obtained by processing the hemodynamic characteristic data, and the imaging abnormality assessment index is obtained by processing the hemodynamic abnormality assessment index as well as the morphological abnormality assessment index and the texture characteristic abnormality assessment index;

[0163] The calculation formula of the hemodynamic abnormality assessment index is:

[0164]

[0165] in, is the hemodynamic abnormality assessment index, v b , R i and P i are blood flow velocity, resistance index and pulsatility index respectively, and η1, η2 and η3 are preset weight coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database);

[0166] The calculation formula of the image abnormality assessment index is:

[0167]

[0168] in, is the imaging abnormality assessment index, and are the hemodynamic abnormality assessment index, morphological abnormality assessment index and texture feature abnormality assessment index respectively, and γ1, γ2 and γ3 are preset weight coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database).

[0169] According to an embodiment of the present invention, the step of obtaining an ovarian cancer risk assessment index based on the biological information detection data, the susceptibility assessment index and the image abnormality assessment index, and obtaining an ovarian cancer risk assessment level, includes:

[0170] Inputting the CA125 data, HE4 data, susceptibility assessment index and abnormal imaging assessment index into a preset ovarian cancer risk assessment model for processing to obtain an ovarian cancer risk assessment index;

[0171] The ovarian cancer risk assessment index is compared with a preset ovarian cancer risk assessment index threshold, and the range level to which the threshold comparison result belongs is used as the ovarian cancer risk assessment level.

[0172] It should be noted that CA125 data, HE4 data, susceptibility assessment index and imaging abnormality assessment index are input into a preset ovarian cancer risk assessment model for processing to obtain an ovarian cancer risk assessment index;

[0173] The calculation formula of the ovarian cancer risk assessment model is:

[0174]

[0175] Among them, W OV is the ovarian cancer risk assessment index, C a and H e They are CA125 data and HE4 data, r c and are the susceptibility assessment index and the imaging abnormality assessment index, respectively; ε1 and ε2 are the preset weight coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database).

[0176] According to an embodiment of the present invention, it also includes:

[0177] Obtain the test data of ovarian cancer patients before and after treatment, including tumor area before treatment, CA125 data before treatment, HE4 data before treatment, residual tumor area, CA125 data after treatment, and HE4 data after treatment;

[0178] The treatment effect evaluation index is obtained according to the tumor area before treatment, CA125 data before treatment, HE4 data before treatment, residual tumor area, CA125 data after treatment and HE4 data after treatment.

[0179] It should be noted that the treatment effect evaluation was performed based on the tumor area and serum marker test data of ovarian cancer patients before and after treatment;

[0180] The calculation formula of the treatment effect evaluation index is:

[0181]

[0182] Among them, r d is the treatment effect evaluation index, S z , S z ′、C a , C a ′、H e and H e ′ are respectively the tumor area before treatment, the residual tumor area, the CA125 data before treatment, the CA125 data after treatment, the HE4 data before treatment and the HE4 data after treatment, λ1, λ2 and λ3 are the preset characteristic coefficients (which can be obtained by querying the preset ovarian cancer risk detection platform database).

[0183] The present invention discloses an ovarian cancer risk prediction method and system that integrates visual images and clinical characteristics. By collecting patient examination image data and clinical characteristics such as patient age, genetic medical history, and serum marker levels, a multimodal data set is constructed, and multi-dimensional fusion analysis is performed to achieve the purpose of improving the prediction accuracy of early ovarian cancer risk.

[0184] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0185] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0186] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

Claims

1. A method for predicting ovarian cancer risk by integrating visual images and clinical features, characterized in that: The following steps are involved: Obtaining visual image data and clinical characteristic data of the patient, extracting morphological characteristic data, texture characteristic data and hemodynamic characteristic data based on the visual image data, the clinical characteristic data including basic patient data and bioinformatics detection data; Obtain the susceptibility assessment index based on the patient's basic data and bioinformatics test data; Obtaining a morphological abnormality assessment index according to the morphological feature data processing, and obtaining a texture feature abnormality assessment index according to the texture feature data processing; Obtaining a hemodynamic abnormality assessment index according to the hemodynamic characteristic data processing, and obtaining an image abnormality assessment index according to the hemodynamic abnormality assessment index and the morphological abnormality assessment index and the texture characteristic abnormality assessment index processing; The bioinformation detection data, the susceptibility assessment index and the image abnormality assessment index are processed to obtain an ovarian cancer risk assessment index, and the ovarian cancer risk assessment grade is processed to obtain an ovarian cancer risk assessment grade.

2. The method for predicting ovarian cancer risk by integrating visual images and clinical features according to claim 1, characterized in that: The method of obtaining visual image data and clinical characteristic data of the patient, extracting morphological characteristic data, texture characteristic data and hemodynamic characteristic data according to the visual image data, wherein the clinical characteristic data includes basic data of the patient and bio-information detection data, includes: The morphological feature data include ovarian volume, circumference and major-minor axis ratio, the texture feature data include contrast, correlation value and entropy value, and the hemodynamic feature data include blood flow velocity, resistance index and pulsatility index; The basic data of the patient include the patient's age and number of births; The bioinformatics detection data includes CA125 data, HE4 data and the number of mutated genes.

3. The method for predicting ovarian cancer risk by integrating visual images and clinical features according to claim 2, characterized in that: The method of obtaining the susceptibility assessment index based on the patient's basic data and biological information detection data includes: The susceptibility assessment index is obtained according to the patient's age, number of births and the number of mutated genes.

4. The method for predicting ovarian cancer risk by integrating visual images and clinical features according to claim 3, characterized in that: The step of obtaining a morphological abnormality assessment index according to the morphological feature data processing and obtaining a texture feature abnormality assessment index according to the texture feature data processing comprises: The morphological abnormality assessment index is obtained according to the ovarian volume, circumference and the ratio of the major axis to the minor axis; The texture feature abnormality assessment index is obtained according to the contrast, correlation value and entropy value processing.

5. The method for predicting ovarian cancer risk by integrating visual images and clinical features according to claim 4, characterized in that: The method of obtaining a hemodynamic abnormality assessment index according to the hemodynamic characteristic data processing, and obtaining an image abnormality assessment index according to the hemodynamic abnormality assessment index and the morphological abnormality assessment index and the texture characteristic abnormality assessment index processing includes: Obtaining a hemodynamic abnormality assessment index according to the blood flow velocity, resistance index and pulsatility index; The image abnormality assessment index is obtained according to the hemodynamic abnormality assessment index, the morphological abnormality assessment index and the texture feature abnormality assessment index.

6. The method for predicting ovarian cancer risk by integrating visual images and clinical features according to claim 5, characterized in that: The step of obtaining an ovarian cancer risk assessment index based on the biological information detection data, the susceptibility assessment index and the image abnormality assessment index, and obtaining an ovarian cancer risk assessment grade, includes: Inputting the CA125 data, HE4 data, susceptibility assessment index and abnormal imaging assessment index into a preset ovarian cancer risk assessment model for processing to obtain an ovarian cancer risk assessment index; The ovarian cancer risk assessment index is compared with a preset ovarian cancer risk assessment index threshold, and the range level to which the threshold comparison result belongs is used as the ovarian cancer risk assessment level.

7. The method for predicting ovarian cancer risk by integrating visual images and clinical features according to claim 6, characterized in that: Also includes: Obtain the test data of ovarian cancer patients before and after treatment, including tumor area before treatment, CA125 data before treatment, HE4 data before treatment, residual tumor area, CA125 data after treatment, and HE4 data after treatment; The treatment effect evaluation index is obtained according to the tumor area before treatment, CA125 data before treatment, HE4 data before treatment, residual tumor area, CA125 data after treatment and HE4 data after treatment.

8. An ovarian cancer risk prediction system integrating visual images and clinical features, characterized in that: The method comprises a memory and a processor, wherein the memory stores a program of a method for predicting the risk of ovarian cancer by integrating visual images and clinical characteristics, and when the method for predicting the risk of ovarian cancer by integrating visual images and clinical characteristics is executed by the processor, the following steps are implemented: Obtaining visual image data and clinical characteristic data of the patient, extracting morphological characteristic data, texture characteristic data and hemodynamic characteristic data based on the visual image data, the clinical characteristic data including basic patient data and bioinformatics detection data; Obtain the susceptibility assessment index based on the patient's basic data and bioinformatics test data; Obtaining a morphological abnormality assessment index according to the morphological feature data processing, and obtaining a texture feature abnormality assessment index according to the texture feature data processing; Obtaining a hemodynamic abnormality assessment index according to the hemodynamic characteristic data processing, and obtaining an image abnormality assessment index according to the hemodynamic abnormality assessment index and the morphological abnormality assessment index and the texture characteristic abnormality assessment index processing; The bioinformation detection data, the susceptibility assessment index and the image abnormality assessment index are processed to obtain an ovarian cancer risk assessment index, and the ovarian cancer risk assessment grade is processed to obtain an ovarian cancer risk assessment grade.

9. The ovarian cancer risk prediction system integrating visual images and clinical features according to claim 8, characterized in that: The method of obtaining visual image data and clinical characteristic data of the patient, extracting morphological characteristic data, texture characteristic data and hemodynamic characteristic data according to the visual image data, wherein the clinical characteristic data includes basic data of the patient and bio-information detection data, includes: The morphological feature data include ovarian volume, circumference and major-minor axis ratio, the texture feature data include contrast, correlation value and entropy value, and the hemodynamic feature data include blood flow velocity, resistance index and pulsatility index; The basic data of the patient include the patient's age and number of births; The bioinformatics detection data includes CA125 data, HE4 data and the number of mutated genes.

10. The ovarian cancer risk prediction system integrating visual images and clinical features according to claim 9, characterized in that: The method of obtaining the susceptibility assessment index based on the patient's basic data and biological information detection data includes: The susceptibility assessment index is obtained according to the patient's age, number of births and the number of mutated genes.

Images