A method and device for analyzing pathological characteristics of breast ductal carcinoma in situ

By screening important pathological characteristics and classification models of breast ductal carcinoma in situ, the problem of the inability to analyze nuclear magnetic resonance imaging characteristics in the prior art is solved, and accurate prediction and evaluation of cancer progress is achieved.

CN117975181BActive Publication Date: 2024-11-12SHENZHEN PEOPLES HOSPITAL +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311705944.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-12-12
Publication Date
2024-11-12
Estimated Expiration
2043-12-12

AI Technical Summary

Technical Problem

The prior art cannot effectively analyze the MRI imaging characteristics of breast ductal carcinoma in situ, resulting in the inability to accurately predict cancer progression and metastasis, and lack of MRI imaging characteristics that are highly related to ductal carcinoma in situ.

Method used

By obtaining the sample pathological feature set and sample label of sample catheter carcinoma in situ, important pathological features and their importance rankings are determined, and combined with classification models, the optimal pathological feature set is screened to improve prediction accuracy.

Benefits of technology

MRI features are provided that are highly correlated with the progression of catheter carcinoma in situ, helping healthcare workers to quickly and accurately evaluate patients' condition and avoid excessive or mistreatment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117975181B_ABST
    Figure CN117975181B_ABST
Patent Text Reader

Abstract

The present application provides a method and device for analyzing the pathological characteristics of breast ductal carcinoma in situ. It includes: obtaining a sample pathological feature set and a sample label of a sample ductal carcinoma in situ breast; determining at least two important pathological features and the importance ranking corresponding to each important pathological feature based on the sample pathological feature set and the sample label of the sample ductal carcinoma in situ breast; adding important pathological features to the classification model in sequence according to the importance ranking, and calculating the accuracy of the classification model; determining the optimal pathological feature set according to the accuracy and the number of important pathological features added to the classification model. By analyzing the importance of the features to screen out the important pathological features, and analyzing the number of features and the prediction accuracy of the corresponding classification model to screen out the optimal pathological features, it is possible to provide magnetic resonance imaging features that are highly correlated with the progression of ductal carcinoma in situ, which helps medical staff to quickly and accurately assess the patient's condition.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and device for analyzing pathological characteristics of breast ductal carcinoma in situ. Background Art

[0002] Ductal carcinoma in situ (DCIS) refers to abnormal proliferation of epithelial cells in the breast ducts, but does not extend beyond the surrounding basement membrane. The Chinese Anti-Cancer Association's Guidelines and Norms for the Diagnosis and Treatment of Breast Cancer describe DCIS as "a precursor lesion of invasive breast cancer, a type of non-systemic localized intraductal lesion". In 2003, the WHO officially separated DCIS, including DCIS with microinvasion, from breast cancer and classified it as a precancerous lesion, called ductal intraepithelial neoplasia. The American Comprehensive Cancer Network's Breast Cancer Clinical Practice Guidelines have established a separate diagnosis and treatment standard for DCIS that is different from invasive breast cancer. In today's society, people are under great pressure from life and work, and they get married and have children later, while the age of onset of breast cancer is getting younger and younger. Many women of childbearing age have a stronger need for breast preservation and pay more attention to the beauty of breast appearance. Therefore, how to correctly understand DCIS and distinguish it from invasive cancer, avoid overtreatment and incorrect chemotherapy, and improve the quality of life of patients, has practical and important clinical significance.

[0003] The role of combined imaging in invasive breast cancer has been recognized, but its application in DCIS has not received enough attention. On the one hand, due to the confusion in the understanding of DCIS in the past, it was simply attributed to malignant lesions together with invasive breast cancer, and the imaging manifestations of DCIS were not analyzed separately and systematically. On the other hand, due to the differences in the pathology of DCIS in the past, the pathological types that can reflect the local recurrence and progression of DCIS to invasive cancer are not unified, resulting in a lack of pathological comparison of imaging manifestations. With the unification of pathological understanding and the clinical emphasis on DCIS as a separate group of diseases, imaging has begun to play an increasingly important role in DCIS. In traditional mammography, DCIS may show signs of dense calcification, and radial scars may appear as star-shaped changes; in breast color Doppler ultrasound examination, intraductal papillary lesions may show ductal dilatation and abnormal intraductal sound and image, and DCIS may detect solid light masses, but it is not easy to find carcinoma in situ that only shows calcification; with the development of magnetic resonance imaging (MRI) technology, the application of breast MRI has become more and more mature, especially the application of breast dynamic enhancement technology and functional imaging, which enables MRI to provide not only morphological manifestations of breast lesions, but also vascular information and molecular microscopic activity of lesions. However, there are currently few MRI studies on DCIS, and no relevant research can summarize the MRI imaging features that are highly correlated with the infiltration and metastasis of DCIS lesions. Summary of the invention

[0004] In view of the above-mentioned problems, the present application is proposed to provide a method and device for analyzing pathological characteristics of breast ductal carcinoma in situ that overcomes the problems or at least partially solves the problems, including:

[0005] A method for analyzing pathological characteristics of breast ductal carcinoma in situ, comprising:

[0006] Obtaining a sample pathological feature set and a sample label of a sample ductal carcinoma in situ breast; wherein the sample pathological feature set includes at least three nuclear magnetic resonance imaging features; and the sample label includes the presence of cancer progression and the absence of cancer progression;

[0007] Determine at least two important pathological features and the importance ranking corresponding to each important pathological feature according to the sample pathological feature set and the sample label of the sample ductal carcinoma in situ breast;

[0008] According to the importance ranking, the important pathological features are added to the classification model in sequence, and the accuracy of the classification model is calculated;

[0009] An optimal pathology feature set is determined based on the accuracy and the number of important pathology features added to the classification model.

[0010] Preferably, the step of determining at least two important pathological features and the importance ranking corresponding to each important pathological feature based on the sample pathological feature set and sample label of the sample ductal carcinoma in situ breast comprises:

[0011] Numerically representing the sample pathological characteristics and sample labels of the sample ductal carcinoma in situ breast;

[0012] According to the numerically expressed pathological characteristics of the samples, determining the characteristic correlation coefficient between each pathological characteristic of the samples and the pathological characteristics of the other samples;

[0013] According to the numerically represented sample pathological features and sample labels, determine the label correlation coefficient between each sample pathological feature and the sample label;

[0014] When the feature correlation coefficient between any two sample pathological features is greater than a first threshold, setting the sample pathological feature with a lower label correlation coefficient as a redundant pathological feature;

[0015] Setting all sample pathological features except the redundant pathological features as candidate pathological features;

[0016] Setting a preset number of the candidate pathological features with the highest label correlation coefficients as important pathological features;

[0017] According to the label correlation coefficient, the importance ranking corresponding to each important pathological feature is determined.

[0018] Preferably, the step of determining the characteristic correlation coefficient between each pathological feature of the sample and the pathological features of the remaining samples based on the numerically represented pathological features of the sample comprises:

[0019] The characteristic correlation coefficient of each sample pathological feature and the pathological features of other samples was calculated according to the following formula:

[0020] ;

[0021] in, is the feature correlation coefficient between feature x and feature y, is the value of the feature x of the i-th sample, is the value of the feature y of the i-th sample, is the average value of feature x of all samples, is the average value of feature y of all samples, and n is the number of samples.

[0022] Preferably, the step of determining the label correlation coefficient between each sample pathological feature and the sample label based on the numerically represented sample pathological features and the sample label comprises:

[0023] According to the sample pathological features and sample labels after numerical representation, the rank of each sample pathological feature of each sample ductal carcinoma in situ breast in all samples is determined, as well as the rank of each sample ductal carcinoma in situ breast sample label in all samples;

[0024] The label correlation coefficient between each sample pathological feature and the sample label is calculated according to the following formula:

[0025] ;

[0026] in, is the label correlation coefficient between feature x and label z, is the rank of the feature x of the i-th sample, is the rank of the label z of the ith sample, and n is the number of samples.

[0027] Preferably, the step of determining the optimal pathological feature set based on the accuracy and the number of important pathological features added to the classification model comprises:

[0028] determining a comprehensive score of the important pathological features added to the classification model according to the accuracy and the number of important pathological features added to the classification model;

[0029] The important pathological feature added to the classification model with the highest comprehensive score is set as the optimal pathological feature.

[0030] Preferably, the step of determining the comprehensive score of the important pathological features added to the classification model based on the accuracy rate and the number of important pathological features added to the classification model comprises:

[0031] When the number of important pathological features added to the classification model is less than the second threshold, the comprehensive score of the important pathological features added to the classification model is calculated according to the following formula:

[0032] ;

[0033] in, is the comprehensive score, a is the accuracy, is the weight of accuracy;

[0034] When the number of important pathological features added to the classification model is greater than or equal to the second threshold, the comprehensive score of the important pathological features added to the classification model is calculated according to the following formula:

[0035] a m;

[0036] in, is the comprehensive score, a is the accuracy, m is the number of important pathological features added to the classification model, , are the weights of accuracy and quantity respectively.

[0037] Preferably, it also includes:

[0038] An optimal classification model is generated based on the optimal pathological feature set.

[0039] A device for analyzing pathological characteristics of breast ductal carcinoma in situ, comprising:

[0040] A feature label acquisition module, used to acquire a sample pathological feature set and a sample label of a sample ductal carcinoma in situ breast; wherein the sample pathological feature set includes at least three nuclear magnetic resonance imaging features; and the sample label includes the presence of cancer progression and the absence of cancer progression;

[0041] An important feature determination module, used to determine at least two important pathological features and an importance ranking corresponding to each important pathological feature based on the sample pathological feature set and sample label of the sample ductal carcinoma in situ breast;

[0042] A classification model processing module, used to sequentially add the important pathological features to the classification model according to the importance ranking, and calculate the accuracy of the classification model;

[0043] The optimal feature determination module is used to determine the optimal pathological feature set according to the accuracy rate and the number of important pathological features added to the classification model.

[0044] A computer device comprises a processor, a memory and a computer program stored in the memory and capable of running on the processor, wherein the computer program implements any of the above methods when executed by the processor.

[0045] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any one of the above items is implemented.

[0046] This application has the following advantages:

[0047] In an embodiment of the present application, with respect to the problem that the prior art cannot analyze the MRI imaging features that are highly correlated with the progression of DCIS, the present application provides a solution for comprehensively analyzing the importance of features, prediction accuracy and number of features to screen out the optimal feature set, specifically: "Obtain a sample pathological feature set and a sample label of a sample DCIS breast; wherein the sample pathological feature set includes at least three MRI imaging features; the sample label includes the presence of cancer progression and the absence of cancer progression; based on the sample pathological feature set and the sample label of the sample DCIS breast, determine at least two important pathological features and an importance ranking corresponding to each important pathological feature; based on the importance ranking, add the important pathological features to the classification model in turn, and calculate the accuracy of the classification model; based on the accuracy and the number of important pathological features added to the classification model, determine the optimal pathological feature set". By analyzing the importance of features to screen out important pathological features, and analyzing the number of features and the prediction accuracy of the corresponding classification model to screen out the optimal pathological features, magnetic resonance imaging features that are highly correlated with the progression of ductal carcinoma in situ can be provided, which helps medical staff to quickly and accurately assess the patient's condition. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the description of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0049] Figure 1 This is a flowchart of the steps of a method for analyzing pathological characteristics of breast ductal carcinoma in situ provided in one embodiment of the present application;

[0050] Figure 2 This is a structural block diagram of a pathological feature analysis device for breast ductal carcinoma in situ provided by an embodiment of the present application;

[0051] Figure 3 It is a structural diagram of a computer device provided in one embodiment of the present application.

[0052] The reference numerals in the drawings of the specification are as follows:

[0053] 12. Computer equipment; 14. External devices; 16. Processing unit; 18. Bus; 20. Network adapter; 22. I / O interface; 24. Display; 28. Memory; 30. Random access memory; 32. Cache memory; 34. Storage system; 40. Program / utility; 42. Program module. DETAILED DESCRIPTION

[0054] In order to make the objects, features and advantages of the present application more obvious and understandable, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0055] Reference Figure 1 , showing a method for analyzing pathological characteristics of breast ductal carcinoma in situ provided by an embodiment of the present application, comprising:

[0056] S110, obtaining a sample pathological feature set and a sample label of a sample ductal carcinoma in situ breast; wherein the sample pathological feature set includes at least three nuclear magnetic resonance imaging features; and the sample label includes the presence of cancer progression and the absence of cancer progression;

[0057] S120, determining at least two important pathological features and an importance ranking corresponding to each important pathological feature according to the sample pathological feature set and the sample label of the sample ductal carcinoma in situ breast;

[0058] S130, adding the important pathological features to the classification model in sequence according to the importance ranking, and calculating the accuracy of the classification model;

[0059] S140: Determine an optimal pathological feature set according to the accuracy rate and the number of important pathological features added to the classification model.

[0060] In an embodiment of the present application, with respect to the problem that the prior art cannot analyze the MRI imaging features that are highly correlated with the progression of DCIS, the present application provides a solution for comprehensively analyzing the importance of features, prediction accuracy and number of features to screen out the optimal feature set, specifically: "Obtain a sample pathological feature set and a sample label of a sample DCIS breast; wherein the sample pathological feature set includes at least three MRI imaging features; the sample label includes the presence of cancer progression and the absence of cancer progression; based on the sample pathological feature set and the sample label of the sample DCIS breast, determine at least two important pathological features and an importance ranking corresponding to each important pathological feature; based on the importance ranking, add the important pathological features to the classification model in turn, and calculate the accuracy of the classification model; based on the accuracy and the number of important pathological features added to the classification model, determine the optimal pathological feature set". By analyzing the importance of features to screen out important pathological features, and analyzing the number of features and the prediction accuracy of the corresponding classification model to screen out the optimal pathological features, magnetic resonance imaging features that are highly correlated with the progression of ductal carcinoma in situ can be provided, which helps medical staff to quickly and accurately assess the patient's condition.

[0061] Next, a method for analyzing pathological characteristics of breast ductal carcinoma in situ in this exemplary embodiment will be further described.

[0062] As described in step S110, a sample pathological feature set and a sample label of the sample ductal carcinoma in situ breast are obtained; wherein the sample pathological feature set includes at least three magnetic resonance imaging features; and the sample label includes the presence of cancer progression and the absence of cancer progression.

[0063] About 200 cases of ductal carcinoma in situ of the breast treated in our hospital over the past decade were collected; the inclusion criteria were: first visit, with complete clinical history, MRI imaging results and diagnosis results; the exclusion criteria were: already diagnosed with ductal carcinoma in situ of the breast, and underwent surgical resection or chemoradiotherapy; the sample pathology feature set and sample labels were obtained from the MRI imaging results and diagnosis results of the above cases.

[0064] Specifically, the sample pathological feature set includes the following types of features: shape features (including round masses, oval masses, lobed masses, irregular masses, smooth edges, irregular edges, and star-shaped edges), enhancement features (including uniform enhancement, uneven enhancement, ring enhancement, intra-mass septal enhancement, mass center enhancement, linear enhancement, plexiform enhancement, regional uneven enhancement, multi-regional enhancement, diffuse enhancement, segmental distribution of lesions, and spotted lesions), breast tissue features (including nipple retraction and depression, T1WI ductal high signal, skin retraction and thickening, trabecular thickening, and subcutaneous edema) and lymph node features (including axillary lymph node enlargement). The sample labels include the occurrence of cancer progression (including infiltration and metastasis) and the absence of cancer progression.

[0065] As described in step S120, at least two important pathological features and the importance ranking corresponding to each important pathological feature are determined based on the sample pathological feature set and the sample label of the sample ductal carcinoma in situ breast.

[0066] To facilitate subsequent calculations and analysis, all sample pathological features and sample labels were converted into numerical form; statistical methods were used to calculate the correlation between each sample pathological feature and other sample pathological features, as well as the correlation between each sample pathological feature and the sample label; based on the calculation results, repeated or highly correlated features were eliminated, and features that were highly correlated with the labels were listed in order to obtain important pathological features and their importance rankings.

[0067] By calculating the correlation between features, it helps to eliminate duplicate or highly correlated features and avoid feature redundancy; by calculating the correlation between features and labels, it helps to screen out features that are highly correlated with labels.

[0068] As described in step S130, the important pathological features are added to the classification model in sequence according to the importance ranking, and the accuracy of the classification model is calculated.

[0069] Prepare an empty classification model (such as a support vector machine or random forest), sort by importance, and add important pathological features to the classification model one by one; use sample data to train and evaluate the classification model. As an example, divide the dataset into a training set and a test set, use the training set to train the model, and then use the test set to evaluate the accuracy of the model; record the number of features added and the accuracy of the model's predictions.

[0070] In a specific implementation, the sample data is divided into a training set and a test set in a ratio of 8:2; an empty support vector machine is created, and the performance and complexity parameters are set as follows: C=1.0; kernel is the kernel function, and gamma is the scale; important pathological features are added to the support vector machine one by one according to the importance, the training set is used to train the support vector machine, and the test set is used to evaluate the accuracy of the trained support vector machine. It should be noted that in the i-th feature addition, the feature values ​​and corresponding labels of the first i important features are selected as the current training data and test data.

[0071] As described in step S140, an optimal pathological feature set is determined based on the accuracy and the number of important pathological features added to the classification model.

[0072] Based on the accuracy of model prediction and the number of features added to the model, the comprehensive score of the features added to the model was calculated, and the optimal pathological features were screened out based on the comprehensive score.

[0073] By calculating the comprehensive score of the added features, the accuracy of the model prediction and the number of features are comprehensively considered, and a classification model with high accuracy can be obtained while ensuring the operational efficiency of the model.

[0074] In one embodiment of the present application, the specific process of the step of "determining at least two important pathological features and the importance ranking corresponding to each important pathological feature based on the sample pathological feature set and sample label of the sample ductal carcinoma in situ breast" can be further explained in combination with the following description.

[0075] The sample pathological characteristics and sample labels of the sample ductal carcinoma in situ breast are numerically represented. Specifically, for a round mass, the diameter or radius of the round mass is measured, and the diameter or radius of the round mass is used as a characteristic value in millimeters; for an oval mass, the length of the major axis and the minor axis of the oval mass are measured, and the length of the major axis and the minor axis of the oval mass are used as characteristic values ​​in millimeters; for lobed masses and irregular masses, the area of ​​the mass is calculated using image processing technology, and the area of ​​the mass is used as a characteristic value in square millimeters; for smooth, irregular, and star-shaped edge features, a binary value is used to indicate its presence or absence, for example, 1 for smooth and 0 for uneven; for uniform enhancement, uneven enhancement, and ring enhancement, use Digital coding indicates its intensity or uses binary values ​​to indicate its presence or absence. For nipple retraction and depression, the depth of the depression is measured and used as the characteristic value in millimeters. For T1WI ductal high signal, skin retraction, thickening, trabecular thickening, and subcutaneous edema, binary values ​​are used to indicate its presence or absence, or quantitative indicators are used to indicate its severity. For axillary lymph node enlargement, the size of the lymph nodes is measured and used as the characteristic value in millimeters, or binary values ​​are used to indicate its presence or absence. For sample labels, binary values ​​are used to indicate their categories, for example, 1 indicates cancer progression and 0 indicates no cancer progression.

[0076] According to the numerically represented sample pathological features, the characteristic correlation coefficient between each sample pathological feature and the remaining sample pathological features is determined. As an example, the characteristic correlation coefficient between each sample pathological feature and the remaining sample pathological features is calculated according to the following formula:

[0077] ; (1)

[0078] in, is the feature correlation coefficient between feature x and feature y, is the value of the feature x of the i-th sample, is the value of the feature y of the i-th sample, is the average value of feature x of all samples, is the average value of feature y of all samples, and n is the number of samples.

[0079] According to the sample pathological features and sample labels after numerical representation, the label correlation coefficient between each sample pathological feature and the sample label is determined. As an example, according to the sample pathological features and sample labels after numerical representation, the rank (i.e., sorting) of each sample pathological feature of each sample ductal carcinoma in situ breast in all samples and the rank of each sample ductal carcinoma in situ breast sample label in all samples are determined; the label correlation coefficient between each sample pathological feature and the sample label is calculated according to the following formula:

[0080] ; (2)

[0081] in, is the label correlation coefficient between feature x and label z, is the rank of the feature x of the i-th sample, is the rank of the label z of the ith sample, and n is the number of samples.

[0082] When the feature correlation coefficient between any two sample pathological features is greater than a first threshold, the sample pathological feature with a lower label correlation coefficient is set as a redundant pathological feature.

[0083] All sample pathological features except the redundant pathological features are set as candidate pathological features.

[0084] A preset number of the candidate pathological features with the highest label correlation coefficients are set as important pathological features.

[0085] According to the label correlation coefficient, the importance ranking corresponding to each important pathological feature is determined.

[0086] In one embodiment of the present application, the specific process of the step of "determining the optimal pathological feature set based on the accuracy rate and the number of important pathological features added to the classification model" can be further explained in combination with the following description.

[0087] According to the accuracy and the number of important pathological features added to the classification model, a comprehensive score of the important pathological features added to the classification model is determined. As an example, when the number of important pathological features added to the classification model is less than the second threshold, the comprehensive score of the important pathological features added to the classification model is calculated according to the following formula:

[0088] ; (3)

[0089] in, is the comprehensive score, a is the accuracy, is the weight of accuracy;

[0090] When the number of important pathological features added to the classification model is greater than or equal to the second threshold, the comprehensive score of the important pathological features added to the classification model is calculated according to the following formula:

[0091] a m; (4)

[0092] in, is the comprehensive score, a is the accuracy, m is the number of important pathological features added to the classification model, , are the weights of accuracy and quantity respectively.

[0093] The important pathological feature added to the classification model with the highest comprehensive score is set as the optimal pathological feature.

[0094] In an embodiment of the present application, the method further includes:

[0095] According to the optimal pathological feature set, an optimal classification model is generated. Specifically, the optimal pathological feature is added to an empty classification model; the classification model is trained using sample data to obtain the optimal classification model.

[0096] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0097] Reference Figure 2 , shows a pathological feature analysis device for breast ductal carcinoma in situ provided by an embodiment of the present application, comprising:

[0098] The feature label acquisition module 210 is used to acquire a sample pathological feature set and a sample label of a sample ductal carcinoma in situ breast; wherein the sample pathological feature set includes at least three nuclear magnetic resonance imaging features; and the sample label includes the presence of cancer progression and the absence of cancer progression;

[0099] An important feature determination module 220 is used to determine at least two important pathological features and an importance ranking corresponding to each important pathological feature based on the sample pathological feature set and sample label of the sample ductal carcinoma in situ breast;

[0100] A classification model processing module 230, for sequentially adding the important pathological features to the classification model according to the importance ranking, and calculating the accuracy of the classification model;

[0101] The optimal feature determination module 240 is used to determine the optimal pathological feature set according to the accuracy rate and the number of important pathological features added to the classification model.

[0102] Reference Figure 3 , shows a computer device of the present application, wherein the computer device 12 is in the form of a general-purpose computing device; the computer device 12 includes: one or more processors or processing units 16, a memory 28, and a bus 18 connecting different system components (including the memory 28 and the processing unit 16).

[0103] The bus 18 may be one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. For example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, an Audio Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0104] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0105] Memory 28 may include computer system readable media in the form of volatile memory, such as random access memory 30 and / or cache memory 32. Computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write to non-removable, non-volatile magnetic media (commonly referred to as a "hard drive"). Although Figure 3 Not shown, a disk drive for reading and writing to a removable non-volatile disk (such as a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (such as a CD-ROM, DVD-ROM or other optical medium) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The memory may include at least one program product having a set (e.g., at least one) of program modules 42, which are configured to perform the functions of the various embodiments of the present application.

[0106] A program / utility 40 having a set (at least one) of program modules 42 may be stored in, for example, a memory, such program modules 42 including an operating system, one or more application programs, other program modules 42, and program data, each of which or some combination may include an implementation of a network environment. The program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0107] The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, cameras, etc.), one or more devices that enable an operator to interact with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed through an I / O interface 22. Furthermore, the computer device 12 may also communicate with one or more networks (e.g., local area networks (LANs)), wide area networks (WANs), and / or public networks (e.g., the Internet) through a network adapter 20. Figure 3 As shown, the network adapter 20 communicates with other modules of the computer device 12 via the bus 18. It should be understood that although Figure 3 Not shown, other hardware and / or software modules may be used in conjunction with the computer device 12, including but not limited to: microcode, device drivers, redundant processing units 16, external disk drive arrays, RAID systems, tape drives, and data backup storage systems 34, etc.

[0108] The processing unit 16 executes various functional applications and data processing by running the programs stored in the memory 28, such as implementing the pathological feature analysis method of breast ductal carcinoma in situ provided in any embodiment of the present application.

[0109] That is, when the processing unit 16 executes the above program, the following can be achieved: obtaining a sample pathological feature set and a sample label of a sample ductal carcinoma in situ breast; wherein the sample pathological feature set includes at least three nuclear magnetic resonance imaging features; the sample label includes the presence of cancer progression and the absence of cancer progression; based on the sample pathological feature set and the sample label of the sample ductal carcinoma in situ breast, determining at least two important pathological features and an importance ranking corresponding to each important pathological feature; based on the importance ranking, adding the important pathological features to the classification model in turn, and calculating the accuracy of the classification model; and determining the optimal pathological feature set based on the accuracy and the number of important pathological features added to the classification model.

[0110] In one embodiment of the present application, a computer-readable storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, the method for analyzing pathological characteristics of breast ductal carcinoma in situ provided in any embodiment of the present application is implemented.

[0111] That is, when the program is executed by the processor, it can achieve: obtaining a sample pathological feature set and a sample label of a sample ductal carcinoma in situ breast; wherein the sample pathological feature set includes at least three magnetic resonance imaging features; the sample label includes the presence of cancer progression and the absence of cancer progression; based on the sample pathological feature set and the sample label of the sample ductal carcinoma in situ breast, determining at least two important pathological features and an importance ranking corresponding to each important pathological feature; based on the importance ranking, adding the important pathological features to the classification model in turn, and calculating the accuracy of the classification model; determining the optimal pathological feature set based on the accuracy and the number of important pathological features added to the classification model.

[0112] Any combination of one or more computer-readable media may be used. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0113] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0114] The computer program code for performing the operation of the present application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the operator's computer, partially on the operator's computer, as an independent software package, partially on the operator's computer, partially on the remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the operator's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet). The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referred to each other.

[0115] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0116] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0117] The above is a detailed introduction to the method and device for analyzing the pathological characteristics of breast ductal carcinoma in situ provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for a person skilled in the art, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for analyzing the pathological characteristics of breast ductal carcinoma in situ, characterized in that: include: Obtaining a sample pathological feature set and a sample label of a sample ductal carcinoma in situ breast; wherein the sample pathological feature set includes at least three nuclear magnetic resonance imaging features; and the sample label includes the presence of cancer progression and the absence of cancer progression; According to the sample pathological feature set and sample label of the sample ductal carcinoma in situ breast, at least two important pathological features and the importance ranking corresponding to each important pathological feature are determined; including: Numerically representing the sample pathological characteristics and sample labels of the sample ductal carcinoma in situ breast; According to the numerically expressed sample pathological characteristics, the characteristic correlation coefficient between each sample pathological characteristic and the remaining sample pathological characteristics is determined; including: calculating the characteristic correlation coefficient between each sample pathological characteristic and the remaining sample pathological characteristics according to the following formula: ; in, is the feature correlation coefficient between feature x and feature y, is the value of the feature x of the i-th sample, is the value of the feature y of the i-th sample, is the average value of feature x of all samples, is the average value of feature y of all samples, and n is the number of samples; According to the sample pathological features and sample labels after numerical representation, the label correlation coefficient between each sample pathological feature and the sample label is determined; including: according to the sample pathological features and sample labels after numerical representation, the rank of each sample pathological feature of each sample ductal carcinoma in situ breast in all samples and the rank of each sample ductal carcinoma in situ breast sample label in all samples are determined; the label correlation coefficient between each sample pathological feature and the sample label is calculated according to the following formula: ; in, is the label correlation coefficient between feature x and label z, is the rank of the feature x of the i-th sample, is the rank of the label z of the i-th sample, and n is the number of samples; When the feature correlation coefficient between any two sample pathological features is greater than a first threshold, setting the sample pathological feature with a lower label correlation coefficient as a redundant pathological feature; Setting all sample pathological features except the redundant pathological features as candidate pathological features; Setting a preset number of the candidate pathological features with the highest label correlation coefficients as important pathological features; Determining the importance ranking corresponding to each important pathological feature according to the label correlation coefficient; According to the importance ranking, the important pathological features are added to the classification model in sequence, and the accuracy of the classification model is calculated; An optimal pathology feature set is determined based on the accuracy and the number of important pathology features added to the classification model.

2. The method according to claim 1, characterized in that The step of determining the optimal pathological feature set according to the accuracy and the number of important pathological features added to the classification model comprises: determining a comprehensive score of the important pathological features added to the classification model according to the accuracy and the number of important pathological features added to the classification model; The important pathological feature added to the classification model with the highest comprehensive score is set as the optimal pathological feature.

3. The method according to claim 2, characterized in that The step of determining the comprehensive score of the important pathological features added to the classification model based on the accuracy rate and the number of important pathological features added to the classification model comprises: When the number of important pathological features added to the classification model is less than the second threshold, the comprehensive score of the important pathological features added to the classification model is calculated according to the following formula: ; in, is the comprehensive score, a is the accuracy, is the weight of accuracy; When the number of important pathological features added to the classification model is greater than or equal to the second threshold, the comprehensive score of the important pathological features added to the classification model is calculated according to the following formula: a m; in, is the comprehensive score, a is the accuracy, m is the number of important pathological features added to the classification model, , are the weights of accuracy and quantity respectively.

4. The method according to claim 1, characterized in that: Also includes: An optimal classification model is generated based on the optimal pathological feature set.

5. A device for analyzing pathological characteristics of breast ductal carcinoma in situ, characterized in that: include: A feature label acquisition module, used to acquire a sample pathological feature set and a sample label of a sample ductal carcinoma in situ breast; wherein the sample pathological feature set includes at least three nuclear magnetic resonance imaging features; and the sample label includes the presence of cancer progression and the absence of cancer progression; The important feature determination module is used to determine at least two important pathological features and the importance ranking corresponding to each important pathological feature according to the sample pathological feature set and sample label of the sample ductal carcinoma in situ breast; including: A first feature determination submodule is used to numerically represent the sample pathological features and sample labels of the sample ductal carcinoma in situ breast; The second feature determination submodule is used to determine the feature correlation coefficient between each sample pathology feature and the other sample pathology features based on the sample pathology features after numerical representation; including: calculating the feature correlation coefficient between each sample pathology feature and the other sample pathology features according to the following formula: ; in, is the feature correlation coefficient between feature x and feature y, is the value of the feature x of the i-th sample, is the value of the feature y of the i-th sample, is the average value of feature x of all samples, is the average value of feature y of all samples, and n is the number of samples; The third feature determination submodule is used to determine the label correlation coefficient between each sample pathological feature and the sample label according to the sample pathological features and sample labels after numerical representation; including: determining the rank of each sample pathological feature of each sample ductal carcinoma in situ breast in all samples and the rank of each sample ductal carcinoma in situ breast sample label in all samples according to the sample pathological features and sample labels after numerical representation; calculating the label correlation coefficient between each sample pathological feature and the sample label according to the following formula: ; in, is the label correlation coefficient between feature x and label z, is the rank of the feature x of the i-th sample, is the rank of the label z of the i-th sample, and n is the number of samples; A fourth feature determination submodule, configured to set a sample pathology feature with a lower label correlation coefficient as a redundant pathology feature when the feature correlation coefficient between any two sample pathology features is greater than a first threshold; a fifth feature determination submodule, configured to set all sample pathological features except the redundant pathological features as candidate pathological features; A sixth feature determination submodule, configured to set a preset number of candidate pathological features having the highest label correlation coefficients as important pathological features; a seventh feature determination submodule, for determining the importance ranking corresponding to each important pathological feature according to the label correlation coefficient; A classification model processing module, used to sequentially add the important pathological features to the classification model according to the importance ranking, and calculate the accuracy of the classification model; The optimal feature determination module is used to determine the optimal pathological feature set according to the accuracy rate and the number of important pathological features added to the classification model.

6. A computer device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program implements the method according to any one of claims 1 to 4 when executed by the processor.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.