Risk assessment method for mammary gland patient based on big data

Abstract

The invention discloses a risk assessment method for a mammary gland patient based on big data, and the method comprises the steps: collecting a pathological image and related data of the mammary gland patient, and carrying out the preprocessing of user interaction data and response data; the related data comprises body data and historical data; performing time-varying feature extraction on the pathological image to obtain change features, and performing classification and staging on the change features to obtain staging data; performing influence analysis on the related data to obtain a variable coefficient, and constructing a mammary gland patient risk assessment model according to the staging data and the variable coefficient; and optimizing the mammary gland patient risk assessment model, inputting to-be-assessed data into the optimized mammary gland patient risk assessment model, and outputting an assessment result. The method not only can improve the accuracy of risk assessment of the mammary gland patient based on big data, but also has good interpretability, and can be directly applied to a mammary gland patient risk assessment system.

Description

Technical Field

[0001] The present invention relates to the field of big data, and in particular to a risk assessment method for breast patients based on big data. Background Art

[0002] Breast cancer is one of the most common malignant tumors in women, with high morbidity and mortality rates worldwide. With the continuous advancement of medical technology, early screening and diagnosis of breast diseases have become the key to reducing breast cancer mortality. However, traditional breast disease risk assessment methods mainly rely on doctors' clinical experience and patients' self-reported symptoms, which is highly subjective and inaccurate.

[0003] In recent years, with the rapid development of big data technology, the medical field has begun to try to use big data for disease risk assessment. Big data has the characteristics of large data volume, diverse types, and fast processing speed, and can comprehensively and objectively reflect the health status of patients. In the field of breast diseases, by collecting multi-dimensional information such as patients' pathological images, physical data, historical data, etc., a more accurate and comprehensive risk assessment model can be constructed.

[0004] However, the existing risk assessment methods for breast patients based on big data still have some shortcomings. For example, the processing of pathological images often remains at the level of simple feature extraction, failing to deeply explore the time-varying features in the images; the processing of related data also lacks systematic impact analysis, resulting in the construction of risk assessment models that are not accurate enough. In addition, the existing risk assessment models also have certain limitations in optimization, making it difficult to adapt to the personalized needs of different patients.

[0005] In response to the above problems, the present invention proposes a breast patient risk assessment method based on big data. The method collects pathological images and related data of breast patients, extracts time-varying features and classifies and stages the images, performs impact analysis on the related data and obtains variable coefficients, thereby constructing a more accurate and comprehensive risk assessment model. At the same time, the method also improves the accuracy and applicability of the model by optimizing the risk assessment model, providing doctors with more reliable decision support. Summary of the invention

[0006] The purpose of the present invention is to provide a risk assessment method for breast patients based on big data.

[0007] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0008] The present invention comprises the following steps:

[0009] Collecting pathological images and related data of breast patients, and preprocessing the user interaction data and the response data; the related data includes physical data and historical data;

[0010] Extracting time-varying features of the pathological image to obtain variation features, and classifying and staging the variation features to obtain staging data;

[0011] Performing an impact analysis on the relevant data to obtain a variable coefficient, and constructing a breast patient risk assessment model based on the staging data and the variable coefficient;

[0012] Optimize the breast patient risk assessment model, input the data to be assessed into the optimized breast patient risk assessment model, and output the assessment result.

[0013] Furthermore, the method of extracting time-varying features from the pathological image to obtain the variation features includes:

[0014] A neural network algorithm is used to extract the features of pathological images to obtain pathological features, and the pathological features are arranged in ascending order according to the time sequence to obtain the time series features;

[0015] Calculate the similarity of time series features:

[0016]

[0017] The similarity between the sth time series feature and the fth time series feature is H sf , the s-th temporal feature of the d-th pathological image is The fth temporal feature of the pth pathological image is The similarity tolerance parameter is The error coefficient is σ, and the descriptor of the fth temporal feature of the dth pathological image is The sth temporal feature descriptor of the pth pathological image is The suppression factor is ρ, and the descriptor of the s-th temporal feature of the d-th pathological image is The descriptor of the fth temporal feature of the pth pathological image is

[0018] Classify the time series features with similarity greater than 0.478 into one category to obtain classification group features, and calculate the weight of the classification group features:

[0019]

[0020] The weight of the feature of the sth classification group is The eigenvector of the s-th classification group is h s , transposed to T, the transposed feature vector of the rth classification group is The contribution of the s-th classification group feature is The importance of the s-th classification group feature is The feature of the sth classification group is h s , classification group feature h s The probability of occurrence is P(h s );

[0021] The classification group features with a weight contribution greater than 0.327 are taken as key features, and the time skewness of the key features is calculated;

[0022]

[0023] The time skewness of the sth time series feature at the c+1th moment and the cth moment is The offset weight is γ, the maximum function is max(·), and the s-th time series feature value at the c+1th time is The sth time series characteristic value at the cth moment is The difference adjustment parameter is ε;

[0024] The key features with time skewness greater than 0.317 are output as change features.

[0025] Furthermore, the method of classifying and staging the change characteristics to obtain staging data includes:

[0026] Obtain the pathology comparison standard, obtain the standard features through the comparison standard, and calculate the similarity between the change features and the standard features:

[0027]

[0028] The number of change features of the pathological image is M, and the number of the yth change feature is M y , the yth change feature vector of the i-th pathological image is v y，i , the a-th change feature vector of the i-th pathological image is v a，i , the i-th research data of the c-th simulation is u c，i , the i-th research data of the τ-th initial center set is u τ，i , the control factor is τ, and the optimization coefficient is The similarity between the yth change feature and the ath change feature of the i-th pathological image is

[0029] The changes with the largest similarity are taken as the elements of the initial center set to obtain the initial classification clusters, and the mean distance between the initial classification clusters is calculated:

[0030]

[0031] The mean distance between the initial classification clusters is The number of initial classification clusters is W, and the jth change feature of the initial classification cluster is uj , the xth change feature of the initial classification cluster is u x , change feature u j Heyan Change Characteristics x The distance is dt(u j ,u x );

[0032] Calculate the center compactness of the initial classification clusters:

[0033]

[0034] The central compactness of the initial classification cluster is ε, and the center point of the initial classification cluster is The number of the jth changing features is W1;

[0035] Update the elements of the initial classification cluster by the center compactness until the center compactness reaches the maximum, and use the clustering result as the grouping data, otherwise replace the initial center set;

[0036] Calculate breast relevance for grouped data:

[0037]

[0038] The number of changing features of the grouped data is The breast correlation of the i-th pathological image is The importance of the jth change feature is The jth eigenvector is z j , transposed to T, the identity matrix is ​​E, and the transposed of the jth eigenvector is

[0039] The breast correlation degree less than 0.13 is stage 1; the breast correlation degree greater than 0.13 and less than 0.351 is stage 2; the breast correlation degree greater than 0.351 and less than 0.592 is stage 3; the breast correlation degree greater than 0.592 and less than 0.784 is stage 4; the breast correlation degree greater than 0.784 and less than 1 is stage 5.

[0040] Furthermore, the method of performing influence analysis on the relevant data to obtain the variable coefficient includes:

[0041] Calculate the influence of relevant data:

[0042]

[0043] Among them The influence of relevant data is The initial risk of the relevant data is C o , No. The pathological risk of the relevant data is The number of relevant data is N, the adjustment factor is η, and the The contribution of relevant data is

[0044] Calculate the correlation:

[0045]

[0046] Among them Related data and pathological risk changes The degree of correlation is No. The control coefficient of the relevant data is No. The relevant data is Related data and pathological risk changes The influence of

[0047] The relevant data with a correlation degree greater than 0.352 are regarded as important data, and the variable coefficient is calculated based on the important data:

[0048]

[0049] The proportional control coefficient is The constant of change is The coefficient of variation is The error coefficient is υ, and the relevant data and pathological risk changes The probability of simultaneous occurrence is

[0050] Furthermore, a method for constructing a breast patient risk assessment model based on the staging data and the variable coefficient includes:

[0051] The objective function is constructed based on the period data and variable coefficients, and the expression is:

[0052]

[0053]

[0054] Among them The coefficient of variation for each patient is No. The energy efficiency of each patient is No. The coefficient of variation for each patient is No. Optimization rate per patient No. The loss factor for each patient is No. The number of installment data is No. The data value of each period is No. The breast relevance of stage data is No. The actual risk of breast cancer in each patient is No. The predicted risk of breast cancer in each patient is The loss function is

[0055] The breast patient risk assessment model includes long short-term memory network algorithm, matching algorithm, and deep learning algorithm;

[0056] The long short-term memory network algorithm extracts the time series change characteristics of the input data and selects key features according to the degree of change of the time series change characteristics;

[0057] The matching algorithm compares the key features with breast tumor pathological features to obtain the matching degree, and uses the key features with a matching degree greater than 0.615 as special features to calculate the tumor risk;

[0058] The deep learning algorithm learns the complex model of special features and tumor risk, automatically extracts special features and establishes mapping relationships, and predicts tumor risk based on the mapping relationships.

[0059] Furthermore, the method for optimizing the breast patient risk assessment model comprises:

[0060] Introduce the search sub-population and perform chaotic mapping on the search sub-population. The expression is:

[0061]

[0062] The initial position sequence of the searcher is Rb, and the chaotic mapping sequence of the searcher is Rb +1 , the mapping coefficient is α;

[0063] Calculate the fitness of the searcher:

[0064]

[0065] The penalty factor for the w-th searcher is The objective function of the w-th searcher is The number of searchers is The fitness of the w-th searcher is X w , the balance parameter is β, and the error classification factor is μ;

[0066] Compute the position of the searcher:

[0067]

[0068] The upper bound of the search area is U max , the lower bound of the search area is U min , the random function is rand(·), and the position of the w-th searcher in the t-th iteration is

[0069] The searcher position with the largest fitness is taken as the target position, and the searcher position is updated according to the density factor to obtain the density position. The expression is:

[0070]

[0071] The target location is The density position of the w-th searcher in the t+1th iteration is The position of the w-th searcher in the t-th iteration is The maximum number of iterations is t max , the current iteration number is t, the random numbers from 0 to 1 are r1, r2, r3, the density factor is ζ, the control coefficient is β, the density coefficient is ω, and the position of the w+1th searcher is

[0072] Update the position of the searcher according to the adaptive coefficient and inertia weight to obtain the inertial position. The expression is:

[0073]

[0074]

[0075]

[0076] The inertial position of the w-th searcher at the t+1th iteration is The inertia coefficient of the tth iteration is λ(t), and the maximum value of the inertia coefficient is λ max , the minimum value of the adaptive coefficient is λ min , the inertia weight is l, and the output mapping vector of the tth iteration is R t ;

[0077] Iterate continuously until the maximum number of iterations is reached, otherwise update the particle position and recalculate the fitness.

[0078] In a second aspect, an embodiment of the present application further provides an electronic device, including:

[0079] A processor; and a memory arranged to store computer executable instructions, which when executed cause the processor to perform the method steps described in the first aspect.

[0080] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple applications, the electronic device executes the method steps described in the first aspect.

[0081] The beneficial effects of the present invention are:

[0082] The present invention is a risk assessment method for breast patients based on big data. Compared with the prior art, the present invention has the following technical effects:

[0083] The present invention can improve the accuracy of risk assessment of breast patients based on big data through preprocessing, time-varying feature extraction, classification and staging, impact analysis, model construction and model optimization steps, thereby improving the accuracy of risk assessment of breast patients based on big data, optimizing the risk assessment of multiple breast patients based on big data, greatly saving resources, and improving work efficiency. It can realize automatic risk assessment of breast patients based on big data, and perform data fusion on risk assessment of multiple breast patients based on big data in real time, which is of great significance to risk assessment of breast patients based on big data, can adapt to risk assessment of breast patients based on big data of different standards and risk assessment needs of different breast patients based on big data, and has a certain universality. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 The present invention is a flowchart of the steps of the risk assessment method for breast patients based on big data;

[0085] Figure 2 It is a schematic diagram of the structure of an electronic device in an embodiment of this specification. DETAILED DESCRIPTION

[0086] The present invention is further described below by means of specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0087] The risk assessment method for breast patients based on big data of the present invention comprises the following steps:

[0088] like Figure 1 As shown, in this embodiment, the following steps are included:

[0089] Collecting pathological images and related data of breast patients, and preprocessing the user interaction data and the response data; the related data includes physical data and historical data;

[0090] In the actual evaluation, the historical data is the patient's historical breast data, taking the pathological image of patient XXX detected in 2023 as an example; the relevant data includes staying up late, mood, sleeping posture, male hormone level, family history, eating habits, obesity, mental factors, environmental data, and lifestyle;

[0091] Extracting time-varying features of the pathological image to obtain variation features, and classifying and staging the variation features to obtain staging data;

[0092] In the actual evaluation, the change characteristics are nipple mass, tumor boundary, and blood flow signal; the staging data are stage 3 for nipple mass, stage 4 for tumor boundary, and stage 2 for blood flow signal;

[0093] Performing an impact analysis on the relevant data to obtain a variable coefficient, and constructing a breast patient risk assessment model based on the staging data and the variable coefficient;

[0094] In the actual evaluation, the coefficient of variation was 0.537;

[0095] Optimize the breast patient risk assessment model, input the data to be assessed into the optimized breast patient risk assessment model, and output the assessment result.

[0096] In this embodiment, the method of extracting time-varying features from the pathological image to obtain the variation features includes:

[0097] A neural network algorithm is used to extract the features of pathological images to obtain pathological features, and the pathological features are arranged in ascending order according to the time sequence to obtain the time series features;

[0098] Calculate the similarity of time series features:

[0099]

[0100] The similarity between the sth time series feature and the fth time series feature is H sf , the s-th temporal feature of the d-th pathological image is The fth temporal feature of the pth pathological image is The similarity tolerance parameter is The error coefficient is σ, and the descriptor of the fth temporal feature of the dth pathological image is The sth temporal feature descriptor of the pth pathological image is The suppression factor is ρ, and the descriptor of the s-th temporal feature of the d-th pathological image is The descriptor of the fth temporal feature of the pth pathological image is

[0101] Classify the time series features with similarity greater than 0.478 into one category to obtain classification group features, and calculate the weight of the classification group features:

[0102]

[0103] The weight of the feature of the sth classification group is The eigenvector of the s-th classification group is h s , transposed to T, the transposed feature vector of the rth classification group is The contribution of the s-th classification group feature is The importance of the s-th classification group feature is The feature of the sth classification group is h s , classification group feature h s The probability of occurrence is P(h s );

[0104] The classification group features with a weight contribution greater than 0.327 are taken as key features, and the time skewness of the key features is calculated;

[0105]

[0106] The time skewness of the sth time series feature at the c+1th moment and the cth moment is The offset weight is γ, the maximum function is max(·), and the s-th time series feature value at the c+1th time is The sth time series characteristic value at the cth moment is The difference adjustment parameter is ε;

[0107] The key features with time skewness greater than 0.317 are output as change features.

[0108] In this embodiment, the method for classifying and staging the change characteristics to obtain staging data includes:

[0109] Obtain the pathology comparison standard, obtain the standard features through the comparison standard, and calculate the similarity between the change features and the standard features:

[0110]

[0111] The number of change features of the pathological image is M, and the number of the yth change feature is M y , the yth change feature vector of the i-th pathological image is v y，i , the a-th change feature vector of the i-th pathological image is v a，i， The i-th research data of the c-th simulation is u c，i , the i-th research data of the τ-th initial center set is u τ，i , the control factor is τ, and the optimization coefficient is The similarity between the yth change feature and the ath change feature of the i-th pathological image is

[0112] The changes with the largest similarity are taken as the elements of the initial center set to obtain the initial classification clusters, and the mean distance between the initial classification clusters is calculated:

[0113]

[0114] The mean distance between the initial classification clusters is The number of initial classification clusters is W, and the jth change feature of the initial classification cluster is u j , the xth change feature of the initial classification cluster is u x , change feature u j Heyan Change Characteristics x The distance is dt(u j ,u x );

[0115] Calculate the center compactness of the initial classification clusters:

[0116]

[0117] The central compactness of the initial classification cluster is ε, and the center point of the initial classification cluster is The number of the jth changing features is W1;

[0118] Update the elements of the initial classification cluster by the center compactness until the center compactness reaches the maximum, and use the clustering result as the grouping data, otherwise replace the initial center set;

[0119] Calculate breast relevance for grouped data:

[0120]

[0121] The number of changing features of the grouped data is The breast correlation of the i-th pathological image is The importance of the jth change feature is The jth eigenvector is z j , transposed to T, the identity matrix is ​​E, and the transposed of the jth eigenvector is

[0122] The breast correlation degree less than 0.13 is stage 1; the breast correlation degree greater than 0.13 and less than 0.351 is stage 2; the breast correlation degree greater than 0.351 and less than 0.592 is stage 3; the breast correlation degree greater than 0.592 and less than 0.784 is stage 4; the breast correlation degree greater than 0.784 and less than 1 is stage 5.

[0123] In this embodiment, the method of performing influence analysis on the relevant data to obtain the variable coefficient includes:

[0124] Calculate the influence of relevant data:

[0125]

[0126] Among them The influence of relevant data is The initial risk of the relevant data is C o , No. The pathological risk of the relevant data is The number of relevant data is N, the adjustment factor is η, and the The contribution of relevant data is

[0127] Calculate the correlation:

[0128]

[0129] Among them Related data and pathological risk changes The degree of correlation is No. The control coefficient of the relevant data is No. The relevant data is Related data and pathological risk changes The influence of

[0130] The relevant data with a correlation degree greater than 0.352 are regarded as important data, and the variable coefficient is calculated based on the important data:

[0131]

[0132] The proportional control coefficient is The constant of change is The coefficient of variation is χ, the error coefficient is υ, and the relevant data and pathological risk changes The probability of simultaneous occurrence is

[0133] In this embodiment, the method for constructing a breast patient risk assessment model based on the staging data and the variable coefficient includes:

[0134] The objective function is constructed based on the period data and variable coefficients, and the expression is:

[0135]

[0136]

[0137] Among them The coefficient of variation for each patient is No. The energy efficiency of each patient is No. The coefficient of variation for each patient is No. Optimization rate per patient No. The loss factor for each patient is No. The number of installment data is No. The data value of each period is No. The breast relevance of stage data is No. The actual risk of breast cancer in each patient is No. The predicted risk of breast cancer in each patient is The loss function is

[0138] The breast patient risk assessment model includes long short-term memory network algorithm, matching algorithm, and deep learning algorithm;

[0139] The long short-term memory network algorithm extracts the time series change characteristics of the input data and selects key features according to the degree of change of the time series change characteristics;

[0140] The matching algorithm compares the key features with breast tumor pathological features to obtain the matching degree, and uses the key features with a matching degree greater than 0.615 as special features to calculate the tumor risk;

[0141] The deep learning algorithm learns the complex model of special features and tumor risk, automatically extracts special features and establishes mapping relationships, and predicts tumor risk based on the mapping relationships.

[0142] In this embodiment, the method for optimizing the breast patient risk assessment model includes:

[0143] Introduce the search sub-population and perform chaotic mapping on the search sub-population. The expression is:

[0144]

[0145] The initial position sequence of the searcher is Rb, and the chaotic mapping sequence of the searcher is Rb +1 , the mapping coefficient is α;

[0146] Calculate the fitness of the searcher:

[0147]

[0148] The penalty factor for the w-th searcher is The objective function of the w-th searcher is The number of searchers is The fitness of the w-th searcher is X w , the balance parameter is β, and the error classification factor is μ;

[0149] Compute the position of the searcher:

[0150]

[0151] The upper bound of the search area is U max , the lower bound of the search area is U min , the random function is rand(·), and the position of the w-th searcher in the t-th iteration is

[0152] The searcher position with the largest fitness is taken as the target position, and the searcher position is updated according to the density factor to obtain the density position. The expression is:

[0153]

[0154]

[0155] The target location is The density position of the w-th searcher in the t+1th iteration is The position of the w-th searcher in the t-th iteration is The maximum number of iterations is t max , the current iteration number is t, the random numbers from 0 to 1 are r1, r2, r3, the density factor is ζ, the control coefficient is β, the density coefficient is ω, and the position of the w+1th searcher is

[0156] Update the position of the searcher according to the adaptive coefficient and inertia weight to obtain the inertial position. The expression is:

[0157]

[0158]

[0159]

[0160] The inertial position of the w-th searcher at the t+1th iteration is The inertia coefficient of the tth iteration is λ(t), and the maximum value of the inertia coefficient is λ max , the minimum value of the adaptive coefficient is λ min , the inertia weight is l, and the output mapping vector of the tth iteration is R t ;

[0161] Iterate continuously until the maximum number of iterations is reached, otherwise update the particle position and recalculate the fitness.

[0162] Figure 2 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 2 At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0163] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0164] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0165] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a risk assessment device for breast patients based on big data at the logical level. The processor executes the program stored in the memory and is specifically used to execute any of the above-mentioned risk assessment methods for breast patients based on big data.

[0166] The above application Figure 1The risk assessment method for breast patients based on big data disclosed in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or instructions in the form of software. The above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or executed by a combination of hardware and software modules in a decoding processor. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0167] The electronic device may also perform Figure 1 A risk assessment method for breast patients based on big data was developed and implemented Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.

[0168] An embodiment of the present application also proposes a computer-readable storage medium, which stores one or more programs, and the one or more programs include instructions. When the instructions are executed by an electronic device including multiple applications, they execute any of the aforementioned big data-based risk assessment methods for breast patients.

[0169] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0170] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0171] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0172] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0173] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0174] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0175] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0176] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or 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, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0177] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0178] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A risk assessment method for breast patients based on big data, characterized in that: The following steps are involved: Collecting pathological images and related data of breast patients, and preprocessing the user interaction data and the response data; the related data includes physical data and historical data; Extracting time-varying features of the pathological image to obtain variation features, and classifying and staging the variation features to obtain staging data; Performing an impact analysis on the relevant data to obtain a variable coefficient, and constructing a breast patient risk assessment model based on the staging data and the variable coefficient; Optimize the breast patient risk assessment model, input the data to be assessed into the optimized breast patient risk assessment model, and output the assessment result.

2. The risk assessment method for breast patients based on big data according to claim 1 is characterized in that: The method for extracting time-varying features from the pathological image to obtain changing features includes: A neural network algorithm is used to extract the features of pathological images to obtain pathological features, and the pathological features are arranged in ascending order according to the time sequence to obtain the time series features; Calculate the similarity of time series features: The similarity between the sth time series feature and the fth time series feature is H sf , the s-th temporal feature of the d-th pathological image is The fth temporal feature of the pth pathological image is The similarity tolerance parameter is The error coefficient is σ, and the descriptor of the f-th temporal feature of the d-th pathological image is The sth temporal feature descriptor of the pth pathological image is The suppression factor is ρ, and the descriptor of the s-th temporal feature of the d-th pathological image is The descriptor of the fth temporal feature of the pth pathological image is Classify the time series features with similarity greater than 0.478 into one category to obtain classification group features, and calculate the weight of the classification group features: The weight of the feature of the sth classification group is The eigenvector of the s-th classification group is h s , transposed to T, the transposed feature vector of the rth classification group is The contribution of the s-th classification group feature is The importance of the s-th classification group feature is The feature of the sth classification group is h s , classification group feature h s The probability of occurrence is P(h s ); The classification group features with a weight contribution greater than 0.327 are taken as key features, and the time skewness of the key features is calculated; The time skewness of the sth time series feature at the c+1th moment and the cth moment is The offset weight is γ, the maximum function is max(·), and the s-th time series feature value at the c+1th time is The sth time series eigenvalue at the cth moment is The difference adjustment parameter is ε; The key features with time skewness greater than 0.317 are output as change features.

3. The risk assessment method for breast patients based on big data according to claim 1 is characterized in that: The method for classifying and staging the change characteristics to obtain staging data includes: Obtain the pathology comparison standard, obtain the standard features through the comparison standard, and calculate the similarity between the change features and the standard features: The number of change features of the pathological image is M, and the number of the yth change feature is M y , the yth change feature vector of the i-th pathological image is v y,i , the a-th change feature vector of the i-th pathological image is v a,i , the i-th research data of the c-th simulation is u c,i , the i-th research data of the τ-th initial center set is u τ,i , the control factor is τ, and the optimization coefficient is The similarity between the yth change feature and the ath change feature of the i-th pathological image is The changes with the largest similarity are taken as the elements of the initial center set to obtain the initial classification clusters, and the mean distance between the initial classification clusters is calculated: The mean distance between the initial classification clusters is The number of initial classification clusters is W, and the jth change feature of the initial classification cluster is u j , the xth change feature of the initial classification cluster is u x , change feature u j Characteristics of changes in research x The distance is dt(u j ,u x ); Calculate the center compactness of the initial classification clusters: The central compactness of the initial classification cluster is ε, and the center point of the initial classification cluster is The number of the jth changing features is W1; Update the elements of the initial classification cluster by the center compactness until the center compactness reaches the maximum, and use the clustering result as the grouping data, otherwise replace the initial center set; Calculate breast relevance for grouped data: The number of changing features of the grouped data is The breast correlation of the i-th pathological image is The importance of the jth change feature is The jth eigenvector is z j , transposed to T, the identity matrix is ​​E, and the transposed of the jth eigenvector is The breast correlation degree less than 0.13 is stage 1; the breast correlation degree greater than 0.13 and less than 0.351 is stage 2; the breast correlation degree greater than 0.351 and less than 0.592 is stage 3; the breast correlation degree greater than 0.592 and less than 0.784 is stage 4; the breast correlation degree greater than 0.784 and less than 1 is stage 5.

4. The risk assessment method for breast patients based on big data according to claim 1 is characterized in that: The method of performing influence analysis on the relevant data to obtain the variable coefficient includes: Calculate the influence of relevant data: Among them The influence of relevant data is The initial risk of the relevant data is C o , No. The pathological risk of the relevant data is The number of relevant data is N, the adjustment factor is η, and the The contribution of relevant data is Calculate the correlation: Among them Related data and pathological risk changes The degree of correlation is No. The control coefficient of the relevant data is No. The relevant data is Related data and pathological risk changes The influence of The relevant data with a correlation degree greater than 0.352 are regarded as important data, and the variable coefficient is calculated based on the important data: The proportional control coefficient is The constant of change is The coefficient of variation is χ, the error coefficient is υ, and the relevant data and pathological risk changes The probability of simultaneous occurrence is 5. The risk assessment method for breast patients based on big data according to claim 1 is characterized in that: The method for constructing a breast patient risk assessment model according to the staging data and the variable coefficient comprises: The objective function is constructed based on the period data and variable coefficients, and the expression is: Among them The coefficient of variation for each patient is No. The energy efficiency of each patient is No. The coefficient of variation for each patient is χ κ , No. Optimization rate per patient No. The loss factor for each patient is No. The number of installment data is No. The data value of each period is No. The breast relevance of stage data is No. The actual risk of breast cancer in each patient is No. The predicted risk of breast cancer in each patient is The loss function is The breast patient risk assessment model includes long short-term memory network algorithm, matching algorithm, and deep learning algorithm; The long short-term memory network algorithm extracts the time series change characteristics of the input data and selects key features according to the degree of change of the time series change characteristics; The matching algorithm compares the key features with breast tumor pathological features to obtain the matching degree, and uses the key features with a matching degree greater than 0.615 as special features to calculate the tumor risk; The deep learning algorithm learns the complex model of special features and tumor risk, automatically extracts special features and establishes mapping relationships, and predicts tumor risk based on the mapping relationships.

6. The risk assessment method for breast patients based on big data according to claim 1 is characterized in that: The method for optimizing the breast patient risk assessment model comprises: Introduce the search sub-population and perform chaotic mapping on the search sub-population. The expression is: The initial position sequence of the searcher is R b , the chaotic map sequence of the searcher is R b+1 , the mapping coefficient is α; Calculate the fitness of the searcher: The penalty factor of the w-th searcher is d w , the objective function of the w-th searcher is The number of searchers is The fitness of the w-th searcher is X w , the balance parameter is β, and the error classification factor is μ; Compute the position of the searcher: The upper bound of the search area is U max , the lower bound of the search area is U min , the random function is rand(·), and the position of the w-th searcher in the t-th iteration is The searcher position with the largest fitness is taken as the target position, and the searcher position is updated according to the density factor to obtain the density position. The expression is: The target location is The density position of the w-th searcher in the t+1th iteration is The position of the w-th searcher in the t-th iteration is The maximum number of iterations is t max , the current iteration number is t, the random numbers from 0 to 1 are r1, r2, r3, the density factor is ζ, the control coefficient is β, the density coefficient is ω, and the position of the w+1th searcher is Update the position of the searcher according to the adaptive coefficient and inertia weight to obtain the inertial position. The expression is: The inertial position of the w-th searcher at the t+1th iteration is The inertia coefficient of the tth iteration is λ(t), and the maximum value of the inertia coefficient is λ max , the minimum value of the adaptive coefficient is λ min , the inertia weight is l, and the output mapping vector of the tth iteration is R t ; Iterate continuously until the maximum number of iterations is reached, otherwise update the particle position and recalculate the fitness.

7. An electronic device comprising: processor; as well as A memory arranged to store computer executable instructions, which when executed cause the processor to perform the method according to any one of claims 1 to 6.

8. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, causes the electronic device to execute the method according to any one of claims 1 to 6.

Images