Breast cancer prediction model construction method and device and electronic equipment

Through ultrasound imaging annotation and multimodal imaging omics feature extraction, a breast cancer prediction model was constructed, which solved the problem of insufficient accuracy and stability of the breast cancer prediction model in the prior art, and achieved efficient breast cancer diagnosis.

CN118983059APending Publication Date: 2024-11-19SHANGHAI TENTH PEOPLES HOSPITAL
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411047252.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-01
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

The existing breast cancer prediction models have limitations in prediction accuracy, stability and clinical application. Individual differences and data acquisition standards affect the prediction effect, and the prediction cost of gene expression is high and technical requirements are strict.

Method used

By acquiring ultrasound images and their annotations, the tumor area is segmented using the Encoder-Decoder structural network, key features are extracted in combination with conventional imaging and deep imaging, and breast cancer prediction model is constructed through cascade operations to enhance the generalization of the model Ability and robustness.

Benefits of technology

It improves the accuracy and robustness of breast cancer prediction models, reduces non-tumor regional interference, and can complete a large amount of data processing and analysis in a short time, providing doctors with valuable diagnostic information.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118983059A_ABST
    Figure CN118983059A_ABST
Patent Text Reader

Abstract

The invention provides a breast cancer prediction model construction method and device and electronic equipment, and relates to the technical field of artificial intelligence prediction, and the method comprises the steps: obtaining an ultrasonic image and a label thereof; segmenting a tumor region in the ultrasonic image based on the annotation of the ultrasonic image to obtain a focus segmentation result; performing conventional image omics and depth image omics extraction on the focus segmentation result to obtain key features; and mixing the key features for cascade operation, and constructing a breast cancer prediction model. According to the method, complementarity between different features is fully utilized, the processing capability of the model on complex data is enhanced, and the generalization capability and robustness of the model are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence prediction technology, and in particular to a method, device and electronic equipment for building a breast cancer prediction model. Background Art

[0003] At present, the prediction models of breast cancer are mainly based on multiple data sources such as clinical characteristics and gene expression. However, these methods still have certain limitations in terms of prediction accuracy, stability and clinical application. For example, the prediction models based on clinical characteristics are often affected by individual differences and different data collection standards; although the gene expression prediction model has high sensitivity, it is costly and has strict technical requirements.

[0004] Therefore, a method, device and electronic device for constructing a breast cancer prediction model are proposed. Summary of the invention

[0005] This specification provides a method, device and electronic device for building a breast cancer prediction model, which fully utilizes the complementarity between different features, enhances the model's ability to process complex data, and helps to improve the generalization ability and robustness of the model.

[0006] This specification provides a method for constructing a breast cancer prediction model, comprising:

[0007] Acquire ultrasound images and their annotations;

[0008] Segmenting the tumor area in the ultrasound image based on the annotation of the ultrasound image to obtain a lesion segmentation result;

[0009] Perform conventional imaging omics and deep imaging omics extraction on the lesion segmentation results to obtain key features;

[0010] The key features are mixed and cascaded to construct a breast cancer prediction model.

[0011] Optionally, the ultrasound image includes a two-dimensional grayscale ultrasound image, a color Doppler ultrasound image, an ultrasound elasticity image, and an ultrasound contrast video.

[0012] Optionally, segmenting the tumor area in the ultrasound image based on the annotation of the ultrasound image includes: segmenting the lesion classification result using an Encoder-Decoder structure network.

[0013] Optionally, the lesion segmentation result is subjected to conventional radiomics extraction to obtain key features, including:

[0014] Performing high-dimensional feature extraction on the lesion segmentation result to obtain shape features, intensity features, texture features, wavelet transform, and LOG transform;

[0015] The lesion segmentation results are subjected to vascular morphology and vascular perfusion extraction to obtain arterial richness, blood flow signal classification, vascular velocity, and resistance index.

[0016] Optionally, performing deep imaging omics extraction on the lesion segmentation result to obtain key features includes: performing deep imaging omics extraction based on VGG16 convolutional neural network and / or 3D convolutional neural network to obtain key features.

[0017] Optionally, the mixing of the key features and performing a cascade operation to construct a breast cancer prediction model includes:

[0018] Based on the LASSO regression analysis network, the key features of the conventional radiomics are reduced in dimension and irrelevant information is eliminated;

[0019] Screening the key features of the conventional radiomics based on statistical test algorithms and machine learning algorithms to determine the highly correlated key features of the conventional radiomics;

[0020] Screening the key features of the deep imaging omics based on a multimodal image information collaborative network to determine the highly correlated key features of the deep imaging omics;

[0021] The high-correlation key features of the conventional imaging omics and the high-correlation key features of the deep imaging omics are mixed and cascaded to construct a breast cancer prediction model.

[0022] This specification provides a breast cancer prediction model construction device, comprising:

[0023] An acquisition module, used for acquiring ultrasound images and annotations thereof;

[0024] A segmentation module, used to segment the tumor area in the ultrasound image based on the annotation of the ultrasound image to obtain a lesion segmentation result;

[0025] An extraction module, used for performing conventional imaging omics and deep imaging omics extraction on the lesion segmentation results to obtain key features;

[0026] A building module is used to mix the key features and perform cascade operations to build a breast cancer prediction model.

[0027] Optionally, the ultrasound image includes a two-dimensional grayscale ultrasound image, a color Doppler ultrasound image, an ultrasound elasticity image, and an ultrasound contrast video.

[0028] Optionally, the segmentation module includes: using an Encoder-Decoder structure network to segment the lesion classification results.

[0029] Optionally, the extraction module includes:

[0030] Performing high-dimensional feature extraction on the lesion segmentation result to obtain shape features, intensity features, texture features, wavelet transform, and LOG transform;

[0031] The lesion segmentation results are subjected to vascular morphology and vascular perfusion extraction to obtain arterial richness, blood flow signal classification, vascular velocity, and resistance index.

[0032] Optionally, the extraction module further includes: performing deep imaging omics extraction based on a VGG16 convolutional neural network and / or a 3D convolutional neural network to obtain key features.

[0033] Optionally, the building block includes:

[0034] Based on the LASSO regression analysis network, the key features of the conventional radiomics are reduced in dimension and irrelevant information is eliminated;

[0035] Screening the key features of the conventional radiomics based on statistical test algorithms and machine learning algorithms to determine the highly correlated key features of the conventional radiomics;

[0036] Screening the key features of the deep imaging omics based on a multimodal image information collaborative network to determine the highly correlated key features of the deep imaging omics;

[0037] The high-correlation key features of the conventional imaging omics and the high-correlation key features of the deep imaging omics are mixed and cascaded to construct a breast cancer prediction model.

[0038] This specification also provides an electronic device, wherein the electronic device includes:

[0039] processor; and,

[0040] A memory storing computer executable instructions, which when executed cause the processor to perform any of the above methods.

[0041] The present specification also provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, and when the one or more programs are executed by a processor, any of the above methods is implemented.

[0042] In this specification, by annotating ultrasound images, the tumor area can be accurately identified and segmented to obtain the lesion segmentation result. This precise regional segmentation helps to reduce the interference of non-tumor areas and improve the accuracy of subsequent feature extraction. Using a combination of conventional imaging genomics and deep imaging genomics, rich key features can be extracted from ultrasound images. Conventional imaging genomics focuses on intuitive features such as morphology, intensity, and texture, while deep imaging genomics can automatically learn and extract deeper abstract features. The combination of the two can fully capture the various feature information of the lesion and improve the performance of the prediction model. Mixing and cascading the key features extracted by conventional imaging genomics and deep imaging genomics can make full use of the complementarity between different features, enhance the model's ability to process complex data, and help improve the generalization ability and robustness of the model. The breast cancer prediction model constructed based on mixed features can accurately predict the occurrence of breast cancer and provide doctors with valuable diagnostic information. At the same time, due to the use of advanced methods of deep learning and imaging genomics, the model training process is efficient and can complete the processing and analysis of a large amount of data in a short time. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments 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 creative work.

[0044] Figure 1 A schematic diagram of the principle of a method for constructing a breast cancer prediction model provided in an embodiment of this specification;

[0045] Figure 2 A schematic diagram of an Encoder-Decoder structure network provided in an embodiment of the present application;

[0046] Figure 3 A schematic diagram of optimized fusion modeling of key imaging omics features of multiple omics provided in an embodiment of the present application;

[0047] Figure 4 A schematic diagram of the structure of a breast cancer prediction model building device provided in an embodiment of this specification;

[0048] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of this specification;

[0049] Figure 6 A schematic diagram of a computer-readable medium provided in accordance with an embodiment of the present specification. DETAILED DESCRIPTION

[0050] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles of the present invention defined in the following description can be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not deviate from the spirit and scope of the present invention.

[0051] The following is combined with Figure 1-6 The exemplary embodiments of the present invention are described more fully. However, the exemplary embodiments can be implemented in various forms, and it should not be understood that the present invention is limited to the embodiments set forth herein. On the contrary, providing these exemplary embodiments can make the present invention more comprehensive and complete, and it is more convenient to fully convey the inventive concept to those skilled in the art. The same reference numerals in the figures represent the same or similar elements, components or parts, and thus their repeated description will be omitted.

[0052] Under the premise of being consistent with the technical concept of the present invention, the features, structures, characteristics or other details described in a specific embodiment do not exclude that they can be combined in one or more other embodiments in a suitable manner.

[0053] In the description of specific embodiments, the features, structures, characteristics or other details described in the present invention are intended to enable those skilled in the art to fully understand the embodiments. However, it does not exclude that those skilled in the art can practice the technical solutions of the present invention without one or more of the specific features, structures, characteristics or other details.

[0054] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0055] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0056] The term "and / or" or "and / or" includes all combinations of any one or more of the associated listed items.

[0057] Figure 1 A schematic diagram of the principle of a method for constructing a breast cancer prediction model provided in an embodiment of this specification, the method may include:

[0058] S110: Acquire ultrasound images and annotations;

[0059] Optionally, the ultrasound image includes a two-dimensional grayscale ultrasound image, a color Doppler ultrasound image, an ultrasound elasticity image, and an ultrasound contrast video.

[0060] In the specific implementation of this specification, the ultrasound image data of the patient's breast tumor and axillary lymph nodes are collected, and the instruments and equipment used in the collection process, the collection conditions and the storage format of the images are recorded. At the same time, the patient's clinical information, breast tumor pathology information, immunohistochemical gene information test results, axillary lymph node pathology results, and patient survival prognosis information are sorted out.

[0061] S120: Segmenting the tumor region in the ultrasound image based on the annotation of the ultrasound image to obtain a lesion segmentation result;

[0062] Optionally, segmenting the tumor area in the ultrasound image based on the annotation of the ultrasound image includes: segmenting the lesion classification result using an Encoder-Decoder structure network.

[0063] In the specific implementation of this specification, Figure 2 As shown in the figure, the main function of the Encoder part is to convert the input lesion image or related sequence data into a fixed-length context vector. This vector can capture the semantic and grammatical information of the input data, especially the features related to the lesion. The Encoder is usually implemented using a recurrent neural network (RNN) or its variants (such as LSTM, GRU), which gradually reads the elements of the input sequence and updates its hidden state, and finally obtains a context vector containing the lesion feature information.

[0064] The decoder generates an output sequence, i.e., the segmentation result of the lesion, based on the context vector generated by the encoder. The decoder can also be implemented using RNN or its variants, which decodes the information in the context vector and gradually outputs the segmentation result. In this process, the decoder can learn how to accurately locate and segment the lesion area based on the feature information in the context vector.

[0065] In practical applications, the Encoder-Decoder structure network can be combined with other neural network models such as convolutional neural network (CNN) to form a more complex network architecture to further improve the accuracy and efficiency of lesion segmentation. For example, CNN can be used for feature extraction, and then the extracted features can be input into the Encoder-Decoder structure network for segmentation, which effectively solves the problems of obvious noise, low resolution and unclear lesion boundaries in ultrasound images, and realizes accurate segmentation of lesions in images.

[0066] S130: performing conventional imaging omics and deep imaging omics extraction on the lesion segmentation result to obtain key features;

[0067] Optionally, the lesion segmentation result is subjected to conventional radiomics extraction to obtain key features, including:

[0068] Performing high-dimensional feature extraction on the lesion segmentation result to obtain shape features, intensity features, texture features, wavelet transform, and LOG transform;

[0069] The lesion segmentation results are subjected to vascular morphology and vascular perfusion extraction to obtain arterial richness, blood flow signal classification, vascular velocity, and resistance index.

[0070] In the specific implementation of this specification, the shape characteristics of the lesion are described by calculating the perimeter, area, circularity, aspect ratio and other parameters of the lesion area. The structure and morphology of the lesion are analyzed by methods such as area filling and hole detection. The grayscale distribution of the lesion area is statistically analyzed, including statistics such as mean, standard deviation, skewness, and kurtosis. The grayscale relationship between pixels inside the lesion is analyzed to reflect the texture and contrast information of the lesion. The texture features of the lesion area are extracted by methods such as grayscale co-occurrence matrix, autocorrelation function, and Fourier transform. These features can reflect the roughness, directionality, and periodicity of the lesion. The lesion image is subjected to wavelet decomposition to obtain wavelet coefficients of different scales and directions, which can reveal the detailed information of the lesion at different frequencies and spatial positions. The lesion image is subjected to LOG filtering to extract the edge information of the lesion. The LOG transformation can enhance the edges and details in the image, which helps to further analyze the morphology and structure of the lesion.

[0071] Optionally, performing deep imaging omics extraction on the lesion segmentation result to obtain key features includes: performing deep imaging omics extraction based on VGG16 convolutional neural network and / or 3D convolutional neural network to obtain key features.

[0072] In the specific implementation of this specification, the VGG16 convolutional neural network, due to its deep structure and specific convolution layer design, can effectively extract multi-level features from the input image. These features, from low-level edge and texture information to high-level abstract features, can be gradually extracted and recognized by the VGG16 network. The characteristic of VGG16 is that it uses a smaller 3x3 convolution kernel and a convolution operation with a step size of 1, which helps the network learn more local features, increases the depth of the network, and improves the expressiveness of features.

[0073] At the same time, 3D convolutional neural networks are also a very effective tool when processing three-dimensional medical imaging data. Compared with 2D convolutional neural networks, 3D convolutional neural networks can directly process three-dimensional data and extract feature information in three-dimensional space. This is especially important for medical imaging analysis because medical imaging data is usually three-dimensional and contains rich spatial information.

[0074] When using VGG16 or 3D convolutional neural networks for deep imaging omics extraction, the following steps are usually required: necessary preprocessing of the input medical imaging data, such as denoising and normalization, to eliminate the heterogeneity between data and improve the training effect of the model; use the labeled data to train the VGG16 or 3D convolutional neural network. During the training process, the network will automatically learn how to extract key features from the input data; after the training is completed, the trained model can be used to extract features from new medical imaging data.

[0075] It is worth noting that the feature extraction process is not simply taking the output of a certain layer in the network as a feature. In fact, each layer in the deep neural network extracts features at different levels, and these features may have different importance in different tasks. Therefore, in practical applications, it may be necessary to select an appropriate feature extraction strategy based on specific task requirements.

[0076] S140: Mix the key features and perform a cascade operation to construct a breast cancer prediction model.

[0077] Optionally, the mixing of the key features and performing a cascade operation to construct a breast cancer prediction model includes:

[0078] Based on the LASSO regression analysis network, the key features of the conventional radiomics are reduced in dimension and irrelevant information is eliminated;

[0079] Screening the key features of the conventional radiomics based on statistical test algorithms and machine learning algorithms to determine the highly correlated key features of the conventional radiomics;

[0080] Screening the key features of the deep imaging omics based on a multimodal image information collaborative network to determine the highly correlated key features of the deep imaging omics;

[0081] The high-correlation key features of the conventional imaging omics and the high-correlation key features of the deep imaging omics are mixed and cascaded to construct a breast cancer prediction model.

[0082] In the specific implementation of this specification, the statistical test algorithm includes the Mann-Whitney U test and the chi-square test. The machine learning algorithm includes the RFE-SVM algorithm, the random forest algorithm, and the AdaBoost algorithm. Figure 3 As shown in the figure, a multimodal image information collaborative network is used to comprehensively utilize image information of different modalities, and the selected conventional imaging genomics features and deep network features are mixed for cascade operation to reduce the response differences of different networks (ResNet, DenseNet, SENet networks, etc.) to the same lesion, supervise each other, and make different convolutional neural network parameter updates independent of each other for different data during back-propagation gradient transmission, making the lesion information more comprehensive and the network classification output more credible.

[0083] In this specification, by annotating ultrasound images, the tumor area can be accurately identified and segmented to obtain the lesion segmentation result. This precise regional segmentation helps to reduce the interference of non-tumor areas and improve the accuracy of subsequent feature extraction. Using a combination of conventional imaging genomics and deep imaging genomics, rich key features can be extracted from ultrasound images. Conventional imaging genomics focuses on intuitive features such as morphology, intensity, and texture, while deep imaging genomics can automatically learn and extract deeper abstract features. The combination of the two can fully capture the various feature information of the lesion and improve the performance of the prediction model. Mixing and cascading the key features extracted by conventional imaging genomics and deep imaging genomics can make full use of the complementarity between different features, enhance the model's ability to process complex data, and help improve the generalization ability and robustness of the model. The breast cancer prediction model constructed based on mixed features can accurately predict the occurrence of breast cancer and provide doctors with valuable diagnostic information. At the same time, due to the use of advanced methods of deep learning and imaging genomics, the model training process is efficient and can complete the processing and analysis of a large amount of data in a short time.

[0084] Figure 4 A schematic diagram of the principle of a breast cancer prediction model construction device provided in an embodiment of this specification, the device may include:

[0085] An acquisition module, used for acquiring ultrasound images and annotations thereof;

[0086] A segmentation module, used to segment the tumor area in the ultrasound image based on the annotation of the ultrasound image to obtain a lesion segmentation result;

[0087] An extraction module, used for performing conventional imaging omics and deep imaging omics extraction on the lesion segmentation results to obtain key features;

[0088] A building module is used to mix the key features and perform cascade operations to build a breast cancer prediction model.

[0089] Optionally, the ultrasound image includes a two-dimensional grayscale ultrasound image, a color Doppler ultrasound image, an ultrasound elasticity image, and an ultrasound contrast video.

[0090] Optionally, the segmentation module includes: using an Encoder-Decoder structure network to segment the lesion classification results.

[0091] Optionally, the extraction module includes:

[0092] Performing high-dimensional feature extraction on the lesion segmentation result to obtain shape features, intensity features, texture features, wavelet transform, and LOG transform;

[0093] The lesion segmentation results are subjected to vascular morphology and vascular perfusion extraction to obtain arterial richness, blood flow signal classification, vascular velocity, and resistance index.

[0094] Optionally, the extraction module further includes: performing deep imaging omics extraction based on a VGG16 convolutional neural network and / or a 3D convolutional neural network to obtain key features.

[0095] Optionally, the building block includes:

[0096] Based on the LASSO regression analysis network, the key features of the conventional radiomics are reduced in dimension and irrelevant information is eliminated;

[0097] Screening the key features of the conventional radiomics based on statistical test algorithms and machine learning algorithms to determine the highly correlated key features of the conventional radiomics;

[0098] Screening the key features of the deep imaging omics based on a multimodal image information collaborative network to determine the highly correlated key features of the deep imaging omics;

[0099] The high-correlation key features of the conventional imaging omics and the high-correlation key features of the deep imaging omics are mixed and cascaded to construct a breast cancer prediction model.

[0100] The functions of the device in the embodiment of the present invention have been described in the above method embodiment, so for details not provided in the description of this embodiment, please refer to the relevant description in the above embodiment, and no further description will be given here.

[0101] Based on the same inventive concept, an embodiment of this specification also provides an electronic device.

[0102] The following describes an electronic device embodiment of the present invention, which can be regarded as a specific physical implementation of the method and device embodiments of the present invention. The details described in the electronic device embodiment of the present invention should be regarded as a supplement to the above method or device embodiments; details not disclosed in the electronic device embodiment of the present invention can be implemented with reference to the above method or device embodiments.

[0103] Figure 5This is a schematic diagram of the structure of an electronic device provided in an embodiment of this specification. Figure 5 The electronic device 300 according to the embodiment of the present invention is described. Figure 5 The electronic device 300 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0104] like Figure 5 As shown, the electronic device 300 is in the form of a general computing device. The components of the electronic device 300 may include, but are not limited to: at least one processing unit 310, at least one storage unit 320, a bus 330 connecting different system components (including the storage unit 320 and the processing unit 310), a display unit 340, etc.

[0105] The storage unit stores program codes, which can be executed by the processing unit 310, so that the processing unit 310 performs the steps according to various exemplary embodiments of the present invention described in the above processing method section of this specification. For example, the processing unit 310 can perform the following steps: Figure 1 Steps shown.

[0106] The storage unit 320 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 3201 and / or a cache memory unit 3202 , and may further include a read-only memory unit (ROM) 3203 .

[0107] The storage unit 320 may also include a program / utility 3204 having a set (at least one) of program modules 3205, such program modules 3205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include the implementation of a network environment.

[0108] Bus 330 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0109] The electronic device 300 may also communicate with one or more external devices 400 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), one or more devices that enable viewers to interact with the electronic device 300, and / or any device that enables the electronic device 300 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed through an input / output (I / O) interface 350. Furthermore, the electronic device 300 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 360. The network adapter 360 may communicate with other modules of the electronic device 300 through the bus 330. It should be understood that although Figure 5 Not shown, other hardware and / or software modules may be used in conjunction with the electronic device 300, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0110] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the exemplary embodiments described in the present invention can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation method of the present invention can be embodied in the form of a software product, which can be stored in a computer-readable storage medium (which can be a CD-ROM, USB flash drive, mobile hard disk, etc.) or on a network, including a number of instructions to enable a computing device (which can be a personal computer, server, or network device, etc.) to execute the above method according to the present invention. When the computer program is executed by a data processing device, the computer-readable medium can implement the above method of the present invention, that is: Figure 1 The method shown.

[0111] Figure 6 A schematic diagram of a computer-readable medium provided in accordance with an embodiment of the present specification.

[0112] accomplish Figure 1The computer program of the method shown can be stored on one or more computer readable media. The computer readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable 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 of the above.

[0113] The computer readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by an instruction execution system, an apparatus, or a device or used in combination with it. The program code contained on the readable storage medium may be transmitted with any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0114] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the viewer computing device, partially on the viewer device, as a stand-alone software package, partially on the viewer computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the viewer computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0115] In summary, the present invention can be implemented in hardware, or in a software module running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that general data processing devices such as microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components in the embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and a computer program product) for executing part or all of the methods described herein. Such a program implementing the present invention can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0116] The specific embodiments described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the present invention is not inherently related to any specific computer, virtual device or electronic device, and various general devices can also implement the present invention. The above description is only a specific 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 principles of the present invention should be included in the protection scope of the present invention.

[0117] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0118] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for constructing a breast cancer prediction model, characterized in that: include: Acquire ultrasound images and their annotations; Segmenting the tumor area in the ultrasound image based on the annotation of the ultrasound image to obtain a lesion segmentation result; Perform conventional imaging omics and deep imaging omics extraction on the lesion segmentation results to obtain key features; The key features are mixed and cascaded to construct a breast cancer prediction model.

2. The method for constructing a breast cancer prediction model according to claim 1, wherein: The ultrasonic images include two-dimensional grayscale ultrasonic images, color Doppler ultrasonic images, ultrasonic elasticity images, and ultrasonic contrast imaging videos.

3. The method for constructing a breast cancer prediction model according to claim 2, wherein: The segmenting of the tumor area in the ultrasound image based on the annotation of the ultrasound image includes: segmenting the lesion segmentation result by using an Encoder-Decoder structure network.

4. The method for constructing a breast cancer prediction model according to claim 3, wherein: The lesion segmentation results are subjected to conventional radiomics extraction to obtain key features, including: Performing high-dimensional feature extraction on the lesion segmentation result to obtain shape features, intensity features, texture features, wavelet transform, and LOG transform; The lesion segmentation results are subjected to vascular morphology and vascular perfusion extraction to obtain arterial richness, blood flow signal classification, vascular velocity, and resistance index.

5. The method for constructing a breast cancer prediction model according to claim 4, wherein: The step of performing deep imaging omics extraction on the lesion segmentation result to obtain key features also includes: performing deep imaging omics extraction based on a VGG16 convolutional neural network and / or a 3D convolutional neural network to obtain key features.

6. The method for constructing a breast cancer prediction model according to claim 5, wherein: The step of mixing the key features and performing a cascade operation to construct a breast cancer prediction model includes: Based on the LASSO regression analysis network, the key features of the conventional radiomics are reduced in dimension and irrelevant information is eliminated; Screening the key features of the conventional radiomics based on statistical test algorithms and machine learning algorithms to determine the highly correlated key features of the conventional radiomics; Screening the key features of the deep imaging omics based on a multimodal image information collaborative network to determine the highly correlated key features of the deep imaging omics; The high-correlation key features of the conventional imaging omics and the high-correlation key features of the deep imaging omics are mixed and cascaded to construct a breast cancer prediction model.

7. A breast cancer prediction model construction device, characterized in that: include: An acquisition module, used for acquiring ultrasound images and annotations thereof; A segmentation module, used to segment the tumor area in the ultrasound image based on the annotation of the ultrasound image to obtain a lesion segmentation result; An extraction module, used for performing conventional imaging omics and deep imaging omics extraction on the lesion segmentation results to obtain key features; A building module is used to mix the key features and perform cascade operations to build a breast cancer prediction model.

8. The breast cancer prediction model construction device according to claim 7, characterized in that: The ultrasonic images include two-dimensional grayscale ultrasonic images, color Doppler ultrasonic images, ultrasonic elasticity images, and ultrasonic contrast imaging videos.

9. An electronic device, wherein: The electronic device includes: processor; and, A memory storing computer executable instructions which, when executed, cause the processor to perform the method according to any one of claims 1-6.

10. A computer-readable storage medium, wherein: The computer-readable storage medium stores one or more programs, and when the one or more programs are executed by a processor, the method of any one of claims 1 to 6 is implemented.

Citation Information

Cited By

  • Intelligent focus puncture positioning method and device for ultrasonic image of remote fibular ligament

    CN119970169A

  • An intelligent puncture positioning method and device for anterior talofibular ligament lesion based on ultrasound imaging

    CN119970169B