Breast detector and detection system thereof

Abstract

The invention relates to the technical field of medical instruments, and particularly discloses a mammary gland detector and a detection system thereof, and the mammary gland detector comprises a near-infrared light source module which is used for transmitting and receiving a near-infrared light signal penetrating through a target mammary gland tissue, and converting the received near-infrared light signal into an electric signal; the multi-modal data processing module is used for carrying out real-time analysis on the converted electric signals based on various algorithm models to obtain abnormal physiological function change detection information of the mammary gland, and extracting additional physiological function information and blood vessel heat distribution characteristics of an organism; the imaging display module is used for obtaining and displaying an effective imaging result based on an imaging instruction input by a user, the mammary gland abnormal physiological function change detection information obtained in real time, the additional physiological function information of the body and the blood vessel heat distribution characteristics; the mammary gland detector can efficiently and accurately detect mammary gland abnormity, clear and visual detection results are provided for a user, and early discovery of mammary gland problems is facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and particularly relates to a breast detector and its detection system. Background Art

[0002] With the improvement of people's health awareness and the increase in the incidence of breast diseases, the demand for breast detection technology is increasing day by day. At present, there are various technical means in the field of breast detection: Mammography: This is a widely used traditional detection method. By irradiating the breast with X-rays and using the differences in the absorption of X-rays by different tissues to form an image, it can detect lesions such as calcifications and masses in the breast. It plays an important role in the early screening of breast diseases, especially for detecting microcalcifications in the breast with high sensitivity. Ultrasonic detection technology: Utilizes the propagation characteristics of ultrasonic waves in breast tissue and constructs the structural image of the breast by receiving the reflected waves. This technology can clearly show the hierarchical structure of the breast, breast ducts, and breast lobules, etc., and has high accuracy in differentiating the cystic or solid nature of breast masses. It is also simple to operate and radiation-free, making it suitable for breast examinations of women of all ages. Magnetic resonance imaging (MRI) technology: Based on the principle of magnetic resonance, it performs multi-plane and multi-sequence imaging of the breast. MRI has extremely high resolution for breast soft tissues and can detect some micro-lesions that are difficult to detect by other detection means. It is particularly valuable for the qualitative diagnosis of breast diseases and is often used for further detailed evaluation of breast diseases. Near-infrared breast detection technology: Based on the absorption characteristics of near-infrared light by human tissues, it uses near-infrared light to penetrate breast tissue and obtains internal breast information by detecting the near-infrared light signal after passing through the breast tissue. This technology has the advantages of being radiation-free, simple to operate, and relatively low in price, and has also been applied to a certain extent in the detection of breast diseases.

[0003] Although existing breast detection technologies have their respective advantages, there are still some deficiencies: Most existing technologies can only obtain certain aspects of breast information. For example, mammography mainly focuses on detecting morphological abnormalities such as masses and calcifications; ultrasound detection mainly focuses on the structural characteristics of the breast; while MRI provides relatively detailed soft tissue images, but still focuses on the structural aspect. These single detection information is difficult to comprehensively reflect the physiological and pathological states of the breast, which may lead to misdiagnosis or missed diagnosis of some complex breast diseases. Insufficient acquisition of functional information: Most current detection technologies have limited acquisition of information on the physiological function changes of the breast. For example, it is impossible to monitor in real time the physiological function changes such as breast tissue metabolism and blood perfusion, and these functional information is of great significance for early detection of the potential risks of breast diseases and assessment of the disease development process. Lack of multi-modal data fusion analysis: The data obtained by different detection technologies are independent of each other and lack effective integration and analysis. Clinically, doctors often need to interpret the results obtained by different detection methods separately and then make a comprehensive judgment based on experience. This not only increases the workload of doctors, but also due to the lack of standardized multi-modal data fusion analysis methods, it is easily affected by subjective factors, resulting in a certain degree of restriction on the accuracy and reliability of the diagnosis results.

[0004] Therefore, the present invention proposes a breast detector and its detection system. Summary of the Invention

[0005] The present invention provides a breast detector and its detection system, including a near-infrared light source module that can emit and receive near-infrared light signals and convert them into electrical signals to provide basic data for detection. The multi-modal data processing module analyzes the electrical signals in real time through a variety of algorithm models, can accurately obtain detection information on abnormal physiological function changes in the breast, and can also extract additional physiological function and vascular thermal distribution characteristics, improving the comprehensiveness and accuracy of detection. This analysis method of multiple algorithm models can integrate a variety of information, not only limited to traditional structural information, but also can explore the physiological function information of the breast, realizing the fusion processing of multi-modal data and overcoming the problem of single detection information in the prior art. This enables the detector to more comprehensively reflect the physiological state of the breast, makes up for the deficiency of the prior art in obtaining functional information, helps to early detect the potential risks of breast diseases and accurately evaluate the disease development process. The imaging display module generates and displays an effective imaging result based on the imaging instruction input by the user and a variety of information obtained by the multi-modal data processing module. This imaging method can present complex detection data in an intuitive and easy-to-understand image form to doctors, providing more comprehensive and accurate diagnostic basis for doctors, overcoming the problems of the results of the prior imaging technology being not intuitive and comprehensive enough, improving the accuracy and efficiency of diagnosis, and helping to make clinical treatment decisions. This breast detector can efficiently and accurately detect breast abnormalities, provide clear and intuitive detection results for users, and help to early detect breast problems.

[0006] The present invention provides a breast detector, including:

[0007] A near-infrared light source module, configured to emit and receive the near-infrared light signal after passing through the target breast tissue, and convert the received near-infrared light signal into an electrical signal;

[0008] A multi-modal data processing module, configured to perform real-time analysis on the converted electrical signal based on a variety of algorithm models, obtain detection information on abnormal physiological function changes in the breast, and extract additional physiological function information of the body and vascular thermal distribution characteristics;

[0009] An imaging display module, configured to obtain and display an effective imaging result based on the imaging instruction input by the user, the detection information on abnormal physiological function changes in the breast obtained in real time, and the additional physiological function information of the body and vascular thermal distribution characteristics.

[0010] Preferably, the near-infrared light source module includes:

[0011] A near-infrared light signal emitting sub-module, configured to emit a near-infrared light signal within a specific wavelength range based on a preset power density;

[0012] A near-infrared light signal receiving sub-module, configured to receive the near-infrared light signal after passing through the target breast tissue based on a high-sensitivity photoelectric sensor array;

[0013] A near-infrared light signal conversion sub-module for converting the received near-infrared light signal after passing through the target breast tissue into an electrical signal.

[0014] Preferably, the multi-modal data processing module includes:

[0015] A physiological function analysis sub-module for analyzing the converted electrical signal based on a physiological function analysis algorithm model to obtain breast abnormal physiological function change detection information;

[0016] An additional physiological function information analysis sub-module for analyzing the spectral characteristics of the converted electrical signal, inferring the periodic change information of the metabolic state and hemodynamic information of the target breast tissue, and obtaining additional physiological function information of the body based on the periodic change information of the metabolic state and hemodynamic information of the target breast tissue;

[0017] A blood vessel distribution analysis sub-module for extracting the blood vessel thermal distribution characteristics in the target breast tissue from the converted electrical signal based on a blood vessel thermal distribution analysis model.

[0018] Preferably, the physiological function analysis sub-module includes:

[0019] A breast-related data collection sub-module for collecting a large amount of breast-related data from different individuals covering normal breast tissue states and various breast diseases, as well as basic research data related to breast physiological functions, as target research data;

[0020] A typical feature extraction sub-module for extracting all breast physiological function features from the target research data and screening out multiple typical breast physiological function features from all breast physiological function features, where the multiple typical breast physiological function features include multiple typical normal breast physiological function features and multiple typical abnormal breast physiological function features;

[0021] A physiological function analysis algorithm model construction sub-module for screening out the training data of each typical breast physiological function feature from the target research data and constructing a physiological function analysis algorithm model based on the training data of each typical breast physiological function feature;

[0022] An electrical signal analysis sub-module for analyzing the converted electrical signal based on the physiological function analysis algorithm model to obtain breast abnormal physiological function change detection information.

[0023] Preferably, the typical feature extraction sub-module includes:

[0024] A feature centering unit, which is used to extract all breast physiological function features from all samples in the target research data, center all breast physiological function features in each sample, and obtain all centered breast physiological function features in all samples;

[0025] An eigenvalue decomposition unit, which is used to generate a breast physiological function feature matrix based on all centered breast physiological function features in all samples, calculate the covariance matrix of the breast physiological function feature matrix, and then perform eigenvalue decomposition on the covariance matrix of the breast physiological function feature matrix to obtain all eigenvalues of the covariance matrix;

[0026] A decision tree generation unit, which is used to generate multiple decision trees based on a random forest and all breast physiological function features in all samples;

[0027] An importance score calculation unit, which is used to calculate the importance score of each breast physiological function feature based on all decision trees;

[0028] A typical feature screening unit, which is used to screen out multiple typical breast physiological function features from all breast physiological function features based on all eigenvalues and the importance score of each breast physiological function feature in each sample.

[0029] Preferably, the importance score calculation unit includes:

[0030] A decision tree parameter acquisition subunit, which is used to determine the total number of all decision trees, the total number of nodes in the decision trees, and the number of samples included after splitting each node using each breast physiological function feature;

[0031] An importance score subunit, which is used to calculate the importance score of each breast physiological function feature based on the total number of all decision trees, the total number of nodes in the decision trees, and the number of samples included in each node.

[0032] Preferably, the typical feature screening unit includes:

[0033] An eigenvalue sorting subunit, which is used to sort all eigenvalues of the covariance matrix from smallest to largest to obtain an eigenvalue sequence, and disassemble the eigenvalue sequence to obtain a sub-eigenvalue sequence corresponding to each eigenvalue ordinal number;

[0034] An accumulated contribution rate calculation subunit, which is used to regard the quotient of the mean value of all eigenvalues included in the sub-eigenvalue sequence corresponding to each eigenvalue ordinal number and the mean value of all eigenvalues of the covariance matrix as the accumulated contribution rate corresponding to each eigenvalue ordinal number;

[0035] A typical feature screening subunit, configured to screen out multiple typical breast physiological function features from all breast physiological function features based on the cumulative contribution rate corresponding to each eigenvalue ordinal number and the importance score of each breast physiological function feature in each sample.

[0036] Preferably, the electrical signal analysis sub-module includes:

[0037] A feature extraction unit, configured to analyze the converted electrical signals based on a physiological function analysis algorithm model to obtain the breast physiological function features of the target breast tissue;

[0038] An abnormal physiological function change judgment unit, configured to judge whether there is an abnormal physiological function change in the target breast tissue based on the breast physiological function features of the target breast tissue;

[0039] A first breast abnormality detection information output unit, configured to, if it is determined that there is an abnormal physiological function change in the target breast tissue, determine the type and severity of the abnormal physiological function change as breast abnormal physiological function change detection information;

[0040] A second breast abnormality detection information output unit, configured to, if it is determined that there is no abnormal physiological function change in the target breast tissue, use the fact that there is no abnormal physiological function change in the target breast tissue as breast abnormal physiological function change detection information.

[0041] Preferably, the imaging display module includes:

[0042] A dynamic imaging sub-module, configured to generate an initial dynamic imaging result based on an optimized image reconstruction algorithm, in combination with the breast abnormal physiological function change detection information obtained in real time, the additional physiological function information of the body, and the vascular thermal distribution characteristics;

[0043] A photographing and displaying sub-module, configured to generate and display an effective imaging result based on an imaging instruction input by a user and the initial dynamic imaging result.

[0044] The present invention provides a breast detection system, including any one of the above breast detectors and an adapter matching any one of the above breast detectors.

[0045] The beneficial effects of the present invention compared with the prior art are as follows: The near-infrared light source module can emit and receive near-infrared light signals and convert them into electrical signals, providing basic data for detection. The multi-modal data processing module analyzes the electrical signals in real time through various algorithm models, can accurately obtain the detection information of abnormal physiological function changes in the breast, and can also extract additional physiological function and vascular heat distribution characteristics, improving the comprehensiveness and accuracy of detection. This analysis method of multiple algorithm models can fuse various information, not only limited to traditional structural information, but also can explore the physiological function information of the breast, realizing the fusion processing of multi-modal data and overcoming the problem of single detection information in the prior art. This enables the detector to more comprehensively reflect the physiological state of the breast, makes up for the deficiency of the prior art in obtaining functional information, and helps to early detect the potential risks of breast diseases and accurately evaluate the disease development process. The imaging display module generates and displays effective imaging results based on the imaging instructions input by the user and various information obtained by the multi-modal data processing module. This imaging method can present complex detection data in an intuitive and easy-to-understand image form to doctors, providing more comprehensive and accurate diagnostic basis for doctors, overcoming the problems of insufficient intuitiveness and comprehensiveness of the results of existing imaging technologies, improving the accuracy and efficiency of diagnosis, and helping to make clinical treatment decisions. This breast detector can efficiently and accurately detect breast abnormalities, provide clear and intuitive detection results for users, and help to early detect breast problems.

[0046] Other features and advantages of the present invention will be described in the following specification, and, in part, will be obvious from the specification, or can be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in this application document.

[0047] The technical solutions of the present invention will be further described in detail below through the drawings and embodiments. Description of the Drawings

[0048] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention, and do not constitute a limitation to the present invention. In the drawings:

[0049] Figure 1 is a schematic diagram of the internal functional modules of the breast detector in the embodiment of the present invention;

[0050] Figure 2 is a schematic diagram of the execution logic of the near-infrared light source module in the embodiment of the present invention;

[0051] Figure 3 is a schematic diagram of the execution logic of the multi-modal data processing module in the embodiment of the present invention. Detailed Embodiments

[0052] The preferred embodiments of the present invention will be described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only for the purpose of illustrating and explaining the present invention, and are not intended to limit the present invention.

[0053] Embodiment 1:

[0054] The present invention provides a breast detector, referring to Figure 1 , including:

[0055] A near-infrared light source module for emitting and receiving the near-infrared light signal transmitted through the target breast tissue and converting the received near-infrared light signal into an electrical signal;

[0056] A multimodal data processing module for performing real-time analysis on the converted electrical signal based on a variety of algorithm models to obtain breast abnormal physiological function change detection information, and extracting additional physiological function information of the body and vascular thermal distribution characteristics;

[0057] An imaging display module for obtaining an effective imaging result and displaying it based on the imaging instruction input by the user, the breast abnormal physiological function change detection information obtained in real time, and the additional physiological function information of the body and vascular thermal distribution characteristics.

[0058] In this embodiment, the target breast tissue refers to the specific breast tissue that needs to be detected and analyzed, such as the unilateral or bilateral breast of a certain patient undergoing detection.

[0059] In this embodiment, the received near-infrared light signal is converted into an electrical signal. The near-infrared light is received after passing through the breast tissue, and through a specific conversion device or mechanism, this light signal is converted into an electrical signal for subsequent processing and analysis. For example, the intensity change of light is converted into the change of current or voltage through a photoelectric converter.

[0060] In this embodiment, a variety of algorithm models refer to multiple different mathematical calculation methods and modes used for processing and analyzing detection data. For example, a physiological function analysis algorithm model, a vascular thermal distribution analysis model, etc.

[0061] In this embodiment, the breast abnormal physiological function change detection information is the relevant information about the abnormal changes in the breast physiological function obtained through the detection and analysis of the breast tissue, including the type, degree, etc. of the abnormality. For example, the specific situation of abnormal metabolism or abnormal blood perfusion in the breast tissue is detected.

[0062] In this embodiment, the additional physiological function information of the body includes, in addition to the physiological function information of the breast itself, the information of other body functions related to the breast, such as the metabolic state and hemodynamic information of the breast tissue.

[0063] In this embodiment, the vascular thermal distribution feature describes the characteristics and patterns of the heat distribution of blood vessels within breast tissue. For example, in some areas where blood vessels are dense, the heat is higher, while in some areas where blood vessels are sparse, the heat is lower.

[0064] In this embodiment, the imaging instructions input by the user are specific requirements and commands for generating and displaying images issued by the person using this breast detector. For example, it is required to display imaging results at a specific angle, specific part, or of a specific type.

[0065] In this embodiment, the effective imaging result is to display the processed and generated image result that can accurately reflect the information related to breast detection, and it is displayed on the corresponding device for the user to view and analyze.

[0066] The beneficial effects of the above technologies are as follows: The near-infrared light source module can emit and receive near-infrared light signals and convert them into electrical signals, providing basic data for detection. The multi-modal data processing module analyzes the electrical signals in real time through multiple algorithm models, can accurately obtain the detection information of abnormal physiological function changes in the breast, and can also extract additional physiological functions and vascular thermal distribution features, improving the comprehensiveness and accuracy of detection. This analysis method of multiple algorithm models can integrate multiple types of information, not only limited to traditional structural information, but also can explore the physiological function information of the breast, realizing the fusion processing of multi-modal data and overcoming the problem of single detection information in the existing technology. This enables the detector to more comprehensively reflect the physiological state of the breast, makes up for the deficiency in obtaining functional information in the existing technology, helps to early detect the potential risks of breast diseases and accurately evaluate the disease development process. The imaging display module generates and displays an effective imaging result based on the imaging instructions input by the user and the various information obtained by the multi-modal data processing module. This imaging method can present complex detection data in an intuitive and easy-to-understand image form to doctors, providing more comprehensive and accurate diagnostic basis for doctors, overcoming the problems of insufficient intuitiveness and comprehensiveness of the results of existing imaging technologies, improving the accuracy and efficiency of diagnosis, and helping to make clinical treatment decisions. This breast detector can efficiently and accurately detect breast abnormalities, provide clear and intuitive detection results for users, and help to early detect breast problems.

[0067] Embodiment 2:

[0068] Based on Embodiment 1, the near-infrared light source module, refer to Figure 2 , includes:

[0069] The near-infrared light signal emission sub-module is used to emit near-infrared light signals within a specific wavelength range based on a preset power density, so that the near-infrared light signals can penetrate into the skin to a depth of 30 - 100 mm;

[0070] A near-infrared light signal receiving sub-module, which is used to receive the near-infrared light signal transmitted through the target breast tissue based on a high-sensitivity optoelectronic sensor array;

[0071] A near-infrared light signal conversion sub-module, which is used to convert the received near-infrared light signal transmitted through the target breast tissue into an electrical signal.

[0072] In this embodiment, the preset power density refers to the energy radiation amount per unit area that is preset. For example, in a breast detector, the preset power density is 5 milliwatts per square centimeter.

[0073] In this embodiment, the specific wavelength range is a predetermined wavelength interval. For example, in near-infrared light detection, the specific wavelength range may be from 700 nanometers to 900 nanometers.

[0074] In this embodiment, the high-sensitivity optoelectronic sensor array is a device composed of multiple optoelectronic sensors with high-sensitivity detection capabilities arranged. For example, an array composed of 10 optoelectronic sensors in a row that can accurately detect weak light signals.

[0075] The beneficial effects of the above technical solutions are as follows: The near-infrared light signal emitting sub-module can emit near-infrared light signals within a specific wavelength range according to the preset power density, ensuring the stability and pertinence of the detection light source. The near-infrared light signal receiving sub-module uses a high-sensitivity optoelectronic sensor array to receive light signals, improving the accuracy and reliability of signal reception. The near-infrared light signal conversion sub-module effectively converts the received light signal into an electrical signal, providing convenience for subsequent data processing and analysis. Each sub-module of the near-infrared light source module works in coordination, and can provide high-quality detection signals for the breast detector, improving the accuracy and effectiveness of detection.

[0076] Embodiment 3:

[0077] Based on Embodiment 1, the multi-modal data processing module, referring to Figure 3 , includes:

[0078] A physiological function analysis sub-module, which is used to analyze the converted electrical signal based on a physiological function analysis algorithm model to obtain breast abnormal physiological function change detection information;

[0079] An additional physiological function information analysis sub-module, which is used to analyze the spectral characteristics of the converted electrical signal, infer the periodic change information of the metabolic state and hemodynamic information of the target breast tissue, and obtain additional physiological function information of the body based on the periodic change information of the metabolic state and hemodynamic information of the target breast tissue;

[0080] A blood vessel distribution analysis sub-module, which is used to extract the blood vessel thermal distribution characteristics in the target breast tissue from the converted electrical signal based on a blood vessel thermal distribution analysis model.

[0081] In this embodiment, the physiological function analysis algorithm model is a mathematical calculation model specifically used to analyze breast physiological function, which performs calculation and analysis by inputting relevant data, such as a neural network model trained based on a large amount of breast physiological function data.

[0082] In this embodiment, the converted electrical signal is analyzed based on the physiological function analysis algorithm model to obtain abnormal breast physiological function change detection information, and the electrical signal is processed using a specific algorithm model to determine whether the breast physiological function is abnormal and the specific situation of the abnormality. For example, abnormal blood oxygen supply to breast tissue is found through analysis.

[0083] In this embodiment, the spectral characteristics of the converted electrical signal are analyzed to infer the metabolic state of the target breast tissue and the periodic change information of the hemodynamic information. The frequency distribution characteristics of the electrical signal are studied to infer the metabolism of the breast tissue and the periodic law of blood flow. For example, the metabolic rate is accelerated or the periodic fluctuation of the blood flow rate is judged according to the spectral characteristics. First, the spectral analysis is performed on the converted electrical signal. The spectral characteristics reflect the energy distribution of the electrical signal at different frequencies. Then, based on these spectral characteristics, the metabolic state of the target breast tissue and the periodic changes of the hemodynamic information are inferred. In breast tissue, changes in metabolic state and changes in hemodynamics will affect the absorption and scattering of near-infrared light, which are reflected in the converted electrical signal. For example, if the spectrum of the electrical signal shows regular intensity changes in certain specific frequency bands, it may mean that the metabolic activity of the breast tissue is periodically accelerated or slowed down. For hemodynamics, the periodic fluctuations in the spectral characteristics may be related to the periodic changes in the flow rate and flow of blood. Through careful analysis and interpretation of these spectral features, we can infer the periodic changes in metabolism and blood flow of breast tissue, providing an important basis for evaluating the physiological function of the breast.

[0084] In this embodiment, the metabolic state of the target breast tissue refers to the chemical processes such as synthesis and decomposition of substances in the breast tissue, such as the consumption of nutrients and the rate of product generation. For example, vigorous metabolism means active cell activity, and slow metabolism may mean functional impairment.

[0085] In this embodiment, the hemodynamic information is data related to the flow of blood in breast tissue, including blood flow velocity, blood volume, vascular pressure, etc. For example, if a faster blood flow velocity is detected in a certain area, it may indicate that the metabolism of the area is active.

[0086] In this embodiment, additional physiological function information of the body is obtained based on the periodic change information of the metabolic state and hemodynamic information of the target breast tissue. According to the periodic changes in breast tissue metabolism and blood flow, other related physiological function conditions are further understood. For example, the hormone secretion level is inferred based on the metabolism and blood flow conditions.

[0087] In this embodiment, the vascular thermal distribution characteristics in the target breast tissue are extracted from the converted electrical signals based on the vascular thermal distribution analysis model. The characteristics of the vascular heat distribution in the breast tissue are found from the electrical signals using a special model. For example, it is determined that the temperature in the area with dense blood vessels is higher and the temperature in the sparse area is lower.

[0088] In this embodiment, the periodic change information of the metabolic state and hemodynamic information of the target breast tissue refers to the regular change characteristics of breast tissue metabolism and blood flow over time. For example, during certain stages within a physiological cycle, metabolism is enhanced and blood flow is accelerated, while in other stages, it is the opposite.

[0089] The beneficial effects of the above technical solutions are as follows: The physiological function analysis sub-module analyzes the electrical signals through a specific algorithm model and can accurately obtain the detection information of abnormal physiological function changes in the breast, which helps to timely detect functional abnormalities in the breast. The additional physiological function information analysis sub-module obtains the periodic change information such as the metabolic state and hemodynamics by analyzing the spectral characteristics of the electrical signals, enriching the dimension and depth of the detection. The vascular distribution analysis sub-module extracts the vascular thermal distribution characteristics based on a special model, providing an important basis for comprehensively understanding the health status of breast tissue. The sub-modules of the multi-modal data processing module cooperate with each other and can deeply analyze the electrical signals from multiple angles, providing more comprehensive and accurate information for breast detection.

[0090] Embodiment 4:

[0091] Based on Embodiment 3, the physiological function analysis sub-module includes:

[0092] The breast-related data collection sub-module is used to collect a large amount of breast-related data from different individuals, covering normal breast tissue states and various breast diseases, as well as basic research data related to breast physiological functions, as the target research data;

[0093] The typical feature extraction sub-module is used to extract all breast physiological function features from the target research data and screen out multiple typical breast physiological function features from all breast physiological function features. Among them, multiple typical breast physiological function features include multiple typical normal breast physiological function features and multiple typical abnormal breast physiological function features;

[0094] A sub-module for building a physiological function analysis algorithm model, which is used to screen out training data for each typical breast physiological function feature from the target research data, and build a physiological function analysis algorithm model based on the training data for each typical breast physiological function feature;

[0095] An electrical signal analysis sub-module, which is used to analyze the converted electrical signals based on the physiological function analysis algorithm model to obtain breast abnormal physiological function change detection information.

[0096] In this embodiment, the normal breast tissue state refers to the tissue morphology, structure, and physiological activities of the breast when it is disease-free and has normal functions. For example, the normal arrangement of breast cells, normal blood supply, and metabolic level, etc.

[0097] In this embodiment, various breast diseases cover a variety of breast disorders with different types and manifestations, such as mastitis, breast hyperplasia, breast fibroids, breast cancer, etc.

[0098] In this embodiment, breast-related data refers to various information and values related to the breast, including but not limited to breast structure data, physiological function data, disease-related data, etc. For example, the size and morphological measurement values of the breast, biochemical indicators of breast tissue, etc.

[0099] In this embodiment, the basic research data related to breast physiological functions is the data obtained from research on the principles, mechanisms, influencing factors, etc. of normal breast physiological functions. For example, research data on the metabolic mechanism of breast cells, research data on the influence of hormones on breast physiological functions, etc.

[0100] In this embodiment, all breast physiological function features are extracted from the target research data, and various characteristics and indicators that can reflect breast physiological functions are found from the existing research data. For example, the blood oxygen content of the breast, the metabolic rate of nutrients, etc.

[0101] In this embodiment, typical breast physiological function features are the characteristics of breast physiological functions that are representative and can significantly distinguish normal and abnormal physiological functions.

[0102] In this embodiment, typical normal breast physiological function features represent the common and obvious physiological function characteristics of normal breast tissue. For example, the stable metabolic rate and uniform blood distribution of normal breast tissue, etc.

[0103] In this embodiment, typical abnormal breast physiological function features reflect the significant physiological function change characteristics when the breast is abnormal. For example, the abnormal increase in blood supply and abnormal active metabolism in the breast tumor area, etc.

[0104] In this embodiment, training data for each typical breast physiological function feature is screened out from the target research data, and a data set for training a model to identify each typical breast physiological function feature is selected from a large amount of research data. The implementation process includes the following steps: First, it is necessary to clarify the specific definitions and criteria of the typical breast physiological function features to be screened. These are feature descriptions with significant discriminability based on previous research results, expert experience, or obtained through data analysis. Then, a comprehensive analysis and processing of the target research data are carried out. This involves cleaning and preprocessing the data to remove noise and outliers and ensure the quality and accuracy of the data. Next, according to the defined feature criteria, data screening and classification are performed using data mining, machine learning, or statistical methods. For example, feature engineering techniques can be used to extract variables and indicators related to typical breast physiological function features from the original data. For the preliminary data screened out, further evaluation and verification are required. This can be done through cross-validation, comparison with known sample data, etc., to ensure that the screened data can effectively represent each typical breast physiological function feature. During the screening process, the balance of the data also needs to be considered, that is, to ensure that there are sufficient and balanced training data for different types of typical features to avoid bias in the model training process. Finally, after multiple screenings and optimizations, the final training data set for each typical breast physiological function feature used for training is obtained.

[0105] In this embodiment, a physiological function analysis algorithm model is built based on the training data of each typical breast physiological function characteristic. Using the selected specific training data, an algorithm model capable of analyzing breast physiological functions is constructed through mathematical methods and calculation processes. The implementation process is roughly as follows: First, a suitable algorithm model architecture is selected. This may be common machine learning algorithms such as decision trees, neural networks, support vector machines, or combinations and improvements thereof. Then, the training data of each selected typical breast physiological function characteristic is input into the selected algorithm model. During the training process, the algorithm automatically adjusts the parameters of the model according to the input data to minimize the error between the predicted result and the actual result. This is usually achieved through optimization algorithms such as gradient descent. For complex models, multiple iterative trainings may be required to continuously adjust the parameters to improve the performance and accuracy of the model. During the training process, some techniques are also used to prevent overfitting, such as regularization and early stopping. Regularization penalizes the complexity of the model to avoid the model being too complex and only memorizing the training data; early stopping stops the training when the performance of the model on the validation set begins to decline. At the same time, techniques such as cross-validation are used to evaluate the performance of the model on different subsets of data to ensure that the model has good generalization ability, that is, it can make accurate predictions on new and unseen data. After multiple trainings and optimizations, an algorithm model that can accurately analyze and predict based on the input breast physiological function characteristic data is obtained. Finally, the built model also needs to be finally evaluated and verified. For example, an independent test data set is used to test indicators such as the accuracy, sensitivity, and specificity of the model to determine whether the model meets the requirements of practical applications. If not, it may be necessary to readjust the model architecture or optimize the training process again.

[0106] The beneficial effects of the above technical solutions are as follows: The breast-related data collection sub-module collects a large amount of comprehensive data, providing rich materials for subsequent analysis and ensuring the reliability of the analysis results. The typical feature extraction sub-module screens out a variety of typical breast physiological function characteristics, making the analysis more targeted and effective. The physiological function analysis algorithm model building sub-module builds a model based on the training data of typical features, improving the accuracy and adaptability of the model. The electrical signal analysis sub-module uses the built model for analysis and can accurately obtain detection information on abnormal physiological function changes in the breast. The physiological function analysis sub-module can provide accurate and valuable physiological function detection results for the breast detector through systematic data collection, feature extraction, model building, and electrical signal analysis.

[0107] Embodiment 5:

[0108] Based on Embodiment 4, the typical feature extraction sub-module includes:

[0109] A feature centering unit, configured to extract all breast physiological function features from all samples in the target research data, center all breast physiological function features in each sample, and obtain all centered breast physiological function features in all samples;

[0110] An eigenvalue decomposition unit, configured to generate a breast physiological function feature matrix based on all centered breast physiological function features in all samples, calculate the covariance matrix of the breast physiological function feature matrix, and then perform eigenvalue decomposition on the covariance matrix of the breast physiological function feature matrix to obtain all eigenvalues of the covariance matrix;

[0111] A decision tree generation unit, configured to generate multiple decision trees based on random forests and all breast physiological function features in all samples;

[0112] An importance score calculation unit, configured to calculate the importance score of each breast physiological function feature based on all decision trees;

[0113] A typical feature screening unit, configured to screen out multiple typical breast physiological function features from all breast physiological function features based on all eigenvalues and the importance scores of each breast physiological function feature in each sample.

[0114] In this embodiment, all breast physiological function features are extracted from all samples in the target research data. For all samples covered in the target research data, various characteristics and indicators that can reflect breast physiological functions are found. For example, features such as changes in blood oxygen content and metabolic rate are extracted from breast detection data samples of multiple patients. Or feature extraction based on medical images: texture features, such as gray-level co-occurrence matrix (GLCM) features, are extracted from breast ultrasound images. Or feature extraction based on physiological electrical signals: for breast physiological electrical signals, time-domain features such as mean and variance are extracted. The mean and variance can reflect the average level and fluctuation degree of the electrical signal, which are related to the physiological electrical activities of breast tissues. Or feature extraction based on metabolic data: if there is data on the concentration of breast tissue metabolites, such as data on the change in glucose concentration over time, the metabolic rate can be calculated, and the metabolic rate reflects the metabolic function characteristics of breast tissues.

[0115] In this embodiment, the samples in the target research data refer to the data sets of individuals or cases used for research. For example, breast-related detection data of women of different ages and different health conditions.

[0116] In this embodiment, all breast physiological function features in each sample are centralized to obtain all centralized breast physiological function features in all samples: that is, the value of each breast physiological function feature in each sample is subtracted by the mean value of this breast physiological function feature in all samples to obtain the centralized feature. For example, if the value of a certain feature in a sample is originally 10 and the mean value of this feature in the sample is 5, the value of this feature becomes 5 after centralization.

[0117] In this embodiment, a breast physiological function feature matrix is generated based on all centralized breast physiological function features in all samples. The centralized breast physiological function features of all samples are arranged in a matrix form according to certain rules. Suppose there are n samples and each sample has p features, forming a data matrix X n×p 。

[0118] In this embodiment, the covariance matrix of the breast physiological function feature matrix is calculated. Through mathematical operations, a matrix composed of the covariances between various features in this feature matrix is obtained. It reflects the degree of linear relationship between different features. The formula is:

[0119] In this embodiment, the covariance matrix S of the breast physiological function feature matrix is subjected to eigenvalue decomposition to obtain all eigenvalues of the covariance matrix. The covariance matrix is decomposed into eigenvalues and eigenvectors using mathematical methods. The obtained eigenvalues reflect the variance magnitude of the data in the direction of the corresponding eigenvectors. The formula is: S = V∧V T , where ∧ is a diagonal matrix and the diagonal elements are λ 1 ≥λ 2 ≥…≥λ P , and V is an eigenvector matrix.

[0120] In this embodiment, multiple decision trees are generated based on random forests and all breast physiological function features in all samples. Using the random forest algorithm, combined with all breast physiological function features of all samples, multiple decision trees for classification and prediction are constructed. The implementation process is roughly as follows: First, prepare the data. Organize and preprocess all breast physiological function feature data in all samples to ensure that the data quality and format meet the requirements. Then, randomly select features. When constructing each decision tree, randomly select a part of the features from all breast physiological function features as candidate features for splitting nodes. Next, randomly sample the samples. For the construction of each decision tree, instead of using all samples, randomly select a part of the samples (usually sampling with replacement) from all samples as the training set for this decision tree. Select the splitting node. For each node, according to the selected candidate features, calculate the information gain or other evaluation metrics of each feature, select the optimal feature and its splitting point for node splitting, and divide the samples into different subsets. Repeat the splitting process. Continuously repeat the above steps until the stopping condition is met, such as reaching the preset maximum depth, too few samples in the node, etc. Construct multiple decision trees. Construct multiple different decision trees according to the above steps, and each decision tree is constructed based on randomly selected features and samples. During the construction of the decision tree, since the selection of features and samples is random, different decision trees will be different, and finally a random forest is formed. This random forest can be used to classify and predict new breast physiological function feature data, and combining the results of multiple decision trees can improve the accuracy and stability of prediction.

[0121] In this embodiment, the importance score of each breast physiological function feature is given a numerical value representing its importance in analysis and judgment through a specific calculation method. The higher the score, the greater the impact of the feature on the result.

[0122] The beneficial effects of the above technical solutions are as follows: The feature centering unit centralizes the breast physiological function features, eliminates data bias, and improves data quality and stability. The eigenvalue decomposition unit provides an important basis and foundation for subsequent feature screening by calculating the eigenvalues of the covariance matrix. The decision tree generation unit uses random forests to generate multiple decision trees, which can comprehensively evaluate the importance of each feature. The importance score calculation unit calculates the importance score of each feature, making feature screening more scientific and accurate. The typical feature screening unit comprehensively screens out typical features based on eigenvalues and importance scores, improving the representativeness and effectiveness of features. The typical feature extraction sub-module can accurately screen out representative typical breast physiological function features through a series of rigorous processing steps, providing strong support for subsequent physiological function analysis and detection.

[0123] Embodiment 6:

[0124] Based on Example 5, the importance score calculation unit includes:

[0125] A decision tree parameter acquisition subunit, configured to determine the total number of all decision trees, the total number of nodes in the decision trees, and the number of samples included after splitting each node using each breast physiological function feature;

[0126] An importance score subunit, configured to calculate the importance score of each breast physiological function feature based on the total number of all decision trees, the total number of nodes in the decision trees, and the number of samples included in each node:

[0127]

[0128] Where I j is the importance score of the j-th breast physiological function feature, T is the total number of all decision trees, n t is the total number of nodes in the t-th decision tree, N tj (i) is the number of samples included after splitting the i-th node in the t-th decision tree using the j-th breast physiological function feature, N tj is the number of samples included after splitting the root node of the t-th decision tree using the j-th breast physiological function feature, H(p) is the information entropy function and H(p) = -p lg(p) - (1 - p) lg(1 - p), where p is a probability value and p takes or N tj (i left ) is the number of samples included in the left child node after splitting the i-th node in the t-th decision tree using the j-th breast physiological function feature, N tj (i right ) is the number of samples included in the right child node after splitting the i-th node in the t-th decision tree using the j-th breast physiological function feature.

[0129] In this embodiment, determining the total number of all decision trees, the total number of nodes in the decision trees, and the number of samples included after splitting each node using each breast physiological function feature means that a comprehensive statistics and analysis need to be performed on the decision trees in the constructed random forest.

[0130] Among them, the total number of all decision trees is the number of decision trees included in the random forest. For example, if 100 decision trees are constructed, then the total number is 100;

[0131] The total number of nodes in the decision trees refers to the sum of the nodes in all decision trees. Nodes include root nodes, internal nodes, and leaf nodes;

[0132] Among them, the number of samples included in each node after splitting using each breast physiological function feature is the number of samples included in each child node after splitting, which is statistically counted for each breast physiological function feature every time it is used for node splitting. Through such statistics, the partitioning of samples by each feature during different node splits can be understood, thereby evaluating the importance of the feature and the structural characteristics of the decision tree.

[0133] The beneficial effects of the above technical solution are as follows: The decision tree parameter acquisition subunit clarifies the key parameters of the decision tree, providing necessary data support for calculating the importance score. The importance score subunit comprehensively considers factors such as the total number of decision trees, the total number of nodes, and the number of samples in each node through a specific calculation formula, and can accurately quantify the importance of each breast physiological function feature. This method of calculating the importance score based on rigorous mathematical calculations avoids the errors of subjective judgment and makes the evaluation of feature importance more objective and scientific. The accurately calculated importance score helps to effectively screen out typical features that play a key role in breast physiological function analysis from numerous features. The importance score calculation unit can provide an accurate and reliable basis for the screening of typical features, thereby improving the detection accuracy and effectiveness of the breast detector.

[0134] Embodiment 7:

[0135] Based on Embodiment 5, the typical feature screening unit includes:

[0136] The eigenvalue sorting subunit is used to sort all the eigenvalues of the covariance matrix from smallest to largest to obtain an eigenvalue sequence, and disassemble the eigenvalue sequence to obtain a sub-eigenvalue sequence corresponding to each eigenvalue ordinal number;

[0137] The cumulative contribution rate calculation subunit is used to regard the quotient of the mean value of all the eigenvalues included in the sub-eigenvalue sequence corresponding to each eigenvalue ordinal number and the mean value of all the eigenvalues of the covariance matrix as the cumulative contribution rate corresponding to each eigenvalue ordinal number;

[0138] The typical feature screening subunit is used to screen out multiple typical breast physiological function features from all breast physiological function features based on the cumulative contribution rate corresponding to each eigenvalue ordinal number and the importance score of each breast physiological function feature in each sample.

[0139] In this embodiment, the eigenvalue sequence is disassembled to obtain a sub-eigenvalue sequence corresponding to each eigenvalue ordinal number: the subsequence formed by all eigenvalues in the eigenvalue sequence whose sorting ordinal numbers are less than or equal to each eigenvalue ordinal number is regarded as the sub-eigenvalue sequence corresponding to that eigenvalue ordinal number. For example, if the eigenvalue sequence is [1, 2, 3, 4, 5, 6, 7, 8, 9], then the subsequence corresponding to the ordinal number 3 is [1, 2, 3], the subsequence corresponding to the ordinal number 2 is [1, 2], and the subsequence corresponding to the ordinal number 1 is [1].

[0140] In this embodiment, the eigenvalue ordinal number: is the sequential number assigned to each eigenvalue in the eigenvalue sequence. For example, in the above example, the ordinal number of eigenvalue 1 is 1, the ordinal number of eigenvalue 2 is 2, and so on.

[0141] In this embodiment, based on the cumulative contribution rate corresponding to each eigenvalue ordinal number and the importance score of each breast physiological function feature in each sample, multiple typical breast physiological function features are screened out from all breast physiological function features: comprehensively considering the cumulative contribution rate corresponding to each eigenvalue ordinal number (reflecting the contribution degree of the corresponding eigenvalue to the whole) and the importance score of each breast physiological function feature in each sample, to select multiple typical features with representativeness and importance from all breast physiological function features. For example, features with a high cumulative contribution rate and a high importance score are screened as typical breast physiological function features.

[0142] The beneficial effects of the above technical solutions are as follows: The eigenvalue sorting subunit sorts and disassembles the eigenvalues, providing an ordered data structure for subsequent calculation of the cumulative contribution rate. The cumulative contribution rate calculation subunit can quantitatively evaluate the contribution degree of each eigenvalue in the overall features by calculating the cumulative contribution rate corresponding to each eigenvalue ordinal number. The typical feature screening subunit comprehensively considers the cumulative contribution rate and the importance score to screen typical features, making the screening process more comprehensive and accurate. This multi-dimensional screening method can more effectively select typical features with representativeness and importance from numerous breast physiological function features. The typical feature screening unit can accurately screen out typical features that are of great significance for breast physiological function analysis through scientific calculation and comprehensive evaluation, improving the detection performance and diagnostic accuracy of the breast detector.

[0143] Embodiment 8:

[0144] On the basis of Embodiment 4, the electrical signal analysis sub-module includes:

[0145] The feature extraction unit is used to analyze the converted electrical signal based on the physiological function analysis algorithm model to obtain the breast physiological function features of the target breast tissue;

[0146] An abnormal physiological function change judgment unit, configured to judge whether there is an abnormal physiological function change in the target breast tissue based on the breast physiological function characteristics of the target breast tissue;

[0147] A first breast abnormality detection information output unit, configured to, if it is determined that there is an abnormal physiological function change in the target breast tissue, determine the type and severity of the abnormal physiological function change as breast abnormal physiological function change detection information;

[0148] A second breast abnormality detection information output unit, configured to, if it is determined that there is no abnormal physiological function change in the target breast tissue, use the fact that there is no abnormal physiological function change in the target breast tissue as breast abnormal physiological function change detection information.

[0149] In this embodiment, the breast physiological function characteristics of the target breast tissue are obtained by analyzing the converted electrical signals based on a physiological function analysis algorithm model, that is: a specially designed physiological function analysis algorithm model is used to process the already converted electrical signals, and various characteristics and information regarding the physiological functions of the target breast tissue are extracted from them. For example, characteristics such as the blood supply speed and metabolic activity level of the breast tissue are obtained through algorithm analysis.

[0150] In this embodiment, the breast physiological function characteristics of the target breast tissue refer to various characteristics and indicators regarding physiological functions possessed by specific breast tissues. For example, the renewal rate of breast tissue cells, the uptake and utilization efficiency of nutrients, etc.

[0151] In this embodiment, judging whether there is an abnormal physiological function change in the target breast tissue based on the breast physiological function characteristics of the target breast tissue is: according to the obtained physiological function characteristics of the target breast tissue, comparing them with the normal physiological function range or standard, so as to determine whether there is an abnormal change in the physiological function of this tissue. For example, if the blood supply speed is much lower than the normal range, or the metabolic activity level is too high, it may be judged that there is an abnormality.

[0152] In this embodiment, determining the type and severity of abnormal physiological function changes means that after it is determined that there are abnormal physiological function changes, the specific category of this abnormality is further clarified, such as whether it is abnormal blood supply, abnormal metabolism or other types of abnormalities, and its severity is evaluated, such as mild, moderate or severe abnormalities. The implementation process usually includes the following steps: First, a comprehensive classification system for abnormal physiological function types needs to be established. This may be based on previous medical research, clinical experience and relevant pathological knowledge, and classify abnormal physiological function changes into different categories, such as abnormal blood supply (including insufficient blood supply, hyperemia, etc.), abnormal metabolism (such as too fast or too slow metabolism), abnormal tissue structure, etc. Then, the physiological function characteristics of the target breast tissue obtained are compared with the normal range. This may involve numerical analysis of various detection indicators, such as blood flow rate, metabolite concentration, morphological parameters of tissue structure, etc. If these indicators deviate from the normal range, it is initially judged that there is an abnormality. For the situation initially judged to be abnormal, the specific deviation pattern and degree are further analyzed. For example, if the blood flow rate is abnormal, it is necessary to determine whether it is higher or lower than the normal level and how much the deviation is. The type of abnormality is determined by synthesizing the abnormal conditions of multiple physiological function characteristics. It may be necessary to consider the changes of multiple related indicators at the same time to accurately classify the type of abnormality. For example, simultaneous abnormalities in blood flow rate and metabolite concentration may point to a specific disease type. When evaluating the severity, different grades or thresholds are usually set. According to the degree of deviation of abnormal indicators from the normal range, the severity is divided into mild, moderate, severe, etc. This may require referring to a large amount of clinical data and statistical analysis to determine a reasonable classification standard. It may also be combined with other auxiliary examination results and the patient's clinical symptoms to further confirm and correct the judgment of the type and severity of the abnormality. For example, the patient's self-feeling, the doctor's palpation results, etc. Finally, a clear conclusion about the type and severity of abnormal physiological function changes is obtained, providing a basis for subsequent diagnosis, treatment and prognosis evaluation.

[0153] The beneficial effects of the above technical solutions are as follows: The feature extraction unit can extract the physiological function characteristics of the breast from the converted electrical signals, providing a basis for subsequent judgment. The abnormal physiological function change judgment unit can accurately judge whether there is an abnormality in the target breast tissue, improving the accuracy of detection. The first breast abnormality detection information output unit determines the type and severity of the abnormality when it is determined that there is an abnormality, providing detailed information for subsequent diagnosis and treatment. The second breast abnormality detection information output unit gives a clear result when it is determined that there is no abnormality, avoiding misjudgment. The electrical signal analysis sub-module can comprehensively and accurately output the detection information of abnormal physiological function changes in the breast through an orderly analysis and judgment process, enhancing the detection value and application effect of the breast detector.

[0154] Embodiment 9:

[0155] Based on Embodiment 1, the imaging display module includes:

[0156] A dynamic imaging sub-module, which is used to generate an initial dynamic imaging result based on an optimized image reconstruction algorithm, combined with the detection information of the abnormal physiological function changes of the breast obtained in real time, the additional physiological function information of the body, and the vascular thermal distribution characteristics;

[0157] A photographing and displaying sub-module, which is used to generate and display an effective imaging result based on the imaging instruction input by the user and the initial dynamic imaging result.

[0158] In this embodiment, generating the initial dynamic imaging result based on the optimized image reconstruction algorithm, combined with the detection information of the abnormal physiological function changes of the breast obtained in real time, the additional physiological function information of the body, and the vascular thermal distribution characteristics means: applying an improved and optimized image reconstruction calculation method, and at the same time integrating various information such as the detection data of the abnormal physiological function changes of the breast collected in real time, other relevant physiological function information of the body, and the characteristics of the vascular thermal distribution, to create the initial dynamic image display result. For example, converting data such as the metabolic state and blood flow of breast tissue into dynamic images through an algorithm.

[0159] In this embodiment, the initial dynamic imaging result is the dynamic image initially generated through the above comprehensive processing, which preliminarily shows the relevant information of the breast tissue, but may need to be further adjusted and improved according to the user's needs. For example, it may be a dynamic image that initially presents the vascular distribution and functional changes in the breast.

[0160] In this embodiment, generating the effective imaging result from the imaging instruction input by the user and the initial dynamic imaging result means that the user inputs specific requirements and instructions regarding image generation according to their own needs, and the system combines these instructions with the initial dynamic imaging result, and after further processing and adjustment, finally forms an imaging result that meets the user's requirements, has practical application value, and is accurate and effective. For example, if the user requests to highlight the vascular distribution in a specific area or identify abnormal areas with a specific color, the system processes the initial image according to these instructions to obtain the final effective imaging result.

[0161] The beneficial effects of the above technical solutions are as follows: The dynamic imaging sub-module uses an optimized image reconstruction algorithm to generate the initial dynamic imaging result, which can present the relevant information of the breast more clearly and accurately. Combining various information detected in real time makes the imaging result more comprehensive and reliable, providing a more comprehensive diagnostic basis for doctors or users. The photographing and displaying sub-module generates and displays an effective imaging result according to the user's imaging instruction, meeting the personalized needs of the user and improving the user experience. The imaging display module can efficiently and flexibly present the detailed condition of the breast to the user, helping to detect problems in a timely manner and make accurate diagnoses.

[0162] Example 10:

[0163] The present invention provides a breast detection system, including any one of the breast detectors in Examples 1 to 9 and an adapter matching with any one of the breast detectors in Examples 1 to 9.

[0164] In this embodiment, the adapter refers to a device for connecting the breast detector and other devices or systems. Its function is to ensure that the breast detector can be compatible and communicate effectively with different external environments or other components, which may involve functions such as power adaptation, conversion of data transmission formats, and matching of interfaces, so as to ensure the stable operation and coordinated work of the entire breast detection system. For example, converting the specific output signal of the breast detector into a format that can be recognized and processed by a computer, or providing a suitable power input for the detector.

[0165] The beneficial effects of the above technical solutions are as follows: The breast detection system includes an advanced breast detector, which can provide comprehensive and accurate breast detection functions. The adapted adapter ensures good connection and compatibility between the detector and other devices or systems, enabling the detection system to operate stably. It improves the convenience and operability of detection, making it convenient to use in different scenarios. It enhances the overall performance and reliability of the system, providing a more perfect solution for breast detection. This breast detection system has good practicability and application value, which helps to improve the detection level and diagnostic efficiency of breast diseases.

[0166] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these changes and modifications.

Claims

1. A breast detector, characterized in that: include: A near-infrared light source module, used to emit and receive near-infrared light signals after passing through target breast tissue, and convert the received near-infrared light signals into electrical signals; Multimodal data processing module, used to perform real-time analysis on the converted electrical signals based on multiple algorithm models, obtain detection information on abnormal physiological function changes of the breast, and extract additional physiological function information and vascular thermal distribution characteristics of the body; The imaging display module is used to obtain and display effective imaging results based on the imaging instructions input by the user, the real-time detection information of abnormal physiological function changes of the breast, the body's additional physiological function information and vascular thermal distribution characteristics.

2. The breast detector according to claim 1, characterized in that: Near infrared light source module, including: A near-infrared light signal transmitting submodule, used for transmitting a near-infrared light signal in a specific wavelength range based on a preset power density; A near-infrared light signal receiving submodule, used for receiving near-infrared light signals after passing through target breast tissue based on a high-sensitivity photoelectric sensor array; The near-infrared light signal conversion submodule is used to convert the received near-infrared light signal after passing through the target breast tissue into an electrical signal.

3. The breast detector according to claim 1, characterized in that: Multimodal data processing module, including: The physiological function analysis submodule is used to analyze the converted electrical signals based on the physiological function analysis algorithm model to obtain detection information of abnormal physiological function changes of the breast; An additional physiological function information analysis submodule is used to analyze the spectrum characteristics of the converted electrical signal, infer the metabolic state of the target breast tissue and the periodic change information of the hemodynamic information, and obtain additional physiological function information of the body based on the metabolic state of the target breast tissue and the periodic change information of the hemodynamic information; The blood vessel distribution analysis submodule is used to extract the blood vessel thermal distribution characteristics in the target breast tissue from the converted electrical signal based on the blood vessel thermal distribution analysis model.

4. The breast detector according to claim 3, characterized in that: Physiological function analysis submodule, including: The breast-related data collection submodule is used to collect a large amount of breast-related data from different individuals covering normal breast tissue status and various breast diseases, as well as basic research data related to breast physiological functions, as target research data; A typical feature extraction submodule is used to extract all breast physiological function features from the target research data, and screen out a variety of typical breast physiological function features from all breast physiological function features, wherein the multiple typical breast physiological function features include a variety of typical normal breast physiological function features and a variety of typical abnormal breast physiological function features; The physiological function analysis algorithm model building submodule is used to screen out the training data of each typical breast physiological function characteristic in the target research data, and build a physiological function analysis algorithm model based on the training data of each typical breast physiological function characteristic; The electrical signal analysis submodule is used to analyze the converted electrical signals based on the physiological function analysis algorithm model to obtain detection information of abnormal breast physiological function changes.

5. The breast detector according to claim 4, characterized in that: Typical feature extraction submodules include: A feature centralization unit is used to extract all breast physiological function features from all samples in the target research data, centralize all breast physiological function features in each sample, and obtain all centralized breast physiological function features in all samples; An eigenvalue decomposition unit is used to generate a breast physiological function feature matrix based on all centralized breast physiological function features in all samples, calculate the covariance matrix of the breast physiological function feature matrix, and then perform eigenvalue decomposition on the covariance matrix of the breast physiological function feature matrix to obtain all eigenvalues ​​of the covariance matrix; A decision tree generation unit, used for generating multiple decision trees based on random forests and all breast physiological function features in all samples; An importance score calculation unit, used for calculating the importance score of each breast physiological function feature based on all decision trees; The typical feature screening unit is used to screen out a plurality of typical breast physiological function features from all breast physiological function features based on all feature values ​​and the importance score of each breast physiological function feature in each sample.

6. The breast detector according to claim 5, characterized in that: Importance score calculation unit, including: A decision tree parameter acquisition subunit is used to determine the total number of all decision trees and the total number of nodes in the decision trees and the number of samples contained in each node after splitting each node using each breast physiological function feature; The importance score subunit is used to calculate the importance score of each breast physiological function feature based on the total number of all decision trees, the total number of nodes in the decision tree, and the number of samples contained in each node.

7. The breast detector according to claim 5, characterized in that: Typical feature screening unit, including: The eigenvalue sorting subunit is used to sort all the eigenvalues ​​of the covariance matrix from small to large to obtain an eigenvalue sequence, and to decompose the eigenvalue sequence to obtain a sub-eigenvalue sequence corresponding to each eigenvalue ordinal number; The cumulative contribution rate calculation subunit is used to take the quotient of the mean of all eigenvalues ​​contained in the sub-eigenvalue sequence corresponding to each eigenvalue sequence and the mean of all eigenvalues ​​of the covariance matrix as the cumulative contribution rate corresponding to each eigenvalue sequence; The typical feature screening subunit is used to screen out a variety of typical breast physiological function features from all breast physiological function features based on the cumulative contribution rate corresponding to each eigenvalue ordinal number and the importance score of each breast physiological function feature in each sample.

8. The breast detector according to claim 4, characterized in that: The electrical signal analysis submodule includes: A feature extraction unit, used to analyze the converted electrical signal based on a physiological function analysis algorithm model to obtain breast physiological function characteristics of the target breast tissue; An abnormal physiological function change judging unit, used to judge whether the target breast tissue has an abnormal physiological function change based on the breast physiological function characteristics of the target breast tissue; a first breast abnormality detection information output unit, configured to determine the type and severity of the abnormal physiological function change as breast abnormal physiological function change detection information if it is determined that the target breast tissue has an abnormal physiological function change; The second breast abnormality detection information output unit is used to use the absence of abnormal physiological function change in the target breast tissue as the breast abnormal physiological function change detection information if it is determined that the target breast tissue does not have abnormal physiological function change.

9. The breast detector according to claim 1, characterized in that: Imaging display module, comprising: The dynamic imaging submodule is used to generate initial dynamic imaging results based on an optimized image reconstruction algorithm and in combination with real-time detection information of abnormal physiological function changes of the breast, additional physiological function information of the body, and vascular thermal distribution characteristics; The photographing and displaying submodule is used to generate and display effective imaging results based on the imaging instructions input by the user and the initial dynamic imaging results.

10. A breast detection system, characterized in that: The invention comprises the breast detector according to any one of claims 1 to 9 and an adapter matching the breast detector according to any one of claims 1 to 9.

Images