Internet-based rehabilitation nursing system and method for breast cancer patients

Through the Internet-based rehabilitation nursing system for breast cancer patients, using intelligent algorithms and natural language processing technology, the problem that traditional rehabilitation nursing model cannot meet the needs of breast cancer patients is solved, and personalized online consultation services for breast cancer patients are realized, improving coverage and response speed of rehabilitation care.

CN118507083BActive Publication Date: 2024-11-05JILIN UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410949633.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-16
Publication Date
2024-11-05
Estimated Expiration
2044-07-16

AI Technical Summary

Technical Problem

The traditional rehabilitation care model cannot meet the diverse needs of breast cancer patients, especially those in remote areas, and the communication between patients and professional medical staff is limited.

Method used

Develop an Internet-based rehabilitation care system for breast cancer patients, using intelligent algorithms and natural language processing technology to semantic code and analyze users' online consultation questions, and match standard questions with high adaptability through standard question databases, and return to the corresponding Reference answer.

Benefits of technology

Through the Internet platform, breaking through geographical restrictions and providing personalized online consultation services for breast cancer patients, improving the coverage and response speed of rehabilitation care, and meeting the diverse needs of patients.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118507083B_ABST
    Figure CN118507083B_ABST
Patent Text Reader

Abstract

The present application discloses an Internet-based rehabilitation nursing system and method for breast cancer patients, which relates to the field of rehabilitation nursing. It uses the Internet platform to break through geographical restrictions and build an Internet-based online consultation solution for breast cancer patients' rehabilitation nursing. More specifically, using intelligent algorithms and natural language processing technology, semantic encoding and analysis are performed on users' online consultation questions, and standard questions with high adaptability are matched in the standard question library, so as to return reference answers corresponding to alternative standard questions as consultation replies. In this way, personalized consultation services on breast cancer are provided to breast cancer patients.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of rehabilitation nursing, and more specifically, to an Internet-based rehabilitation nursing system and method for breast cancer patients. Background Art

[0002] Breast cancer is one of the most common malignant tumors in women. Its treatment and rehabilitation care have an important impact on the patient's quality of life and survival rate. However, the traditional rehabilitation care model is often limited by factors such as geography and time, and cannot meet the diverse rehabilitation needs of patients. Specifically, traditional rehabilitation care services cannot cover all patients, especially those in remote areas. In addition, there are time and space limitations for communication between patients and professional medical staff.

[0003] With the rapid development of Internet technology, people's way of obtaining and communicating health information has also undergone profound changes. The Internet platform can break through geographical restrictions and provide services to all breast cancer patients. Therefore, it is of great practical significance to develop an Internet-based rehabilitation nursing system for breast cancer patients and provide online consultation. Summary of the invention

[0004] In order to solve the above technical problems, this application is proposed.

[0005] According to one aspect of the present application, there is provided an Internet-based rehabilitation nursing system for breast cancer patients, comprising:

[0006] An online consultation question module is used to obtain online consultation questions input by users;

[0007] An alternative standard question extraction module is used to extract a first alternative standard question from a standard question library;

[0008] A semantic encoding and contextual reinforcement module, used for performing semantic encoding and contextual semantic reinforcement on the online consultation question and the first candidate standard question to obtain a sequence of word-granular semantic encoding vectors for reinforcing the online consultation question and a sequence of word-granular semantic encoding vectors for reinforcing the first candidate standard question;

[0009] A matching coefficient calculation module, used to calculate the feature sequence matching coefficient between the sequence of the enhanced online consultation question word granularity semantic coding vector and the sequence of the enhanced first candidate standard question word granularity semantic coding vector;

[0010] A reply module, configured to return a reference answer corresponding to the first candidate standard question as a consultation reply in response to the matching coefficient between the feature sequences being greater than or equal to a preset threshold;

[0011] Wherein, the semantic encoding and context enhancement module includes:

[0012] A consultation question word granularity semantic encoding unit, used for performing word segmentation processing on the online consultation question and then passing it through a semantic encoder including a word embedding layer to obtain a sequence of online consultation question word granularity semantic encoding vectors;

[0013] A word granularity semantic encoding unit for candidate standard questions, configured to obtain a sequence of word granularity semantic encoding vectors of the first candidate standard questions after word segmentation processing is performed on the first candidate standard questions through the semantic encoder including the word embedding layer;

[0014] A sequence context semantic reinforcement unit is used to input the sequence of online consultation question word granularity semantic coding vectors and the sequence of the first alternative standard question word granularity semantic coding vectors into a saliency-global based context semantic reinforcement module to obtain the sequence of reinforced online consultation question word granularity semantic coding vectors and the sequence of reinforced first alternative standard question word granularity semantic coding vectors.

[0015] In the above-mentioned Internet-based breast cancer patient rehabilitation nursing system, the sequence context semantic reinforcement unit includes:

[0016] A saliency extraction subunit, used for extracting an online consultation question word granular semantic salient feature vector of a sequence of online consultation question word granular semantic encoding vectors;

[0017] A global extraction subunit, used for extracting a global feature vector of the granular semantics of the online consultation question words of the sequence of the granular semantic encoding vectors of the online consultation question words;

[0018] A neighborhood encoding and activation subunit, used for performing neighborhood encoding and activation on the online consultation question word granular semantic salient feature vector and the online consultation question word granular semantic global feature vector to obtain an online consultation question word granular semantic salient feature context-associated activation feature vector and an online consultation question word granular semantic global feature context-associated activation feature vector;

[0019] A non-linear transformation subunit is used for fusing the online consultation question word granular semantic prominence feature context association activation feature vector and the online consultation question word granular semantic global feature context association activation feature vector to obtain the online consultation question word granular semantic prominence-global feature context association activation feature vector, and then performing non-linear activation on the online consultation question word granular semantic prominence-global feature context association activation feature vector to obtain the online consultation question word granular semantic prominence-global feature context association weight feature vector;

[0020] The semantic reinforcement subunit is used to use the online consultation question word granular semantic prominence-global feature context association weight feature vector as the weight vector, perform positional dot multiplication on each online consultation question word granular semantic coding vector in the sequence of online consultation question word granular semantic coding vectors, and add the sequence obtained by positional dot multiplication to the sequence of online consultation question word granular semantic coding vectors element by element to obtain the sequence of reinforced online consultation question word granular semantic coding vectors.

[0021] In the above Internet-based breast cancer patient rehabilitation nursing system, the saliency extraction subunit is used to:

[0022] The maximum value of each online consultation question word granular semantic coding vector in the sequence of the online consultation question word granular semantic coding vector is calculated to obtain the online consultation question word granular semantic salient feature vector composed of each of the maximum values.

[0023] In the above Internet-based breast cancer patient rehabilitation nursing system, the global extraction subunit is used to:

[0024] The average value of each online consultation question word granular semantic coding vector in the sequence of the online consultation question word granular semantic coding vector is calculated to obtain the online consultation question word granular semantic global feature vector composed of each of the average values.

[0025] In the above Internet-based breast cancer patient rehabilitation nursing system, the neighborhood coding and activation subunit is used to:

[0026] Performing one-dimensional convolution coding, nonlinear activation and linear transformation on the granular semantic salient feature vector of the online consultation question words to obtain the granular semantic salient feature context-related activation feature vector of the online consultation question words;

[0027] The online consultation question word granular semantic global feature vector is subjected to one-dimensional convolution coding, nonlinear activation and linear transformation to obtain the online consultation question word granular semantic global feature context-related activation feature vector.

[0028] In the above Internet-based breast cancer patient rehabilitation nursing system, the nonlinear transformation subunit is used to:

[0029] The online consultation question word granular semantic prominence-global feature context association activation feature vector is subjected to nonlinear activation based on tanh activation function and nonlinear activation based on sigmoid activation function to obtain the online consultation question word granular semantic prominence-global feature context association weight feature vector.

[0030] In the above Internet-based breast cancer patient rehabilitation nursing system, the matching coefficient calculation module is used to:

[0031] Calculate the cosine similarity matrix between the sequence of the enhanced online consultation question word granularity semantic coding vectors and the sequence of the enhanced first alternative standard question word granularity semantic coding vectors to obtain a first cosine similarity matrix, and calculate the cosine similarity matrix between the sequence of the enhanced first alternative standard question word granularity semantic coding vectors and the sequence of the enhanced online consultation question word granularity semantic coding vectors to obtain a second cosine similarity matrix;

[0032] Multiplying the weight hyperparameter by the accumulated value of each eigenvalue of the first cosine similarity matrix to obtain an optimized evaluation result word granularity semantic matching item;

[0033] Subtracting the weight hyperparameter from one to obtain a weight hyperparameter complement, and multiplying the weight hyperparameter complement by the accumulated value of each eigenvalue of the second cosine similarity matrix to obtain an optimized first candidate learning resource word granularity semantic matching item;

[0034] The matching coefficient between the feature sequences is obtained by adding the word granularity semantic matching item of the optimized evaluation result and the word granularity semantic matching item of the optimized first candidate learning resource.

[0035] According to another aspect of the present application, a method for rehabilitation nursing of breast cancer patients based on the Internet is provided, which comprises:

[0036] Obtain online consultation questions input by users;

[0037] Extracting a first candidate standard question from a standard question bank;

[0038] Performing semantic encoding and contextual semantic enhancement on the online consultation question and the first candidate standard question to obtain a sequence of enhanced online consultation question word granularity semantic encoding vectors and a sequence of enhanced first candidate standard question word granularity semantic encoding vectors;

[0039] Calculating a feature sequence matching coefficient between the sequence of the enhanced online consultation question word granularity semantic encoding vectors and the sequence of the enhanced first candidate standard question word granularity semantic encoding vectors;

[0040] In response to the matching coefficient between the feature sequences being greater than or equal to a preset threshold, returning a reference answer corresponding to the first candidate standard question as a consultation reply;

[0041] The online consultation question and the first candidate standard question are semantically encoded and contextually semantically enhanced to obtain a sequence of enhanced online consultation question word granularity semantic encoding vectors and a sequence of enhanced first candidate standard question word granularity semantic encoding vectors, including:

[0042] After word segmentation processing is performed on the online consultation question, a semantic encoder including a word embedding layer is used to obtain a sequence of word-granular semantic encoding vectors of the online consultation question;

[0043] After word segmentation processing is performed on the first candidate standard question, the first candidate standard question is passed through the semantic encoder including the word embedding layer to obtain a sequence of word granularity semantic encoding vectors of the first candidate standard question;

[0044] The sequence of the online consultation question word granularity semantic coding vectors and the sequence of the first alternative standard question word granularity semantic coding vectors are input into a saliency-global based contextual semantic enhancement module to obtain the sequence of enhanced online consultation question word granularity semantic coding vectors and the sequence of enhanced first alternative standard question word granularity semantic coding vectors.

[0045] Compared with the prior art, the present application provides an Internet-based breast cancer patient rehabilitation nursing system and method, which can break through geographical restrictions by utilizing the Internet platform to build an Internet-based breast cancer patient rehabilitation nursing online consultation program. More specifically, the user's online consultation questions are semantically encoded and analyzed using intelligent algorithms and natural language processing technology, and standard questions with high adaptability are matched in the standard question library, thereby returning reference answers corresponding to the alternative standard questions as consultation replies. In this way, breast cancer patients are provided with personalized consultation services on breast cancer. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0047] Figure 1 is a block diagram of an Internet-based breast cancer patient rehabilitation nursing system according to an embodiment of the present application;

[0048] Figure 2 is a system architecture diagram of an Internet-based breast cancer patient rehabilitation nursing system according to an embodiment of the present application;

[0049] Figure 3 A block diagram of a semantic encoding and context enhancement module in an Internet-based breast cancer patient rehabilitation nursing system according to an embodiment of the present application;

[0050] Figure 4The present invention is a flowchart of an Internet-based rehabilitation nursing method for breast cancer patients according to an embodiment of the present application. DETAILED DESCRIPTION

[0051] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0052] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0053] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0054] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0055] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0056] Traditional rehabilitation nursing services cannot cover all patients, especially those in remote areas. In addition, there are time and space limitations for patients to communicate with professional medical staff. With the rapid development of Internet technology, people's access to and communication of health information has also undergone profound changes. The use of the Internet platform can break through geographical restrictions and provide services to all breast cancer patients. Therefore, it is of great practical significance to develop an Internet-based rehabilitation nursing system for breast cancer patients and provide online consultation.

[0057] In the technical solution of the present application, an Internet-based rehabilitation nursing system for breast cancer patients is proposed. Figure 1The block diagram of the Internet-based breast cancer patient rehabilitation nursing system according to an embodiment of the present application. Figure 2 FIG. 1 is a system architecture diagram of an Internet-based breast cancer patient rehabilitation nursing system according to an embodiment of the present application. Figure 1 and Figure 2 As shown, according to an embodiment of the present application, an Internet-based rehabilitation and nursing system 300 for breast cancer patients includes: an online consultation question module 310, used to obtain online consultation questions input by users; an alternative standard question extraction module 320, used to extract a first alternative standard question from a standard question library; a semantic encoding and contextual enhancement module 330, used to perform semantic encoding and contextual semantic enhancement on the online consultation question and the first alternative standard question to obtain a sequence of enhanced online consultation question word granularity semantic encoding vectors and a sequence of enhanced first alternative standard question word granularity semantic encoding vectors; a matching coefficient calculation module 340, used to calculate the feature sequence matching coefficient between the sequence of enhanced online consultation question word granularity semantic encoding vectors and the sequence of enhanced first alternative standard question word granularity semantic encoding vectors; a reply module 350, used to return a reference answer corresponding to the first alternative standard question as a consultation reply in response to the feature sequence matching coefficient being greater than or equal to a preset threshold.

[0058] In particular, the online consultation question module 310 is used to obtain online consultation questions input by users. Here, online consultation questions are a direct way for patients to express their own questions and concerns. The consultation needs of each user are unique, and by obtaining the online consultation questions, the specific needs of the user can be identified and understood.

[0059] In particular, the alternative standard question extraction module 320 is used to extract the first alternative standard question from the standard question library. The standard question library is pre-established. In the embodiment of the present application, the standard questions and reference answers of the standard question library are reviewed and confirmed by professional medical staff, ensuring the accuracy and authority of the answers, which helps patients obtain reliable rehabilitation guidance. It is worth mentioning that by pre-establishing a standard question library, the system can quickly screen standard questions with high relevance, thereby shortening the time patients wait for replies and improving the response speed of the service.

[0060] In particular, the semantic encoding and contextual enhancement module 330 is used to perform semantic encoding and contextual semantic enhancement on the online consultation question and the first candidate standard question to obtain a sequence of enhanced online consultation question word granularity semantic encoding vectors and a sequence of enhanced first candidate standard question word granularity semantic encoding vectors. In particular, in a specific example of the present application, Figure 3As shown, the semantic coding and contextual enhancement module 330 includes: a consultation question word granularity semantic coding unit 331, which is used to perform word segmentation processing on the online consultation question and then pass it through a semantic encoder including a word embedding layer to obtain a sequence of online consultation question word granularity semantic coding vectors; an alternative standard question word granularity semantic coding unit 332, which is used to perform word segmentation processing on the first alternative standard question and then pass it through the semantic encoder including a word embedding layer to obtain a sequence of first alternative standard question word granularity semantic coding vectors; a sequence contextual semantic enhancement unit 333, which is used to input the sequence of online consultation question word granularity semantic coding vectors and the sequence of the first alternative standard question word granularity semantic coding vectors into the contextual semantic enhancement module based on significance-globality to obtain the sequence of enhanced online consultation question word granularity semantic coding vectors and the sequence of enhanced first alternative standard question word granularity semantic coding vectors.

[0061] Specifically, the consultation question word granularity semantic encoding unit 331 and the alternative standard question word granularity semantic encoding unit 332 are used to obtain a sequence of online consultation question word granularity semantic encoding vectors after word segmentation processing of the online consultation question; and obtain a sequence of first alternative standard question word granularity semantic encoding vectors after word segmentation processing of the first alternative standard question. Among them, the word segmentation processing can divide the continuous text into independent vocabulary units, so as to clarify the structure of the text data; the word embedding layer can convert the vocabulary into vector form, and these vectors capture the semantic information of the vocabulary; and the contextual semantic feature information of the online consultation question and the first alternative standard question can be learned through the semantic encoder, which is of great significance for understanding the user's consultation needs and the semantic meaning of the first alternative standard question. In a specific example, after word segmentation, the online consultation question is passed through a semantic encoder including a word embedding layer to obtain a sequence of word-granularity semantic coding vectors of the online consultation question, including: word segmentation is performed on the online consultation question to convert the online consultation question into a word sequence composed of multiple words; the embedding layer of the semantic encoder including the word embedding layer is used to map each word in the word sequence into a word embedding vector to obtain a sequence of word embedding vectors; and the converter of the semantic encoder including the word embedding layer is used to perform global context semantic coding on the sequence of word embedding vectors based on the converter idea to obtain a sequence of word-granularity semantic coding vectors of the online consultation question.

[0062] Specifically, the sequence context semantics reinforcement unit 333 is used to input the sequence of the online consultation question word granularity semantic coding vectors and the sequence of the first alternative standard question word granularity semantic coding vectors into a context semantic reinforcement module based on significance-globality to obtain the sequence of the reinforced online consultation question word granularity semantic coding vectors and the sequence of the reinforced first alternative standard question word granularity semantic coding vectors. Usually in the process of understanding a text, people's attention to each word will be different. That is, each word has a different importance for the overall semantic understanding and context information acquisition. For example, in the actual application scenario of the present application, keywords such as professional terms usually carry more information and are of greater importance for identifying the user's consulting needs. Therefore, in order to capture this difference in importance and to better understand the implicit semantic expressions contained in the online consultation questions and the first alternative standard questions, in the technical solution of the present application, the sequence of the online consultation question word granularity semantic coding vectors and the sequence of the first alternative standard question word granularity semantic coding vectors are input into a contextual semantic enhancement module based on saliency-globality to obtain a sequence of enhanced online consultation question word granularity semantic coding vectors and a sequence of enhanced first alternative standard question word granularity semantic coding vectors. Specifically, in an embodiment of the present application, the contextual semantic enhancement module based on saliency-globality captures the saliency-prominent feature information and the global representative feature information in the sequence of the online consultation question word granularity semantic coding vectors and the sequence of the first alternative standard question word granularity semantic coding vectors respectively during the processing process, and performs semantic optimization on the original sequence feature representation based on this saliency-prominent feature information and the global representative feature information, thereby enhancing key information and obtaining a more comprehensive semantic understanding.

[0063] In a specific example of the present application, the sequence of the online consultation question word granularity semantic coding vectors and the sequence of the first candidate standard question word granularity semantic coding vectors are input into a contextual semantic enhancement module based on significance-globality to obtain the sequence of enhanced online consultation question word granularity semantic coding vectors and the sequence of enhanced first candidate standard question word granularity semantic coding vectors, including: first, extracting the online consultation question word granularity semantic salient feature vector of the sequence of the online consultation question word granularity semantic coding vectors. Specifically, the maximum value of each online consultation question word granularity semantic coding vector in the sequence of the online consultation question word granularity semantic coding vector is calculated to obtain the online consultation question word granularity semantic salient feature vector composed of each of the maximum values. Next, extract the online consultation question word granularity semantic global feature vector of the sequence of the online consultation question word granularity semantic coding vector. Specifically, the average value of each online consultation question word granularity semantic coding vector in the sequence of the online consultation question word granularity semantic coding vector is calculated to obtain the online consultation question word granularity semantic global feature vector composed of each of the average values. Then, the online consultation question word granular semantic salient feature vector and the online consultation question word granular semantic global feature vector are subjected to neighborhood coding and activation to obtain the online consultation question word granular semantic salient feature context-associated activation feature vector and the online consultation question word granular semantic global feature context-associated activation feature vector. Specifically, the online consultation question word granular semantic salient feature vector is subjected to one-dimensional convolution coding, nonlinear activation and linear transformation to obtain the online consultation question word granular semantic salient feature context-associated activation feature vector; the online consultation question word granular semantic global feature vector is subjected to one-dimensional convolution coding, nonlinear activation and linear transformation to obtain the online consultation question word granular semantic global feature context-associated activation feature vector. Then, after fusing the online consultation question word granular semantic prominence feature context-associated activation feature vector and the online consultation question word granular semantic global feature context-associated activation feature vector to obtain the online consultation question word granular semantic prominence-global feature context-associated activation feature vector, the online consultation question word granular semantic prominence-global feature context-associated activation feature vector is nonlinearly activated to obtain the online consultation question word granular semantic prominence-global feature context-associated weight feature vector. Specifically, the online consultation question word granular semantic prominence-global feature context-associated activation feature vector is nonlinearly activated based on the tanh activation function and nonlinearly activated based on the sigmoid activation function to obtain the online consultation question word granular semantic prominence-global feature context-associated weight feature vector.Finally, taking the online consultation question word granular semantic prominence-global feature context associated weight feature vector as the weight vector, perform positional dot multiplication on each online consultation question word granular semantic coding vector in the sequence of online consultation question word granular semantic coding vectors, and add the sequence obtained by positional dot multiplication to the sequence of online consultation question word granular semantic coding vectors element by element to obtain the sequence of enhanced online consultation question word granular semantic coding vectors.

[0064] In summary, in the above embodiment, the sequence of the online consultation question word granularity semantic coding vectors and the sequence of the first candidate standard question word granularity semantic coding vectors are input into the contextual semantic enhancement module based on significance-globality to obtain the sequence of enhanced online consultation question word granularity semantic coding vectors and the sequence of enhanced first candidate standard question word granularity semantic coding vectors, including: using the contextual semantic enhancement module based on significance-globality to perform contextual semantic enhancement on the sequence of the online consultation question word granularity semantic coding vectors and the sequence of the first candidate standard question word granularity semantic coding vectors according to the following formula to obtain the sequence of enhanced online consultation question word granularity semantic coding vectors and the sequence of enhanced first candidate standard question word granularity semantic coding vectors; wherein the formula is:

[0065] ;

[0066] ;

[0067] ;

[0068] ;

[0069] ;

[0070] in, is a sequence of granular semantic encoding vectors of the online consultation question words, is the global feature vector of the granular semantics of the online consultation question words, is the semantically salient feature vector of the online consultation question word granularity, It means taking the average value, Indicates taking the maximum value, represents a one-dimensional convolution operation, express Activation function, and are the first weight matrix and the second weight matrix for linear transformation, and They are the global feature context-related activation feature vector of the granular semantics of online consultation question words and the context-related activation feature vector of the granular semantics of online consultation question words. express function, express Activation function, represents dot product, It is a sequence of granular semantic encoding vectors of the enhanced online consultation question words.

[0071] It is worth mentioning that in other specific examples of the present application, the online consultation question and the first alternative standard question can also be semantically encoded and contextually semantically enhanced in other ways to obtain a sequence of word-granularity semantic encoding vectors for enhanced online consultation questions and a sequence of word-granularity semantic encoding vectors for enhanced first alternative standard questions, for example: input the online consultation question and the first alternative standard question; use natural language processing tools (such as BERT, GloVe, etc.) to convert the online consultation question into a word embedding vector representation to capture its semantic information; use the same natural language processing tool to convert the first alternative standard question into a word embedding vector representation; combine the semantic encoding vector of the online consultation question with the contextual information, and the contextual information can be integrated into the question through methods such as attention mechanism or recurrent neural network (RNN) to enhance semantic expression; combine the semantic encoding vector of the first alternative standard question with the contextual information to obtain a sequence of word-granularity semantic encoding vectors for enhanced online consultation questions and a sequence of word-granularity semantic encoding vectors for enhanced first alternative standard questions.

[0072] In particular, the matching coefficient calculation module 340 is used to calculate the feature sequence matching coefficient between the sequence of the enhanced online consultation question word granularity semantic coding vector and the sequence of the enhanced first alternative standard question word granularity semantic coding vector. That is, the semantic similarity and correlation between the user's consultation question semantic features and the first alternative standard question semantic features in the high-dimensional feature distribution space are quantified by the feature sequence matching coefficient. In particular, when quantifying the matching degree and the correlation degree, the feature sequence matching coefficient captures the semantic similarity of the sequence of the enhanced online consultation question word granularity semantic coding vector relative to the sequence of the enhanced first alternative standard question word granularity semantic coding vector, and the semantic similarity of the sequence of the enhanced first alternative standard question word granularity semantic coding vector relative to the sequence of the enhanced online consultation question word granularity semantic coding vector, thereby improving and optimizing the matching quantification process. In addition, weight hyperparameters and the complement of the weight hyperparameters are also used as adjustment values ​​to allow the system to adjust the degree of influence of different features on the matching results.

[0073] In a specific example of the present application, the feature sequence matching coefficient between the sequence of the enhanced online consultation question word granularity semantic coding vectors and the sequence of the enhanced first alternative standard question word granularity semantic coding vectors is calculated, including: calculating the cosine similarity matrix between the sequence of the enhanced online consultation question word granularity semantic coding vectors and the sequence of the enhanced first alternative standard question word granularity semantic coding vectors to obtain a first cosine similarity matrix, and calculating the cosine similarity matrix between the sequence of the enhanced first alternative standard question word granularity semantic coding vectors and the sequence of the enhanced online consultation question word granularity semantic coding vectors. The first cosine similarity matrix is ​​used to obtain a second cosine similarity matrix; a weight hyperparameter is multiplied by the accumulated value of each eigenvalue of the first cosine similarity matrix to obtain an optimized evaluation result word granularity semantic matching item; one is subtracted from the weight hyperparameter to obtain a weight hyperparameter complement, and the weight hyperparameter complement is multiplied by the accumulated value of each eigenvalue of the second cosine similarity matrix to obtain an optimized first candidate learning resource word granularity semantic matching item; the optimized evaluation result word granularity semantic matching item and the optimized first candidate learning resource word granularity semantic matching item are added to obtain the matching coefficient between the feature sequences.

[0074] In summary, in the above embodiment, calculating the feature sequence matching coefficient between the sequence of the enhanced online consultation question word granularity semantic coding vector and the sequence of the enhanced first alternative standard question word granularity semantic coding vector includes: using the feature sequence matching calculation formula to calculate the feature sequence matching coefficient between the sequence of the enhanced online consultation question word granularity semantic coding vector and the sequence of the enhanced first alternative standard question word granularity semantic coding vector; wherein, the feature sequence matching calculation formula is:

[0075] ;

[0076] ;

[0077] ;

[0078] in, The first in the sequence of the word granular semantic encoding vector of the enhanced online consultation question Enhanced online consultation question word granularity semantic encoding vector, The first candidate standard question word in the sequence of the semantic encoding vector of the enhanced first candidate standard question word granularity A reinforced first candidate standard question word granularity semantic encoding vector, represents the transpose of the eigenvector, represents the bi-norm of the eigenvector, is the number of granular semantic encoding vectors of the enhanced online consultation question words, is the number of the semantic encoding vectors of the word granularity of the first candidate standard question, For the The word-granular semantic encoding vector of the enhanced online consultation question and the The cosine similarity between the semantic encoding vectors of the first candidate question word granularity is strengthened. For the The first candidate standard question word granularity semantic encoding vector and the first The cosine similarity between the granular semantic encoding vectors of the enhanced online consultation question words, and They are the cosine similarity matrix between the sequence of semantic encoding vectors of the word granularity of the enhanced online consultation question and the sequence of semantic encoding vectors of the word granularity of the enhanced first candidate standard question, and the cosine similarity matrix between the sequence of semantic encoding vectors of the word granularity of the enhanced first candidate standard question and the sequence of semantic encoding vectors of the word granularity of the enhanced online consultation question, respectively. is the weight hyperparameter, is the matching coefficient between the feature sequences.

[0079] Here, for the sequence of word-granular semantic coding vectors for the enhanced online consultation questions and the sequence of word-granular semantic coding vectors for the enhanced first alternative standard questions, due to the further increase in the difference in semantic features of word segmentation embedding text caused by the source semantic differences between the online consultation questions input by the user and the first alternative standard questions extracted from the standard question library during contextual semantic enhancement based on significance-globality, the sequence of word-granular semantic coding vectors for the enhanced online consultation questions and the sequence of word-granular semantic coding vectors for the enhanced first alternative standard question as a whole will have the problem of insufficient aggregation of text semantic feature distribution, thereby affecting their mapping consistency to the common semantic feature space matching domain, and affecting the calculation accuracy of the feature sequence matching coefficient between the sequence of word-granular semantic coding vectors for the enhanced online consultation questions and the sequence of word-granular semantic coding vectors for the enhanced first alternative standard question.

[0080] Based on this, the present application optimizes the sequence of the enhanced online consultation question word granularity semantic coding vector and the sequence of the enhanced first candidate standard question word granularity semantic coding vector, specifically including:

[0081] Concatenating the sequence of the enhanced online consultation question word granularity semantic coding vectors and the sequence of the enhanced first candidate standard question word granularity semantic coding vectors into an enhanced online consultation question-first candidate standard question source word granularity joint semantic coding vector;

[0082] The enhanced online consultation question-first alternative standard question source word granularity joint semantic coding vector is multiplied by the length of the enhanced online consultation question-first alternative standard question source word granularity joint semantic coding vector and the square root of the length of the enhanced online consultation question-first alternative standard question source word granularity joint semantic coding vector to obtain the enhanced online consultation question-first alternative standard question source word granularity joint full semantic representation vector and the enhanced online consultation question-first alternative standard question source word granularity joint semi-semantic representation vector;

[0083] Perform point subtraction on the enhanced online consultation question-first candidate standard question source word granularity joint full semantic representation vector and the norm of the enhanced online consultation question-first candidate standard question source word granularity joint semantic encoding vector, and calculate the base 2 logarithm of the absolute value of each position of the point subtraction result to obtain the enhanced online consultation question-first candidate standard question source word granularity joint full semantic information vector;

[0084] Subtract the binary norm of the enhanced online consultation question-first alternative standard question source word granularity joint semi-semantic representation vector from the enhanced online consultation question-first alternative standard question source word granularity joint semantic encoding vector, and calculate the base-2 logarithm of the absolute value of each position of the subtraction result to obtain the enhanced online consultation question-first alternative standard question source word granularity joint semi-semantic information vector;

[0085] The weighted sum of the enhanced online consultation question-first alternative standard question source word granularity joint full semantic information vector and the enhanced online consultation question-first alternative standard question source word granularity joint semi-semantic information vector is calculated using the balance hyperparameter as a weight to obtain an optimized enhanced online consultation question-first alternative standard question source word granularity joint semantic encoding vector;

[0086] The optimized enhanced online consultation question-first alternative standard question source word granularity joint semantic coding vector is split into a sequence of optimized enhanced online consultation question word granularity semantic coding vectors and a sequence of optimized enhanced first alternative standard question word granularity semantic coding vectors based on the cascade of the sequence of the enhanced online consultation question word granularity semantic coding vectors and the sequence of the enhanced first alternative standard question word granularity semantic coding vectors.

[0087] That is, the present application takes the enhanced online consultation question-first alternative standard question source word granularity joint semantic coding vector, which is a joint semantic representation of the sequence of the enhanced online consultation question word granularity semantic coding vector and the sequence of the enhanced first alternative standard question word granularity semantic coding vector, as a feature set, in which the changing semantic representation is based on the feature value of the enhanced online consultation question-first alternative standard question source word granularity joint semantic coding vector.

[0088] In this way, in order to dynamically aggregate the semantic set as a whole composed of different changing semantics of the source word granularity joint semantic coding vector of the enhanced online consultation question-first alternative standard question without ignoring the individual semantic change information, the individual features of the source word granularity joint semantic coding vector of the enhanced online consultation question-first alternative standard question and the collective expression of its aggregation scale are used as full-width semantic features and half-width semantic features, and the low-rank negative correlation of different dimensions of the overall semantics of the feature set of the source word granularity joint semantic coding vector of the enhanced online consultation question-first alternative standard question is used as the phase and scaled to dynamically adjust the changing relationship of the semantic content of the source word granularity joint semantic coding vector of the enhanced online consultation question-first alternative standard question.

[0089] As a result, the overall aggregation of semantic information expression of the feature set of the enhanced online consultation question-first alternative standard question source word granularity joint semantic coding vector is improved, and the problem of insufficient aggregation of text semantic feature distribution in the sequence of the enhanced online consultation question word granularity semantic coding vector and the sequence of the enhanced first alternative standard question word granularity semantic coding vector is improved, thereby enhancing the calculation accuracy of the feature sequence matching coefficient between the sequence of the enhanced online consultation question word granularity semantic coding vector and the sequence of the enhanced first alternative standard question word granularity semantic coding vector.

[0090] In particular, the reply module 350 is used to return the reference answer corresponding to the first candidate standard question as a consultation reply in response to the matching coefficient between the feature sequences being greater than or equal to a preset threshold. Here, by setting the preset threshold, it can be ensured that the reference answer corresponding to the candidate standard question will only be considered as a suitable answer when the candidate standard question is semantically relevant enough to the user's question. This helps to provide information that is highly relevant to user needs and ensures that the reply obtained by the user is highly accurate.

[0091] As described above, the Internet-based breast cancer patient rehabilitation nursing system 300 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with an Internet-based breast cancer patient rehabilitation nursing algorithm. In a possible implementation, the Internet-based breast cancer patient rehabilitation nursing system 300 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the Internet-based breast cancer patient rehabilitation nursing system 300 can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the Internet-based breast cancer patient rehabilitation nursing system 300 can also be one of the many hardware modules of the wireless terminal.

[0092] Alternatively, in another example, the Internet-based breast cancer patient rehabilitation and nursing system 300 and the wireless terminal may also be separate devices, and the Internet-based breast cancer patient rehabilitation and nursing system 300 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0093] Furthermore, an Internet-based rehabilitation nursing method for breast cancer patients is also provided.

[0094] Figure 4 FIG. 1 is a flow chart of an Internet-based breast cancer patient rehabilitation nursing method according to an embodiment of the present application. Figure 4 As shown, according to the embodiment of the present application, the Internet-based rehabilitation nursing method for breast cancer patients includes the following steps: S1, obtaining an online consultation question input by a user; S2, extracting a first candidate standard question from a standard question library; S3, performing semantic encoding and contextual semantic enhancement on the online consultation question and the first candidate standard question to obtain a sequence of enhanced online consultation question word granularity semantic encoding vectors and a sequence of enhanced first candidate standard question word granularity semantic encoding vectors; S4, calculating a feature sequence matching coefficient between the sequence of enhanced online consultation question word granularity semantic encoding vectors and the sequence of enhanced first candidate standard question word granularity semantic encoding vectors; S5, in response to the feature sequence matching coefficient being greater than or equal to a preset threshold, returning a reference answer corresponding to the first candidate standard question as a consultation reply.

[0095] In summary, the Internet-based rehabilitation nursing method for breast cancer patients according to the embodiment of the present application is explained, which can break through the geographical restrictions by utilizing the Internet platform, and construct an Internet-based online consultation solution for rehabilitation nursing for breast cancer patients. More specifically, the user's online consultation questions are semantically encoded and analyzed using intelligent algorithms and natural language processing technology, and standard questions with high adaptability are matched in the standard question library, thereby returning reference answers corresponding to the alternative standard questions as consultation replies. In this way, personalized consultation services about breast cancer are provided to breast cancer patients.

[0096] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. An Internet-based rehabilitation nursing system for breast cancer patients, characterized in that: include: An online consultation question module is used to obtain online consultation questions input by users; An alternative standard question extraction module is used to extract a first alternative standard question from a standard question library; A semantic encoding and contextual reinforcement module, used for performing semantic encoding and contextual semantic reinforcement on the online consultation question and the first candidate standard question to obtain a sequence of word-granular semantic encoding vectors for reinforcing the online consultation question and a sequence of word-granular semantic encoding vectors for reinforcing the first candidate standard question; A matching coefficient calculation module, used to calculate the feature sequence matching coefficient between the sequence of the enhanced online consultation question word granularity semantic coding vector and the sequence of the enhanced first candidate standard question word granularity semantic coding vector; A reply module, configured to return a reference answer corresponding to the first candidate standard question as a consultation reply in response to the matching coefficient between the feature sequences being greater than or equal to a preset threshold; Wherein, the semantic encoding and context enhancement module includes: A consultation question word granularity semantic encoding unit, used for performing word segmentation processing on the online consultation question and then passing it through a semantic encoder including a word embedding layer to obtain a sequence of online consultation question word granularity semantic encoding vectors; A word granularity semantic encoding unit for candidate standard questions, configured to obtain a sequence of word granularity semantic encoding vectors of the first candidate standard questions after word segmentation processing is performed on the first candidate standard questions through the semantic encoder including the word embedding layer; A sequence contextual semantics reinforcement unit, used for inputting the sequence of the online consultation question word granular semantic coding vectors and the sequence of the first candidate standard question word granular semantic coding vectors into a contextual semantic reinforcement module based on significance-globality to obtain the sequence of the reinforced online consultation question word granular semantic coding vectors and the sequence of the reinforced first candidate standard question word granular semantic coding vectors; Wherein, the sequence context semantic enhancement unit includes: A saliency extraction subunit, used for extracting an online consultation question word granular semantic salient feature vector of a sequence of online consultation question word granular semantic encoding vectors; A global extraction subunit, used for extracting a global feature vector of the granular semantics of the online consultation question words of the sequence of the granular semantic encoding vectors of the online consultation question words; A neighborhood encoding and activation subunit, used for performing neighborhood encoding and activation on the online consultation question word granular semantic salient feature vector and the online consultation question word granular semantic global feature vector to obtain an online consultation question word granular semantic salient feature context-associated activation feature vector and an online consultation question word granular semantic global feature context-associated activation feature vector; A non-linear transformation subunit is used for fusing the online consultation question word granular semantic prominence feature context association activation feature vector and the online consultation question word granular semantic global feature context association activation feature vector to obtain the online consultation question word granular semantic prominence-global feature context association activation feature vector, and then performing non-linear activation on the online consultation question word granular semantic prominence-global feature context association activation feature vector to obtain the online consultation question word granular semantic prominence-global feature context association weight feature vector; A semantic reinforcement subunit, used to use the online consultation question word granular semantic prominence-global feature context association weight feature vector as a weight vector, perform positional dot multiplication on each online consultation question word granular semantic coding vector in the sequence of the online consultation question word granular semantic coding vector, and perform element-by-element addition of the sequence obtained by positional dot multiplication and the sequence of the online consultation question word granular semantic coding vector to obtain the sequence of the reinforced online consultation question word granular semantic coding vector; Wherein, the saliency extraction subunit is used to: Calculating the maximum value of each online consultation question word granular semantic coding vector in the sequence of the online consultation question word granular semantic coding vector to obtain the online consultation question word granular semantic salient feature vector composed of each of the maximum values; Wherein, the neighborhood coding and activation subunit is used to: Performing one-dimensional convolution coding, nonlinear activation and linear transformation on the granular semantic salient feature vector of the online consultation question words to obtain the granular semantic salient feature context-related activation feature vector of the online consultation question words; The online consultation question word granular semantic global feature vector is subjected to one-dimensional convolution coding, nonlinear activation and linear transformation to obtain the online consultation question word granular semantic global feature context-related activation feature vector.

2. The Internet-based breast cancer patient rehabilitation nursing system according to claim 1 is characterized in that: The global extraction subunit is used for: The average value of each online consultation question word granular semantic coding vector in the sequence of the online consultation question word granular semantic coding vector is calculated to obtain the online consultation question word granular semantic global feature vector composed of each of the average values.

3. The Internet-based breast cancer patient rehabilitation nursing system according to claim 2 is characterized in that: The nonlinear transformation subunit is used for: The online consultation question word granular semantic prominence-global feature context association activation feature vector is subjected to nonlinear activation based on tanh activation function and nonlinear activation based on sigmoid activation function to obtain the online consultation question word granular semantic prominence-global feature context association weight feature vector.

4. The Internet-based breast cancer patient rehabilitation nursing system according to claim 3 is characterized in that: The matching coefficient calculation module is used to: Calculate the cosine similarity matrix between the sequence of the enhanced online consultation question word granularity semantic coding vectors and the sequence of the enhanced first alternative standard question word granularity semantic coding vectors to obtain a first cosine similarity matrix, and calculate the cosine similarity matrix between the sequence of the enhanced first alternative standard question word granularity semantic coding vectors and the sequence of the enhanced online consultation question word granularity semantic coding vectors to obtain a second cosine similarity matrix; Multiplying the weight hyperparameter by the accumulated value of each eigenvalue of the first cosine similarity matrix to obtain an optimized evaluation result word granularity semantic matching item; Subtracting the weight hyperparameter from one to obtain a weight hyperparameter complement, and multiplying the weight hyperparameter complement by the accumulated value of each eigenvalue of the second cosine similarity matrix to obtain an optimized first candidate learning resource word granularity semantic matching item; The word granularity semantic matching item of the optimized evaluation result and the word granularity semantic matching item of the optimized first candidate learning resource are added together to obtain the matching coefficient between the feature sequences.

5. An Internet-based rehabilitation nursing method for breast cancer patients, using the Internet-based rehabilitation nursing system for breast cancer patients according to claim 1, characterized in that: include: Obtain online consultation questions input by users; Extracting a first candidate standard question from a standard question bank; Performing semantic encoding and contextual semantic enhancement on the online consultation question and the first candidate standard question to obtain a sequence of enhanced online consultation question word granularity semantic encoding vectors and a sequence of enhanced first candidate standard question word granularity semantic encoding vectors; Calculating the feature sequence matching coefficient between the sequence of the enhanced online consultation question word granularity semantic encoding vectors and the sequence of the enhanced first candidate standard question word granularity semantic encoding vectors; In response to the matching coefficient between the feature sequences being greater than or equal to a preset threshold, returning a reference answer corresponding to the first candidate standard question as a consultation reply; The online consultation question and the first candidate standard question are semantically encoded and contextually semantically enhanced to obtain a sequence of enhanced online consultation question word granularity semantic encoding vectors and a sequence of enhanced first candidate standard question word granularity semantic encoding vectors, including: After word segmentation processing is performed on the online consultation question, a semantic encoder including a word embedding layer is used to obtain a sequence of word-granular semantic encoding vectors of the online consultation question; After word segmentation processing is performed on the first candidate standard question, the first candidate standard question is passed through the semantic encoder including the word embedding layer to obtain a sequence of word granularity semantic encoding vectors of the first candidate standard question; The sequence of the online consultation question word granularity semantic coding vectors and the sequence of the first alternative standard question word granularity semantic coding vectors are input into a saliency-global based contextual semantic enhancement module to obtain the sequence of enhanced online consultation question word granularity semantic coding vectors and the sequence of enhanced first alternative standard question word granularity semantic coding vectors.