CN116189865A - Hospital reservation registration management system - Google Patents
Hospital reservation registration management system Download PDFInfo
- Publication number
- CN116189865A CN116189865A CN202310324778.7A CN202310324778A CN116189865A CN 116189865 A CN116189865 A CN 116189865A CN 202310324778 A CN202310324778 A CN 202310324778A CN 116189865 A CN116189865 A CN 116189865A
- Authority
- CN
- China
- Prior art keywords
- semantic understanding
- semantic
- vector
- feature vector
- word
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/02—Reservations, e.g. for tickets, services or events
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Tourism & Hospitality (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Epidemiology (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The application relates to the field of intelligent management, and particularly discloses a hospital reservation registration management system, which adopts a neural network model based on deep learning to mine semantic understanding characteristics of information of checked personnel and semantic understanding characteristics of alternative checking items by adopting a context encoder comprising an embedded layer, and further classifies the information by using the semantic characteristics of the information of the checked personnel and associated characteristic distribution information of the semantic characteristics of the alternative checking items, so as to obtain classification results for indicating whether the alternative checking items are recommended to the checked personnel. In this way, the inspection item of the person to be inspected can be accurately recommended based on the classification result.
Description
Technical Field
The present application relates to the field of intelligent management, and more particularly, to a hospital reservation registration management system.
Background
The enterprise physical examination is a welfare provided for staff by enterprises and public institutions, and the staff can know the health condition of the staff in time by doing the enterprise physical examination, so that the loss of human resources can be reduced, and the health physical examination is an effective means for preventing and treating diseases.
Most of the physical examination reservations of the existing medical institutions still adopt an off-line manual registration method, the manual workload of hospitals is large, the work efficiency of physical examination reservation registration business is low, errors are prone to occurring, and the follow-up historical data management is inconvenient.
Thus, a hospital reservation registration management scheme for physical examination of an enterprise is desired.
Disclosure of Invention
The present application has been made in order to solve the above technical problems. The embodiment of the application provides a hospital reservation registration management system, which adopts a neural network model based on deep learning, adopts a context encoder comprising an embedded layer to mine out semantic understanding features of information of checked personnel and semantic understanding features of alternative check items, and further classifies the checked personnel information semantic features and the alternative check items by relevance feature distribution information of the semantic features of the checked personnel information and the semantic features of the alternative check items, so as to obtain classification results for indicating whether the alternative check items are recommended to checked personnel. In this way, the inspection item of the person to be inspected can be accurately recommended based on the classification result.
According to one aspect of the present application, there is provided a hospital reservation registration management system including: the checked personnel information scheduling module is used for acquiring basic information of checked personnel and text description of checking requirements; the basic information semantic understanding module is used for enabling basic information of the detected personnel to pass through a first context encoder comprising a word embedding layer to obtain basic information semantic understanding feature vectors; an inspection requirement semantic understanding module for passing the inspection requirement text description through a second context encoder comprising an embedded layer to obtain an inspection requirement semantic understanding feature vector; the feature fusion module is used for fusing the basic information semantic understanding feature vector and the checking requirement semantic understanding feature vector to obtain an object requirement semantic understanding feature vector; the examination item description retrieving module is used for obtaining text descriptions of alternative examination items; an inspection item semantic understanding module for passing the text description of the alternative inspection item through a third context encoder comprising an embedded layer to obtain an alternative inspection item semantic understanding feature vector; the association module is used for carrying out association coding on the object requirement semantic understanding feature vector and the alternative examination item semantic understanding feature vector so as to obtain a matching feature matrix; the modulation module is used for carrying out feature distribution modulation on the matching feature matrix to obtain an optimized matching feature matrix; and the management result generation module is used for enabling the optimized matching feature matrix to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the candidate examination item is recommended to the person to be examined.
In the hospital reservation registration management system, the basic information semantic understanding module includes: the first word segmentation unit is used for carrying out word segmentation processing on the basic information of the detected person so as to convert the basic information of the detected person into a word sequence consisting of a plurality of words; a first word embedding unit, configured to map each word in the word sequence into a word embedding vector by using an embedding layer of the first context encoder including an embedding layer, so as to obtain a sequence of word embedding vectors; a first context coding unit, configured to perform global context semantic coding on the sequence of word embedding vectors using a converter of the first context encoder including an embedding layer, where the global context semantic coding is based on a converter thought, so as to obtain a plurality of global context semantic feature vectors; and the first cascading unit is used for cascading the plurality of global context semantic feature vectors to obtain the basic information semantic understanding feature vector.
In the above hospital reservation registration management system, the first context encoding unit includes: a query vector construction subunit, configured to perform one-dimensional arrangement on the sequence of word embedding vectors to obtain word feature vectors; a self-attention subunit, configured to calculate a product between the word feature vector and a transpose vector of each word vector in the sequence of word embedding vectors to obtain a plurality of self-attention association matrices; the normalization subunit is used for respectively performing normalization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of normalized self-attention correlation matrices; the attention calculating subunit is used for obtaining a plurality of probability values through a Softmax classification function by each normalized self-attention correlation matrix in the normalized self-attention correlation matrices; an attention applying subunit, configured to weight each word vector in the sequence of word embedding vectors with each probability value in the plurality of probability values as a weight to obtain the plurality of context semantic feature vectors; and the cascading subunit is used for cascading the context semantic feature vectors to obtain the global context semantic feature vectors.
In the hospital reservation registration management system, the check-in requirement semantic understanding module includes: the second word segmentation unit is used for carrying out word segmentation processing on the examination requirement text description so as to convert the examination requirement text description into a word sequence composed of a plurality of words; a second word embedding unit, configured to map each word in the word sequence into a word embedding vector by using an embedding layer of the second context encoder including an embedding layer, so as to obtain a sequence of word embedding vectors; a second context coding unit, configured to perform global context semantic coding on the sequence of word embedding vectors using a converter of the second context encoder including an embedding layer, where the global context semantic coding is based on a converter thought, so as to obtain a plurality of global context semantic feature vectors; and the second cascading unit is used for cascading the plurality of global context semantic feature vectors to obtain the check requirement semantic understanding feature vector.
In the hospital reservation registration management system, the feature fusion module is configured to: fusing the basic information semantic understanding feature vector and the checking requirement semantic understanding feature vector to obtain an object requirement semantic understanding feature vector by the following formula; wherein, the formula is: Wherein (1)>Representing the basic information semantic understanding feature vector, < >>Representing the check requirement semantic understanding feature vector, < >>Representing a cascade function->Representing the object requirement semantic understanding feature vector.
In the hospital reservation registration management system, the examination item semantic understanding module includes: a third word segmentation unit, configured to perform word segmentation processing on the text description of the candidate inspection item to convert the text description of the candidate inspection item into a word sequence composed of a plurality of words; a third word embedding unit, configured to map each word in the word sequence into a word embedding vector by using an embedding layer of the third context encoder including an embedding layer, so as to obtain a sequence of word embedding vectors; a third context coding unit, configured to perform global context semantic coding on the sequence of word embedding vectors using a converter of the third context encoder including an embedding layer, where the global context semantic coding is based on a converter thought, so as to obtain a plurality of global context semantic feature vectors; and a third concatenation unit, configured to concatenate the plurality of global context semantic feature vectors to obtain the candidate inspection item semantic understanding feature vector.
In the hospital reservation registration management system, the association module is configured to: performing association coding on the object requirement semantic understanding feature vector and the alternative examination item semantic understanding feature vector by using the following formula to obtain a matching feature matrix; wherein, the formula is:wherein->Representing the object requirements semantic understanding feature vector,a transpose vector representing the object-required semantic understanding feature vector,>representing the semantic understanding feature vector of the alternative examination item,/->Representing the matching feature matrix,/->Representing vector multiplication. />
In the above hospital reservation registration management system, the modulation module includes: the unfolding unit is used for unfolding the matching feature matrix into matching feature vectors according to rows or columns; the feature optimization unit is used for carrying out vector-normalized Hilbert probability spatialization on the matched feature vectors according to the following formula to obtain optimized matched feature vectors; wherein, the formula is:wherein->Is the matching feature vector,/->Representing the two norms of the matching feature vector, -, for example>Representing the square of the two norms of the matching feature vector,/->Is the +.o of the matching feature vector >Personal characteristic value->An exponential operation representing a vector representing a calculation of a natural exponential function value raised to a power by a eigenvalue of each position in the vector, and +.>Is the +.f of the optimized matching eigenvector>A characteristic value; and a matrix reconstruction unit, configured to perform matrix reconstruction on the optimized matching feature vector to obtain the optimized matching feature matrix.
In the hospital reservation registration management system, the management result generation module includes: the matrix unfolding unit is used for unfolding the optimized matching feature matrix into a classification feature vector based on a row vector or a column vector; the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a plurality of full-connection layers of the classifier so as to obtain coded classification characteristic vectors; and the classification result generating unit is used for enabling the coding classification feature vector to pass through a Softmax classification function of the classifier to obtain the classification result.
According to another aspect of the present application, there is provided a hospital reservation registration management method including:
basic information of a person to be inspected and text description of inspection requirements are obtained; the basic information of the detected personnel passes through a first context encoder comprising a word embedding layer to obtain basic information semantic understanding feature vectors; passing the inspection requirement text description through a second context encoder comprising an embedded layer to obtain an inspection requirement semantic understanding feature vector; fusing the basic information semantic understanding feature vector and the checking requirement semantic understanding feature vector to obtain an object requirement semantic understanding feature vector; acquiring a text description of an alternative examination item; passing the text description of the alternative inspection item through a third context encoder comprising an embedded layer to obtain an alternative inspection item semantic understanding feature vector; performing association coding on the object requirement semantic understanding feature vector and the alternative examination item semantic understanding feature vector to obtain a matching feature matrix; performing feature distribution modulation on the matching feature matrix to obtain an optimized matching feature matrix; and passing the optimized matching feature matrix through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the candidate examination item is recommended to the person to be examined.
According to still another aspect of the present application, there is provided an electronic apparatus including: a processor; and a memory in which computer program instructions are stored which, when executed by the processor, cause the processor to perform the hospital reservation registration management method as described above.
According to yet another aspect of the present application, there is provided a computer readable medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform the hospital appointment management method as described above.
Compared with the prior art, the hospital reservation registration management system provided by the application adopts the neural network model based on deep learning, adopts the context encoder comprising the embedded layer to mine the semantic understanding characteristics of the information of the checked personnel and the semantic understanding characteristics of the alternative checking items, and further classifies the checked personnel by the semantic characteristics of the information of the checked personnel and the associated characteristic distribution information of the semantic characteristics of the alternative checking items, so that a classification result used for indicating whether the alternative checking items are recommended to the checked personnel is obtained. In this way, the aptamer item recommendation of the employee can be accurately performed based on the classification result.
Drawings
The foregoing and other objects, features and advantages of the present application will become more apparent from the following more particular description of embodiments of the present application, as illustrated in the accompanying drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
Fig. 1 is a block diagram of a hospital reservation registration management system according to an embodiment of the present application.
Fig. 2 is a system architecture diagram of a hospital reservation registration management system according to an embodiment of the present application.
Fig. 3 is a block diagram of a basic information semantic understanding module in a hospital reservation registration management system according to an embodiment of the present application.
Fig. 4 is a block diagram of a modulation module in a hospital reservation registration management system according to an embodiment of the present application.
Fig. 5 is a block diagram of a management result generation module in the hospital reservation registration management system according to the embodiment of the present application.
Fig. 6 is a flowchart of a hospital reservation registration management method according to an embodiment of the present application.
Fig. 7 is a block diagram of an electronic device according to an embodiment of the present application.
Description of the embodiments
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
Summary of the application: as described above, most of the physical examination reservations of the existing medical institutions still adopt an off-line manual registration method, the manual workload of hospitals is large, the work efficiency of physical examination reservation registration business is low, errors are prone to occurring, and the subsequent history data management is inconvenient. Thus, a hospital reservation registration management scheme for physical examination of an enterprise is desired.
In constructing a hospital reservation registration management system for enterprise client physical examination, one key technical difficulty is: and recommending the matched physical examination items for the examined person. In existing order types, the inspector is typically provided with a variety of packages for selection, including but not limited to: staff attendance physical examination, staff routine physical examination, family routine physical examination, staff insurance physical examination, and the like. Moreover, the checked personnel can automatically increase physical examination items to construct a customized package, but due to the knowledge of the checked personnel, the checked personnel cannot easily distinguish which physical examination items are suitable, and often select a few irrelevant physical examination items, so that the time and the labor are consumed, and extra load is brought to the operation of a hospital. Accordingly, a hospital reservation registration management system for physical examination of an enterprise is desired that is capable of recommending an adapted physical examination item for a person to be examined based on the condition thereof.
Accordingly, in consideration of the problem of semantic understanding and feature matching of a text in the process of actually recommending an adapted physical examination item to a person to be examined, that is, the problem of respectively carrying out global contextual semantic understanding on basic information and examination requirements of the person to be examined and physical examination items and carrying out matching of semantic features based on respective semantic understanding information is considered, so that the physical examination items recommended to be adapted based on the condition of the person to be examined are realized, and labor is saved while the recommendation of the adapted physical examination items of the staff is accurately carried out. In the process, the difficulty is how to dig out the semantic understanding characteristics of the information of the checked person and the semantic understanding characteristics of the alternative checking items, and how to accurately perform the characteristic matching of the semantic understanding characteristics of the information of the checked person and the semantic understanding characteristics of the alternative checking items, so as to recommend the accurate and adaptive alternative checking items for the checked person.
In recent years, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, text signal processing, and the like. In addition, deep learning and neural networks have also shown levels approaching and even exceeding humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like.
Deep learning and development of a neural network provide new solutions and solutions for mining semantic understanding features of the information of the person to be inspected and semantic understanding features of the alternative inspection items and accurately performing feature matching of the semantic understanding features of the information of the person to be inspected and the semantic understanding features of the alternative inspection items.
Specifically, in the technical scheme of the application, first, basic information of a person to be inspected and a text description of inspection requirements are acquired. Next, considering that since the basic information of the person under test and the inspection requirement text description are both composed of a plurality of words and context semantic feature information having relevance between the words, in order to be able to perform semantic understanding on the basic information of the person under test and the inspection requirement text description accurately, a context encoder including a word embedding layer is used to perform semantic feature extraction on the basic information of the person under test and the inspection requirement text description, respectively, so as to extract respective context semantic understanding feature information, that is, the basic information of the person under test and global semantic information of the inspection requirement text description.
Specifically, word segmentation is carried out on basic information of the detected personnel so as to avoid word sequence confusion during subsequent semantic feature extraction, and the basic information is passed through a first context encoder comprising a word embedding layer so as to extract global context semantic feature information in the basic information of the detected personnel, thereby obtaining basic information semantic understanding feature vectors; and performing word segmentation processing on the check requirement text description, and performing context semantic coding in a second context encoder comprising an embedded layer to extract global context semantic understanding feature information in the check requirement text description, thereby obtaining a check requirement semantic understanding feature vector. And then fusing the basic information semantic understanding feature vector and the checking requirement semantic understanding feature vector, so as to fuse semantic feature information in the basic information of the checked person and semantic feature information in the checking requirement text description, namely comprehensive adaptation features of the checked person about physical examination, and further obtain an object requirement semantic understanding feature vector.
Further, considering that since the text description of the candidate inspection item is also composed of a plurality of words and the respective words also have semantic association features of context, in order to be able to perform deep semantic understanding on the respective types of the candidate inspection item, so as to accurately recommend the adapted candidate inspection item to the person under inspection, in the technical solution of the present application, after the text description of the candidate inspection item is acquired, the text description of the candidate inspection item is also subjected to context semantic encoding in a third context encoder containing an embedded layer, so as to obtain a candidate inspection item semantic understanding feature vector having the text description of the candidate inspection item and based on global context semantic understanding feature information.
And then, after carrying out deep global semantic understanding on the detected personnel information and the candidate inspection item information, recommending a proper candidate inspection item to the detected personnel in order to explore the matching degree between the detected personnel information and the candidate inspection item information.
And then, further classifying the matching feature matrix as a classification feature matrix by a classifier, and classifying by using the semantic features of the detected personnel information and the associated feature distribution information of the semantic features of the candidate inspection items, thereby obtaining a classification result for indicating whether to recommend the candidate inspection items to the detected personnel. That is, in the technical solution of the present application, the labeling of the classifier includes recommending the candidate inspection item to the person to be inspected, and not recommending the candidate inspection item to the person to be inspected, where the classifier determines to which classification label the classification feature matrix belongs through a soft maximum function. In this way, the classification result can be used for accurately recommending the aptamer project of the employee, and meanwhile, the labor is saved.
In particular, in the technical scheme of the application, when the object requirement semantic understanding feature vector and the alternative inspection item semantic understanding feature vector are fused to obtain the matching feature matrix, the object requirement semantic understanding feature vector and the alternative inspection item semantic understanding feature vector are multiplied by positions to obtain feature values of corresponding positions of the matching feature matrix. The object requirement semantic understanding feature vector and the alternative examination item semantic understanding feature vector respectively express the aggregate representation of the basic information and the context semantic of the examination requirement information of the examined person and the context semantic of the text description of the examination item, and the semantic distribution between the two is not consistent, so that after multiplication by location, the overall feature distribution of the matching feature matrix has distribution deviation with the individual feature distribution of the object requirement semantic understanding feature vector and the alternative examination item semantic understanding feature vector, and the accuracy of a classification result obtained by the matching feature matrix through a classifier is affected. Thus, the matching feature matrix is first expanded into matching feature vectors, e.g., denoted as And then +.>The hilbert probability spatialization of vector assignment is specifically expressed as:here, a->Representing the matching feature vector +.>Is>Representing the square thereof, i.e. the matching eigenvector +.>Internal accumulation of oneself, ->Is the matching feature vector +.>Is>Characteristic value, and->Is the optimized matching eigenvector +.>Is>And characteristic values. Here, the vector-generalized Hilbert probability is spatially formulated by the matching feature vector +.>Self-assignment of the matching feature vector in Hilbert space defining the inner product of vectors>And reduces the matching eigenvector +.>Hidden disturbances of the distribution expression of the overall Hilbert spatial topology, thereby increasing said matching feature vector +.>Is converged to the robustness of the natural distribution while relying on metrics to induce the establishment of a probabilistic spatial structure to promote the matching featuresVector->Is dependent on the long-range of the natural distribution across generators. Thus, the matching feature vector is again +.>Restoring to the matching feature matrix improves the accuracy of the classification result obtained by the matching feature matrix through the classifier. In this way, the matched physical examination items can be accurately recommended for the detected personnel based on the actual conditions of the detected personnel, so that the manual workload is saved while the matched physical examination items of the personnel are accurately recommended.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Exemplary System: fig. 1 is a block diagram of a hospital reservation registration management system according to an embodiment of the present application. As shown in fig. 1, a hospital reservation registration management system 300 according to an embodiment of the present application includes: a person under test information scheduling module 310; a basic information semantic understanding module 320; a check demand semantic understanding module 330; a feature fusion module 340; an inspection item description retrieval module 350; a check item semantic understanding module 360; an association module 370; a modulation module 380; and a management result generation module 390.
The information scheduling module 310 of the person under test is configured to obtain basic information of the person under test and a text description of the requirement of the examination; the basic information semantic understanding module 320 is configured to pass basic information of the person under test through a first context encoder that includes a word embedding layer to obtain a basic information semantic understanding feature vector; an inspection requirement semantic understanding module 330 for passing the inspection requirement text description through a second context encoder comprising an embedded layer to obtain an inspection requirement semantic understanding feature vector; the feature fusion module 340 is configured to fuse the basic information semantic understanding feature vector and the inspection requirement semantic understanding feature vector to obtain an object requirement semantic understanding feature vector; an inspection item description retrieval module 350 for obtaining a text description of an alternative inspection item; an inspection item semantic understanding module 360 for passing the text description of the alternative inspection item through a third context encoder comprising an embedded layer to obtain an alternative inspection item semantic understanding feature vector; the association module 370 is configured to perform association encoding on the object requirement semantic understanding feature vector and the candidate inspection item semantic understanding feature vector to obtain a matching feature matrix; the modulation module 380 is configured to perform feature distribution modulation on the matching feature matrix to obtain an optimized matching feature matrix; and a management result generating module 390, configured to pass the optimized matching feature matrix through a classifier to obtain a classification result, where the classification result is used to indicate whether to recommend the candidate inspection item to the person under inspection.
Fig. 2 is a system architecture diagram of a hospital reservation registration management system according to an embodiment of the present application. As shown in fig. 2, in the network architecture, first, basic information and test requirement text description of the tested person are obtained through the tested person information scheduling module 310; next, the basic information semantic understanding module 320 obtains basic information of the checked person by the checked person information scheduling module 310 through a first context encoder including a word embedding layer, so as to obtain a basic information semantic understanding feature vector; the check requirement semantic understanding module 330 passes the check requirement text description acquired by the checked personnel information scheduling module 310 through a second context encoder comprising an embedded layer to obtain a check requirement semantic understanding feature vector; the feature fusion module 340 fuses the basic information semantic understanding feature vector obtained by the basic information semantic understanding module 320 and the inspection requirement semantic understanding feature vector obtained by the inspection requirement semantic understanding module 330 to obtain an object requirement semantic understanding feature vector; the inspection item description retrieval module 350 obtains a text description of an alternative inspection item; then, the inspection item semantic understanding module 360 passes the text description of the alternative inspection item acquired by the inspection item description retrieving module 350 through a third context encoder including an embedded layer to obtain an alternative inspection item semantic understanding feature vector; the association module 370 performs association encoding on the object requirement semantic understanding feature vector obtained by the feature fusion module 340 and the candidate inspection item semantic understanding feature vector obtained by the inspection item semantic understanding module 360 to obtain a matching feature matrix; the modulating module 380 performs feature distribution modulation on the matching feature matrix to obtain an optimized matching feature matrix; further, the management result generation module 390 passes the optimized matching feature matrix through a classifier to obtain a classification result indicating whether to recommend the candidate inspection item to the person under inspection.
Specifically, during the operation of the hospital reservation registration management system 300, the checked-person information scheduling module 310 is configured to obtain basic information of the checked-person and a text description of the check-requirement. It should be understood that, if the adapted physical examination item is to be recommended for the staff member, the personal basic information of the staff member should be acquired first, but in the process of actually recommending the adapted physical examination item for the staff member, the problem of text matching is essentially a problem, that is, global context semantic understanding is respectively performed on the basic information and the inspection requirement of the staff member and the physical examination item, and the matching of semantic features is performed based on the respective semantic understanding information, so that the physical examination item for which the adapted physical examination item is recommended based on the condition of the staff member is realized, so as to accurately perform the recommendation of the adapted physical examination item for the staff member. Thus, in one specific example of the present application, the inspection requirement text description information is also acquired.
Specifically, during the operation of the hospital reservation registration management system 300, the basic information semantic understanding module 320 is configured to pass basic information of the person under examination through a first context encoder including a word embedding layer to obtain a basic information semantic understanding feature vector. In order to accurately perform semantic understanding on the basic information of the person to be inspected, a context encoder including a word embedding layer is used to extract semantic features of the basic information of the person to be inspected, so that context semantic understanding feature information of the basic information of the person to be inspected, namely global semantic information of the basic information of the person to be inspected, is extracted. In the technical scheme of the application, firstly, in order to avoid disordered word sequence during subsequent semantic feature extraction, basic information of a person to be detected is subjected to word segmentation processing to obtain a word sequence composed of a plurality of words, each word in the word sequence is mapped into a word embedding vector by using a first context encoder comprising an embedding layer to obtain a word embedding vector sequence, and then global context semantic encoding is performed on the word embedding vector sequence by using a converter of the first context encoder to obtain a plurality of global context semantic feature vectors; and finally, the global context semantic feature vectors are used for obtaining basic information semantic understanding feature vectors containing basic information of the detected personnel. Wherein the context encoder is a transducer model based encoder.
Fig. 3 is a block diagram of a basic information semantic understanding module in a hospital reservation registration management system according to an embodiment of the present application. As shown in fig. 3, the basic information semantic understanding module 320 includes: a first word segmentation unit 321, configured to perform word segmentation processing on basic information of the person to be detected, so as to convert the basic information of the person to be detected into a word sequence composed of a plurality of words; a first word embedding unit 322, configured to map each word in the word sequence into a word embedding vector by using an embedding layer of the first context encoder including an embedding layer, so as to obtain a sequence of word embedding vectors; a first context coding unit 323, configured to perform global context semantic coding on the sequence of word embedding vectors using a converter of the first context encoder including an embedding layer, where the global context semantic coding is based on a converter concept, so as to obtain a plurality of global context semantic feature vectors; and a first cascade unit 324, configured to cascade the plurality of global context semantic feature vectors to obtain the basic information semantic understanding feature vector. Wherein the first context encoding unit includes: a query vector construction subunit, configured to perform one-dimensional arrangement on the sequence of word embedding vectors to obtain word feature vectors; a self-attention subunit, configured to calculate a product between the word feature vector and a transpose vector of each word vector in the sequence of word embedding vectors to obtain a plurality of self-attention association matrices; the normalization subunit is used for respectively performing normalization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of normalized self-attention correlation matrices; the attention calculating subunit is used for obtaining a plurality of probability values through a Softmax classification function by each normalized self-attention correlation matrix in the normalized self-attention correlation matrices; an attention applying subunit, configured to weight each word vector in the sequence of word embedding vectors with each probability value in the plurality of probability values as a weight to obtain the plurality of context semantic feature vectors; and the cascading subunit is used for cascading the context semantic feature vectors to obtain the global context semantic feature vectors.
Specifically, during operation of the hospital reservation registration management system 300, the examination need semantic understanding module 330 is configured to describe the examination need text through a second context encoder that includes an embedded layer to obtain an examination need semantic understanding feature vector. It should be understood that the inspection requirement text description is also composed of a plurality of words, and context semantic feature information is associated between the words, so in order to accurately perform semantic understanding on the inspection requirement text description, a context encoder including a word embedding layer is used to extract semantic features of the inspection requirement text description, so as to extract context semantic understanding feature information, that is, global semantic information of the inspection requirement text description. Specifically, first, word segmentation processing is performed on the inspection requirement text description to convert the inspection requirement text description into a word sequence composed of a plurality of words; mapping each word in the word sequence into a word embedding vector by using an embedding layer of the second context encoder comprising the embedding layer to obtain a word embedding vector sequence; then, using the converter of the second context encoder comprising the embedding layer to perform global context semantic coding on the sequence of word embedding vectors based on the idea of the converter to obtain a plurality of global context semantic feature vectors; finally, cascading the plurality of global context semantic feature vectors to obtain an inspection requirement semantic understanding feature vector containing the inspection requirement context semantic understanding feature information.
Specifically, during the operation of the hospital reservation registration management system 300, the feature fusion module 340 is configured to fuse the basic information semantic understanding feature vector and the check-in requirement semantic understanding feature vector to obtain an object requirement semantic understanding feature vector. The basic information semantic understanding feature vector and the checking requirement semantic understanding feature vector are fused after the basic information semantic understanding feature vector and the checking requirement semantic understanding feature vector are obtained, so that semantic feature information in basic information of the checked person and semantic feature information in the checking requirement text description are fused, namely comprehensive adaptation features of the checked person about physical examination are obtained, and the object requirement semantic understanding feature vector is obtained. Specifically, in one specific example of the present application, feature information of the two may be fused in a cascade manner, more specifically, the basic information semantic understanding feature vector and the check requirement semantic understanding feature vector are fused in the following formula to obtain an object requirement semantic understanding feature vector; wherein, the formula is:wherein (1) >Representing the basic information semantic understanding feature vector, < >>Representing the check requirement semantic understanding feature vector,representing a cascade function->Representing the object requirement semantic understanding feature vector.
Specifically, during operation of the hospital reservation registration management system 300, the examination item description retrieval module 350 is configured to obtain a textual description of an alternative examination item. It should be understood that, after the object requirement semantic understanding feature vector is obtained, in order to mine the semantic understanding feature of the information of the person to be inspected and the semantic understanding feature of the alternative inspection item, in the technical scheme of the application, the text description of the alternative inspection item is also required to be obtained.
Specifically, during operation of the hospital reservation inventory management system 300, the examination item semantic understanding module 360 is configured to pass the textual description of the candidate examination item through a third context encoder that includes an embedded layer to obtain a candidate examination item semantic understanding feature vector. In order to accurately recommend the matched candidate inspection item to the person to be inspected in order to realize deep semantic understanding of each type of the candidate inspection item, in the technical scheme of the application, after the text description of the candidate inspection item is acquired, the text description of the candidate inspection item is also subjected to context semantic encoding through a third context encoder comprising an embedding layer, so that the candidate inspection item semantic understanding feature vector based on the global context semantic understanding feature information of the text description of the candidate inspection item is obtained. Specifically, to avoid word order confusion during subsequent semantic feature extraction, word segmentation processing is performed on the text description of the candidate inspection item so as to convert the text description of the candidate inspection item into a word sequence composed of a plurality of words; mapping each word in the word sequence into a word embedding vector by using an embedding layer of the third context encoder comprising the embedding layer to obtain a sequence of word embedding vectors; then using the converter of the third context encoder comprising an embedding layer to perform global context semantic coding on the sequence of word embedding vectors based on the idea of the converter to obtain a plurality of global context semantic feature vectors; and finally, cascading the plurality of global context semantic feature vectors to obtain alternative examination item semantic understanding feature vectors containing the alternative examination item context semantic understanding feature information.
Specifically, during the operation of the hospital reservation registration management system 300, the association module 370 is configured to perform association encoding on the object requirement semantic understanding feature vector and the candidate examination item semantic understanding feature vector to obtain a matching feature matrix. That is, after the depth global semantic understanding is performed on the information of the person to be inspected and the information of the candidate inspection item, in order to explore the matching degree between the two, a suitable candidate inspection item is recommended to the person to be inspected, in the technical scheme of the application, the object requirement semantic understanding feature vector and the candidate inspection item semantic understanding feature vector are further associated and encoded, so that the contextual semantic understanding feature of the information of the person to be inspected and the contextual semantic understanding feature of the information of the candidate inspection item are associated, and a matching feature matrix is obtained. More specifically, in one specific example of the present application, the object requirement semantic understanding feature vector and the candidate inspection item semantic understanding feature vector are associated and encoded in the following formula to obtain a matching feature matrix; wherein, the formula is: Wherein->Representing the object requirement semantic understanding feature vector, < >>A transpose vector representing the object-required semantic understanding feature vector,>representing the candidate exam item semantic understanding feature vector,representing the matching feature matrix,/->Representing vector multiplication.
Specifically, during the operation of the hospital reservation management system 300, the modulating module 380 is configured to perform feature distribution modulation on the matching feature matrix to obtain an optimized matching feature matrix. In the technical scheme of the application, when the object demand semantic understanding feature vector and the alternative examination item semantic understanding feature vector are fused to obtain the matching feature matrix, the object demand semantic understanding feature vector and the alternative examination item semantic understanding feature vector are multiplied by positions to obtain feature values of corresponding positions of the matching feature matrix. The object requirement semantic understanding feature vector and the alternative examination item semantic understanding feature vector respectively express the aggregate representation of the basic information and the context semantic of the examination requirement information of the examined person and the context semantic of the text description of the examination item, and the semantic distribution between the two is not consistent, so that after multiplication by location, the overall feature distribution of the matching feature matrix has distribution deviation with the individual feature distribution of the object requirement semantic understanding feature vector and the alternative examination item semantic understanding feature vector, and the accuracy of a classification result obtained by the matching feature matrix through a classifier is affected. Thus, the matching feature matrix is first expanded into matching feature vectors, e.g., denoted as And then +.>The hilbert probability spatialization of vector assignment is specifically expressed as: />Here, a->Representing the matching feature vector +.>Is>Representing the square thereof, i.e. the matching eigenvector +.>Internal accumulation of oneself, ->Is the matching feature vector +.>Is>Characteristic value, and->Is the optimized matching eigenvector +.>Is>And characteristic values. Here, the vector-generalized Hilbert probability is spatially formulated by the matching feature vector +.>Self-assignment of the matching feature vector in Hilbert space defining the inner product of vectors>And reduces the matching eigenvector +.>Hidden disturbances of the distribution expression of the overall Hilbert spatial topology, thereby increasing said matching feature vector +.>Is converged to the robustness of the natural distribution, as well asThe matching feature vector is promoted by means of the establishment of a metric-induced probability spatial structure>Is dependent on the long-range of the natural distribution across generators. Thus, the matching feature vector is again +.>Restoring to the matching feature matrix improves the accuracy of the classification result obtained by the matching feature matrix through the classifier. In this way, the matched physical examination items can be accurately recommended for the detected personnel based on the actual conditions of the detected personnel, so that the manual workload is saved while the matched physical examination items of the personnel are accurately recommended.
Fig. 4 is a block diagram of a modulation module in a hospital reservation registration management system according to an embodiment of the present application. As shown in fig. 4, the modulation module 380 includes: an unfolding unit 381, configured to unfold the matching feature matrix into matching feature vectors according to rows or columns; a feature optimization unit 382, configured to spatially model the hilbert probability of the matching feature vector according to the following formula to obtain an optimized matching feature vector; wherein, the formula is:wherein->Is the matching feature vector,/->Representing the two norms of the matching feature vector, -, for example>Representing the square of the two norms of the matching feature vector,/->Is the +.o of the matching feature vector>Personal characteristic value->An exponential operation representing a vector representing a calculation of a natural exponential function value raised to a power by a eigenvalue of each position in the vector, and +.>Is the +.f of the optimized matching eigenvector>A characteristic value; and a matrix reconstructing unit 383, configured to perform matrix reconstruction on the optimized matching feature vector to obtain the optimized matching feature matrix.
Specifically, during the operation of the hospital reservation registration management system 300, the management result generating module 390 is configured to pass the optimized matching feature matrix through a classifier to obtain a classification result, where the classification result is used to indicate whether to recommend the candidate examination item to the person under examination. That is, the optimized matching feature matrix is used as a classification feature matrix to be classified by a classifier, and the classification is performed by using the semantic features of the information of the person to be inspected and the associated feature distribution information of the semantic features of the candidate inspection item, so as to obtain a classification result for indicating whether the candidate inspection item is recommended to the person to be inspected. More specifically, the optimized matching feature matrix is processed using the classifier to obtain a classification result with the following formula: ,Wherein->Representing projection of the optimized matching feature matrix as a vector,/->To->Weight matrix for all connection layers of each layer, < ->To->Representing the bias vector for each fully connected layer. Specifically, the classifier includes a plurality of fully connected layers and a Softmax layer cascaded with a last fully connected layer of the plurality of fully connected layers. In the classification process of the classifier, the optimized matching feature matrix is first projected as a vector, for example, in a specific example, the optimized matching feature matrix is expanded along a row vector or a column vector to be a classification feature vector; then, performing multiple full-connection coding on the classification feature vectors by using multiple full-connection layers of the classifier to obtain coded classification feature vectors; further, the encoded classification feature vector is input to a Softmax layer of the classifier, i.e. the encoded classification feature vector is classified using the Softmax classification function to obtain a classification label, in a specific example of the present application, the label of the classifier comprises recommending the candidate examination item to the person under examination, and not recommending the candidate examination item to the person under examination, wherein the classifier determines to which classification label the classification feature matrix belongs by a soft maximum function. In this way, the classification result can be used for accurately recommending the aptamer project of the employee, and meanwhile, the labor is saved.
Fig. 5 is a block diagram of a management result generation module in the hospital reservation registration management system according to the embodiment of the present application. As shown in fig. 5, the management result generating module 390 includes: a matrix expansion unit 391, configured to expand the optimized matching feature matrix into a classification feature vector based on a row vector or a column vector; a full-connection encoding unit 392, configured to perform full-connection encoding on the classification feature vector by using multiple full-connection layers of the classifier to obtain an encoded classification feature vector; and a classification result generating unit 393, configured to pass the encoded classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
In summary, the hospital reservation registration management system 300 according to the embodiment of the present application is explained by employing a neural network model based on deep learning, employing a context encoder including an embedded layer to mine out semantic understanding features of subject person information and semantic understanding features of candidate examination items, and further classifying with associated feature distribution information of the semantic features of the subject person information and the semantic features of the candidate examination items, thereby obtaining classification results for indicating whether to recommend the candidate examination items to the subject person. In this way, the aptamer item recommendation of the employee can be accurately performed based on the classification result.
As described above, the hospital reservation registration management system according to the embodiment of the present application can be implemented in various terminal devices. In one example, the hospital reservation registration management system 300 according to embodiments of the present application may be integrated into the terminal device as a software module and/or hardware module. For example, the hospital reservation registration management system 300 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the hospital reservation registration management system 300 could equally be one of the plurality of hardware modules of the terminal device.
Alternatively, in another example, the hospital reservation registration management system 300 and the terminal device may be separate devices, and the hospital reservation registration management system 300 may be connected to the terminal device through a wired and/or wireless network and transmit interactive information in a contracted data format.
An exemplary method is: fig. 6 is a flowchart of a hospital reservation registration management method according to an embodiment of the present application. As shown in fig. 6, the hospital reservation registration management method according to the embodiment of the present application includes the steps of: s110, basic information of a person to be inspected and text description of inspection requirements are acquired; s120, passing the basic information of the detected personnel through a first context encoder comprising a word embedding layer to obtain basic information semantic understanding feature vectors; s130, the check requirement text description is passed through a second context encoder comprising an embedded layer to obtain a check requirement semantic understanding feature vector; s140, fusing the basic information semantic understanding feature vector and the checking requirement semantic understanding feature vector to obtain an object requirement semantic understanding feature vector; s150, acquiring text descriptions of alternative examination items; s160, enabling the text description of the alternative examination item to pass through a third context encoder comprising an embedded layer to obtain an alternative examination item semantic understanding feature vector; s170, carrying out association coding on the object requirement semantic understanding feature vector and the alternative examination item semantic understanding feature vector to obtain a matching feature matrix; s180, performing feature distribution modulation on the matching feature matrix to obtain an optimized matching feature matrix; and S190, enabling the optimized matching feature matrix to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the candidate examination item is recommended to the person to be examined.
In one example, in the hospital reservation registration management method, the step S120 includes: word segmentation processing is carried out on the basic information of the detected personnel so as to convert the basic information of the detected personnel into a word sequence consisting of a plurality of words; mapping each word in the word sequence into a word embedding vector by using an embedding layer of the first context encoder comprising the embedding layer to obtain a sequence of word embedding vectors; performing global context semantic coding on the sequence of word embedding vectors based on a converter thought by using a converter of the first context encoder comprising an embedding layer to obtain a plurality of global context semantic feature vectors; and cascading the plurality of global context semantic feature vectors to obtain the basic information semantic understanding feature vector. Wherein the performing global context semantic coding on the sequence of word embedding vectors using the converter of the first context encoder including the embedding layer based on the converter concept to obtain a plurality of global context semantic feature vectors includes: one-dimensional arrangement is carried out on the sequence of the word embedding vectors to obtain word characteristic vectors; calculating the product between the word characteristic vector and the transpose vector of each word vector in the sequence of word embedding vectors to obtain a plurality of self-attention association matrices; respectively carrying out standardization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of standardized self-attention correlation matrices; obtaining a plurality of probability values by using a Softmax classification function through each normalized self-attention correlation matrix in the normalized self-attention correlation matrices; weighting each word vector in the sequence of word embedding vectors by taking each probability value in the plurality of probability values as a weight so as to obtain the plurality of context semantic feature vectors; cascading the plurality of context semantic feature vectors to obtain the plurality of global context semantic feature vectors.
In one example, in the hospital reservation registration management method, the step S130 includes: word segmentation processing is carried out on the check requirement text description so as to convert the check requirement text description into a word sequence composed of a plurality of words; mapping each word in the word sequence into a word embedding vector by using an embedding layer of the second context encoder comprising the embedding layer to obtain a sequence of word embedding vectors; performing global context semantic coding on the sequence of word embedding vectors based on a converter concept by using a converter of the second context encoder comprising an embedding layer to obtain a plurality of global context semantic feature vectors; and cascading the plurality of global context semantic feature vectors to obtain the check requirement semantic understanding feature vector.
In one example, in the hospital reservation registration management method, the step S140 includes: fusing the basic information semantic understanding feature vector and the checking requirement semantic understanding feature vector to obtain an object requirement semantic understanding feature vector by the following formula; wherein, the formula is:wherein (1)>Representing the basic information semantic understanding feature vector, < > >Representing the check requirement semantic understanding feature vector, < >>Representing a function of the cascade of functions,representing the object requirement semantic understanding feature vector.
In one example, in the hospital reservation registration management method, the step S160 includes: word segmentation processing is carried out on the text description of the alternative examination item so as to convert the text description of the alternative examination item into a word sequence composed of a plurality of words; mapping each word in the word sequence into a word embedding vector by using an embedding layer of the third context encoder comprising the embedding layer to obtain a sequence of word embedding vectors; performing global context semantic coding on the sequence of word embedding vectors based on a converter concept by using a converter of the third context encoder comprising an embedding layer to obtain a plurality of global context semantic feature vectors; and cascading the plurality of global context semantic feature vectors to obtain the candidate inspection item semantic understanding feature vector.
In one example, in the hospital reservation registration management method, the step S170 includes: performing association coding on the object requirement semantic understanding feature vector and the alternative examination item semantic understanding feature vector by using the following formula to obtain a matching feature matrix; wherein, the formula is: Wherein->Representing the object requirement semantic understanding feature vector, < >>A transpose vector representing the object-required semantic understanding feature vector,>representing the semantic understanding feature vector of the alternative examination item,/->Representing the matching feature matrix,/->Representing vector multiplication.
In one example, in the hospital reservation registration management method, the step S180 includes: expanding the matching feature matrix into matching feature vectors according to rows or columns; carrying out vector-normalized Hilbert probability spatialization on the matching feature vectors by using the following formula to obtain optimized matching feature vectors; wherein, the formula is:wherein->Is the matching feature vector,/->Representing the two norms of the matching feature vector, -, for example>Representing the square of the two norms of the matching feature vector,/->Is the +.o of the matching feature vector>Personal characteristic value->An exponential operation representing a vector representing a calculation of a natural exponential function value raised to a power by a eigenvalue of each position in the vector, and +.>Is the +.f of the optimized matching eigenvector>A characteristic value; and performing matrix reconstruction on the optimized matching feature vector to obtain the optimized matching feature matrix.
In one example, in the hospital reservation registration management method, the step S190 includes: expanding the optimized matching feature matrix into classification feature vectors based on row vectors or column vectors; performing full-connection coding on the classification feature vectors by using a plurality of full-connection layers of the classifier to obtain coded classification feature vectors; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
In summary, the hospital reservation registration management method according to the embodiment of the present application is explained by employing a neural network model based on deep learning, employing a context encoder including an embedded layer to mine out semantic understanding features of information of a person under test and semantic understanding features of an alternative examination item, and further classifying with associated feature distribution information of the semantic features of the person under test and the semantic features of the alternative examination item, thereby obtaining a classification result for indicating whether to recommend the alternative examination item to the person under test. In this way, the aptamer item recommendation of the employee can be accurately performed based on the classification result.
Exemplary electronic device: next, an electronic device according to an embodiment of the present application is described with reference to fig. 7.
Fig. 7 illustrates a block diagram of an electronic device according to an embodiment of the present application.
As shown in fig. 7, the electronic device 10 includes one or more processors 11 and a memory 12.
The processor 11 may be a Central Processing Unit (CPU) or other form of processing unit having data processing and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.
In one example, the electronic device 10 may further include: an input device 13 and an output device 14, which are interconnected by a bus system and/or other forms of connection mechanisms (not shown).
The input means 13 may comprise, for example, a keyboard, a mouse, etc.
The output device 14 may output various information including the classification result and the like to the outside. The output means 14 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.
Of course, only some of the components of the electronic device 10 that are relevant to the present application are shown in fig. 7 for simplicity, components such as buses, input/output interfaces, etc. are omitted. In addition, the electronic device 10 may include any other suitable components depending on the particular application.
Exemplary computer program product and computer readable storage Medium: in addition to the methods and apparatus described above, embodiments of the present application may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform steps in the functions of the hospital appointment management method according to the various embodiments of the present application described in the "exemplary systems" section of the present specification.
The computer program product may write program code for performing the operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the present application may also be a computer-readable storage medium, having stored thereon computer program instructions, which when executed by a processor, cause the processor to perform steps in the functions of the hospital appointment registration management method according to the various embodiments of the present application described in the above-mentioned "exemplary systems" section of the present specification.
The computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not limiting, and these advantages, benefits, effects, etc. are not to be considered as necessarily possessed by the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not intended to be limited to the details disclosed herein as such.
The block diagrams of the devices, apparatuses, devices, systems referred to in this application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent to the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (9)
1. A hospital reservation registration management system, comprising: the checked personnel information scheduling module is used for acquiring basic information of checked personnel and text description of checking requirements; the basic information semantic understanding module is used for enabling basic information of the detected personnel to pass through a first context encoder comprising a word embedding layer to obtain basic information semantic understanding feature vectors; an inspection requirement semantic understanding module for passing the inspection requirement text description through a second context encoder comprising an embedded layer to obtain an inspection requirement semantic understanding feature vector; the feature fusion module is used for fusing the basic information semantic understanding feature vector and the checking requirement semantic understanding feature vector to obtain an object requirement semantic understanding feature vector; the examination item description retrieving module is used for obtaining text descriptions of alternative examination items; an inspection item semantic understanding module for passing the text description of the alternative inspection item through a third context encoder comprising an embedded layer to obtain an alternative inspection item semantic understanding feature vector; the association module is used for carrying out association coding on the object requirement semantic understanding feature vector and the alternative examination item semantic understanding feature vector so as to obtain a matching feature matrix; the modulation module is used for carrying out feature distribution modulation on the matching feature matrix to obtain an optimized matching feature matrix; and the management result generation module is used for enabling the optimized matching feature matrix to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the candidate examination item is recommended to the person to be examined.
2. The hospital reservation registration management system of claim 1, wherein the basic information semantic understanding module comprises: the first word segmentation unit is used for carrying out word segmentation processing on the basic information of the detected person so as to convert the basic information of the detected person into a word sequence consisting of a plurality of words; a first word embedding unit, configured to map each word in the word sequence into a word embedding vector by using an embedding layer of the first context encoder including an embedding layer, so as to obtain a sequence of word embedding vectors; a first context coding unit, configured to perform global context semantic coding on the sequence of word embedding vectors using a converter of the first context encoder including an embedding layer, where the global context semantic coding is based on a converter thought, so as to obtain a plurality of global context semantic feature vectors; and the first cascading unit is used for cascading the plurality of global context semantic feature vectors to obtain the basic information semantic understanding feature vector.
3. The hospital reservation registration management system of claim 2, wherein the first context encoding unit comprises: a query vector construction subunit, configured to perform one-dimensional arrangement on the sequence of word embedding vectors to obtain word feature vectors; a self-attention subunit, configured to calculate a product between the word feature vector and a transpose vector of each word vector in the sequence of word embedding vectors to obtain a plurality of self-attention association matrices; the normalization subunit is used for respectively performing normalization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of normalized self-attention correlation matrices; the attention calculating subunit is used for obtaining a plurality of probability values through a Softmax classification function by each normalized self-attention correlation matrix in the normalized self-attention correlation matrices; an attention applying subunit, configured to weight each word vector in the sequence of word embedding vectors with each probability value in the plurality of probability values as a weight to obtain the plurality of context semantic feature vectors; and the cascading subunit is used for cascading the context semantic feature vectors to obtain the global context semantic feature vectors.
4. The hospital reservation inventory management system of claim 3, wherein the check-in requirement semantic understanding module comprises: the second word segmentation unit is used for carrying out word segmentation processing on the examination requirement text description so as to convert the examination requirement text description into a word sequence composed of a plurality of words; a second word embedding unit, configured to map each word in the word sequence into a word embedding vector by using an embedding layer of the second context encoder including an embedding layer, so as to obtain a sequence of word embedding vectors; a second context coding unit, configured to perform global context semantic coding on the sequence of word embedding vectors using a converter of the second context encoder including an embedding layer, where the global context semantic coding is based on a converter thought, so as to obtain a plurality of global context semantic feature vectors; and the second cascading unit is used for cascading the plurality of global context semantic feature vectors to obtain the check requirement semantic understanding feature vector.
5. The hospital reservation inventory management system of claim 4, wherein the feature fusion module is configured to: fusing the basic information semantic understanding feature vector and the checking requirement semantic understanding feature vector to obtain an object requirement semantic understanding feature vector by the following formula; wherein, the formula is: Wherein (1)>Representing the basic information semantic understanding feature vector, < >>Representing the check requirement semantic understanding feature vector, < >>Representing a cascade function->Representing the object requirement semantic understanding feature vector.
6. The hospital reservation inventory management system of claim 5, wherein the examination item semantic understanding module comprises: a third word segmentation unit, configured to perform word segmentation processing on the text description of the candidate inspection item to convert the text description of the candidate inspection item into a word sequence composed of a plurality of words; a third word embedding unit, configured to map each word in the word sequence into a word embedding vector by using an embedding layer of the third context encoder including an embedding layer, so as to obtain a sequence of word embedding vectors; a third context coding unit, configured to perform global context semantic coding on the sequence of word embedding vectors using a converter of the third context encoder including an embedding layer, where the global context semantic coding is based on a converter thought, so as to obtain a plurality of global context semantic feature vectors; and a third concatenation unit, configured to concatenate the plurality of global context semantic feature vectors to obtain the candidate inspection item semantic understanding feature vector.
7. The hospital reservation inventory management system of claim 6, wherein the association module is configured to: performing association coding on the object requirement semantic understanding feature vector and the alternative examination item semantic understanding feature vector by using the following formula to obtain a matching feature matrix; wherein, the formula is:wherein->Representing the object requirement semantic understanding feature vector, < >>A transpose vector representing the object-required semantic understanding feature vector,>representing the semantic understanding feature vector of the alternative examination item,/->Representing the matching feature matrix,/->Representing vector multiplication.
8. The hospital reservation inventory management system of claim 7, wherein the modulation module comprises: the unfolding unit is used for unfolding the matching feature matrix into matching feature vectors according to rows or columns; the feature optimization unit is used for carrying out vector-normalized Hilbert probability spatialization on the matched feature vectors according to the following formula to obtain optimized matched feature vectors; wherein, the formula is:wherein->Is the matching feature vector,/->Representing the two norms of the matching feature vector, -, for example>Representing the square of the two norms of the matching feature vector,/- >Is the +.o of the matching feature vector>Personal characteristic value->An exponential operation representing a vector representing a calculation of a natural exponential function value raised to a power by a eigenvalue of each position in the vector, and +.>Is the +.f of the optimized matching eigenvector>A characteristic value; and a matrix reconstruction unit for reconstructing the matrixAnd performing matrix reconstruction on the optimized matching feature vector to obtain the optimized matching feature matrix.
9. The hospital reservation registration management system of claim 8, wherein the management result generation module comprises: the matrix unfolding unit is used for unfolding the optimized matching feature matrix into a classification feature vector based on a row vector or a column vector; the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a plurality of full-connection layers of the classifier so as to obtain coded classification characteristic vectors; and the classification result generating unit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310324778.7A CN116189865A (en) | 2023-03-30 | 2023-03-30 | Hospital reservation registration management system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310324778.7A CN116189865A (en) | 2023-03-30 | 2023-03-30 | Hospital reservation registration management system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116189865A true CN116189865A (en) | 2023-05-30 |
Family
ID=86436778
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310324778.7A Pending CN116189865A (en) | 2023-03-30 | 2023-03-30 | Hospital reservation registration management system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116189865A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116431004A (en) * | 2023-06-01 | 2023-07-14 | 山东协和学院 | Control method and system for interactive behavior of rehabilitation robot |
CN116682541A (en) * | 2023-06-14 | 2023-09-01 | 杭州医初科技有限公司 | Medical treatment recommendation system and method |
CN117095802A (en) * | 2023-10-17 | 2023-11-21 | 吉林大学 | Intelligent management system and method for accompanying personnel |
CN117171413A (en) * | 2023-09-07 | 2023-12-05 | 滨州八爪鱼网络科技有限公司 | Data processing system and method for digital collection management |
CN117235630A (en) * | 2023-11-15 | 2023-12-15 | 吉林大学 | Intelligent disease area visual management system and method thereof |
CN117789185A (en) * | 2024-02-28 | 2024-03-29 | 浙江驿公里智能科技有限公司 | Automobile oil hole gesture recognition system and method based on deep learning |
CN118266854A (en) * | 2024-05-30 | 2024-07-02 | 美视康健(吉林)医疗设备有限公司 | Stereoscopic vision detection control system and method |
-
2023
- 2023-03-30 CN CN202310324778.7A patent/CN116189865A/en active Pending
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116431004A (en) * | 2023-06-01 | 2023-07-14 | 山东协和学院 | Control method and system for interactive behavior of rehabilitation robot |
CN116431004B (en) * | 2023-06-01 | 2023-08-29 | 山东协和学院 | Control method and system for interactive behavior of rehabilitation robot |
CN116682541A (en) * | 2023-06-14 | 2023-09-01 | 杭州医初科技有限公司 | Medical treatment recommendation system and method |
CN117171413A (en) * | 2023-09-07 | 2023-12-05 | 滨州八爪鱼网络科技有限公司 | Data processing system and method for digital collection management |
CN117171413B (en) * | 2023-09-07 | 2024-03-08 | 滨州八爪鱼网络科技有限公司 | Data processing system and method for digital collection management |
CN117095802A (en) * | 2023-10-17 | 2023-11-21 | 吉林大学 | Intelligent management system and method for accompanying personnel |
CN117095802B (en) * | 2023-10-17 | 2024-01-26 | 吉林大学 | Intelligent management system and method for accompanying personnel |
CN117235630A (en) * | 2023-11-15 | 2023-12-15 | 吉林大学 | Intelligent disease area visual management system and method thereof |
CN117235630B (en) * | 2023-11-15 | 2024-03-05 | 吉林大学 | Intelligent disease area visual management system and method thereof |
CN117789185A (en) * | 2024-02-28 | 2024-03-29 | 浙江驿公里智能科技有限公司 | Automobile oil hole gesture recognition system and method based on deep learning |
CN117789185B (en) * | 2024-02-28 | 2024-05-10 | 浙江驿公里智能科技有限公司 | Automobile oil hole gesture recognition system and method based on deep learning |
CN118266854A (en) * | 2024-05-30 | 2024-07-02 | 美视康健(吉林)医疗设备有限公司 | Stereoscopic vision detection control system and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116189865A (en) | Hospital reservation registration management system | |
US20230153177A1 (en) | Systems and methods to improve data clustering using a meta-clustering model | |
CN109033068B (en) | Method and device for reading and understanding based on attention mechanism and electronic equipment | |
CN110569332B (en) | Sentence feature extraction processing method and device | |
CN107680661B (en) | System and method for estimating medical resource demand | |
CN117251699B (en) | Medical big data analysis method and system based on artificial intelligence | |
CN115796173A (en) | Data processing method and system for supervision submission requirements | |
CN116663568B (en) | Critical task identification system and method based on priority | |
EP3407208A1 (en) | Ontology alignment apparatus, program, and method | |
CN116010586A (en) | Method, device, equipment and storage medium for generating health advice | |
CN116151604A (en) | Office system flow analysis system and method under web environment | |
CN112800217A (en) | Vector relevance matrix-based intelligent assessment method for government affair transaction processing report | |
KR102682244B1 (en) | Method for learning machine-learning model with structured ESG data using ESG auxiliary tool and service server for generating automatically completed ESG documents with the machine-learning model | |
CN117370524A (en) | Training method of reply generation model, reply sentence generation method and device | |
AU2019290658B2 (en) | Systems and methods for identifying and linking events in structured proceedings | |
US11783244B2 (en) | Methods and systems for holistic medical student and medical residency matching | |
JP6811468B2 (en) | Information processing equipment, data retrieval methods, programs, data structures and data processing systems | |
CN113326698A (en) | Method for detecting entity relationship, model training method and electronic equipment | |
Abedini et al. | Epci: an embedding method for post-correction of inconsistency in the RDF knowledge bases | |
CN112396095A (en) | Intelligent assignment method of theme label based on social administration data | |
US20200401877A1 (en) | Maintaining master data using hierarchical classification | |
US20240185043A1 (en) | Generating Synthetic Heterogenous Time-Series Data | |
KR102575752B1 (en) | Examination data classification device and classification method using ensemble classification model | |
CN116913501A (en) | Thyroid cancer pathological image processing system based on machine vision | |
Aryani et al. | Change propagation analysis using domain information |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |