Medical image information management system based on cloud platform

Abstract

The invention relates to the field of medical information, and particularly discloses a medical image information management system based on a cloud platform, which is characterized in that when compatibility evaluation is carried out on a preselected medical institution, besides acquiring information of medical image acquisition equipment, a software version, an imaging protocol and image processing technical information are introduced; semantic coding is carried out on acquisition equipment information of target image information and acquisition information of various pre-selected medical institution images by adopting a semantic embedding coding technology based on deep learning so as to capture implicit relations and context information among data; and furthermore, semantic query matching coding is performed on target image acquisition equipment information and pre-selected medical institution image acquisition information, so that significant correlation characteristics between the target image acquisition equipment information and the pre-selected medical institution image acquisition information can be mined, and compatibility evaluation of the pre-selected medical institution can be realized. In this way, the accuracy of evaluating the compatibility degree of the pre-selected medical institutions can be improved, and the medical service quality and efficiency are further improved.

Description

Technical Field

[0001] The present application relates to the field of medical information technology, and more specifically, to a medical imaging information management system based on a cloud platform. Background Art

[0002] Medical imaging information management systems play a vital role in the modern medical system. In the traditional medical system, the diagnosis and treatment systems of various hospitals are relatively independent, and the medical imaging information of each hospital is mostly limited to a single medical institution, making it difficult to achieve cross-institutional data sharing. As a result, patients need to re-collect medical imaging information when transferred to other hospitals for treatment, which not only wastes patients' time and examination costs, but also reduces the quality of medical services.

[0003] Chinese patent CN117316392A proposes a medical imaging information management system and method based on smart medical care. Based on the patient's basic information, it obtains target imaging information that matches the patient's condition from a medical imaging cloud database, and determines the compatibility of the target imaging information in the corresponding medical institution by judging the similarity between the acquisition device of the target imaging information and the acquisition device of the target medical institution, thereby preventing the repeated acquisition of medical images when the information is common.

[0004] However, when conducting compatibility assessments of medical institutions, the above-mentioned scheme only uses a simple similarity calculation formula to calculate the similarity between medical imaging acquisition devices as the basis for compatibility assessments of medical institutions. This method mainly relies on hardware information such as the model, brand, and specifications of the equipment, but ignores software-level compatibility, such as the version of the image processing software, protocol support, etc., which may result in the evaluation results being incomplete and inaccurate.

[0005] Therefore, an optimized cloud-based medical imaging information management system is expected. Summary of the Invention

[0006] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a medical imaging information management system based on a cloud platform, which, when performing a compatibility assessment of pre-selected medical institutions, in addition to obtaining medical imaging acquisition equipment information, also introduces software version, imaging protocol and image processing technology information, and uses a semantic embedding coding technology based on deep learning to semantically encode the acquisition equipment information of the target imaging information and the acquisition information of each pre-selected medical institution's image to capture the implicit relationship and contextual information between the data. Secondly, the acquisition equipment information of the target imaging information and the acquisition information of the pre-selected medical institution's image are further semantically queried and matched to mine the significant correlation features between the two, thereby realizing the compatibility assessment of the pre-selected medical institution. In this way, the accuracy of the compatibility assessment of the pre-selected medical institution can be improved, providing a scientific basis for the optimal allocation and sharing of medical resources, and further improving the quality and efficiency of medical services.

[0007] According to one aspect of the present application, a cloud-based medical imaging information management system is provided, comprising:

[0008] A medical imaging information storage module is used to collect medical imaging information and store the medical imaging information in a medical imaging cloud database of a cloud platform;

[0009] A target image information acquisition module is used to acquire acquisition device information of the target image information from the medical image cloud database based on the patient's basic information;

[0010] A module for acquiring image information of a pre-selected medical institution, configured to acquire image acquisition information of the pre-selected medical institution, wherein the acquisition information includes acquisition device information, software version, imaging protocol, and image processing technology;

[0011] a semantic feature extraction module, configured to perform semantic embedding coding on each data item in the acquisition device information of the target image information and the acquisition information of the pre-selected medical institution images to obtain a sequence of semantic embedding coding vectors of the target image acquisition device information and semantic embedding coding vectors of the pre-selected medical institution image information;

[0012] A query coding module, used to perform query coding on the sequence of the target image acquisition device information semantic embedding coding vector and the pre-selected medical institution image information semantic embedding coding vector to obtain a target image query response enhanced coding feature, including: a significant feature extraction unit, used to extract the significant features of the sequence of the pre-selected medical institution image information semantic embedding coding vector to obtain a semantic-guided benchmark; a heterogeneous feature enhanced query coding unit, used to perform heterogeneous feature enhanced query coding on the sequence of the target image acquisition device information semantic embedding coding vector and the pre-selected medical institution image information semantic embedding coding vector based on the semantic-guided benchmark to obtain a target image query response enhanced coding vector as the target image query response enhanced coding feature;

[0013] A compatibility evaluation module is used to evaluate the compatibility of the preselected medical institutions based on the enhanced coding features of the target image query response.

[0014] Compared with the existing technology, the medical imaging information management system based on the cloud platform provided by this application, in addition to obtaining the medical imaging acquisition equipment information, also introduces the software version, imaging protocol and image processing technology information when conducting the compatibility assessment of the pre-selected medical institutions, and adopts the semantic embedding coding technology based on deep learning to semantically encode the acquisition equipment information of the target imaging information and the acquisition information of the images of each pre-selected medical institution to capture the implicit relationship and contextual information between the data. Secondly, the acquisition equipment information of the target imaging information and the acquisition information of the pre-selected medical institution images are further semantically queried and matched to mine the significant correlation features between the two, thereby realizing the compatibility assessment of the pre-selected medical institutions. In this way, the accuracy of the compatibility assessment of the pre-selected medical institutions can be improved, providing a scientific basis for the optimal allocation and sharing of medical resources, and further improving the quality and efficiency of medical services. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 is a block diagram of a cloud-based medical imaging information management system according to an embodiment of the present application;

[0017] Figure 2 Schematic diagram of data flow of a cloud-based medical imaging information management system according to an embodiment of the present application;

[0018] Figure 3 This is a block diagram of a query encoding module in a cloud-based medical imaging information management system according to an embodiment of the present application. DETAILED DESCRIPTION

[0019] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0020] As used in this application and the claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0021] Although the present application makes various references to certain modules in the system according to embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are illustrative only, and different aspects of the system and method can use different modules.

[0022] Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the various steps may be processed in reverse order or simultaneously, as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0023] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0024] As mentioned in the above background technology, patent CN117316392A proposes a medical imaging information management system and method based on smart medical care, which includes: collecting and storing medical imaging information, establishing a medical imaging cloud database, the imaging information includes patient basic information, imaging acquisition equipment information, image content information and imaging acquisition medical institution information; obtaining target imaging information corresponding to the patient's basic information based on the medical imaging cloud database; determining the licensed medical institution based on the target imaging information, and the patient determines the target medical institution based on the licensed medical institution; reading the specific content of the target imaging information at the target medical institution based on the patient's basic information; the medical staff of the target medical institution diagnoses the patient's condition based on the specific content of the target imaging information, and obtains the patient's diagnosis results, and the patient understands his or her own condition based on the medical imaging information and the patient's diagnosis results.

[0025] When conducting compatibility assessments of medical institutions, the above-mentioned scheme only uses a simple similarity calculation formula to calculate the similarity between medical imaging acquisition devices as the basis for compatibility assessments of medical institutions. This method mainly relies on hardware information such as the model, brand, and specifications of the equipment, but ignores software-level compatibility, such as the version of the image processing software, protocol support, etc., which may result in the evaluation results being incomplete and inaccurate.

[0026] Based on this, in the technical solution of this application, a medical imaging information management system based on a cloud platform is proposed. Figure 1 This is a block diagram of a cloud-based medical imaging information management system according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of cloud-based medical imaging information management system according to an embodiment of the present application. Figure 1 and Figure 2As shown, according to an embodiment of the present application, a cloud-based medical imaging information management system 300 includes: a medical imaging information storage module 310 for collecting medical imaging information and storing the medical imaging information in a cloud-based medical imaging database of the cloud platform; a target imaging information acquisition module 320 for acquiring the acquisition device information of the target imaging information from the cloud-based medical imaging database based on the basic information of the patient; a pre-selected medical institution imaging information acquisition module 330 for acquiring the acquisition information of the images of the pre-selected medical institution, wherein the acquisition information includes acquisition device information, software version, imaging protocol and image processing technology; a semantic feature extraction module 340 , used to perform semantic embedding coding on the acquisition device information of the target image information and the acquisition information of the pre-selected medical institution image to obtain a sequence of target image acquisition device information semantic embedding coding vectors and pre-selected medical institution image information semantic embedding coding vectors; a query coding module 350, used to perform query coding on the sequence of target image acquisition device information semantic embedding coding vectors and pre-selected medical institution image information semantic embedding coding vectors to obtain target image query response enhanced coding features; a compatibility evaluation module 360, used to evaluate the compatibility of the pre-selected medical institution based on the target image query response enhanced coding features.

[0027] In particular, the medical imaging information storage module 310 is used to collect medical imaging information and store the medical imaging information in the cloud platform's medical imaging cloud database. The medical imaging information includes basic patient information. The medical imaging cloud database refers to a database system that stores medical imaging data on the cloud platform. Storing the medical imaging information in the cloud platform's medical imaging cloud database ensures that all medical imaging information, such as the acquisition device information for the medical imaging, is completely and accurately preserved. In particular, the cloud platform supports data sharing among multiple medical institutions, which can promote the optimal allocation of medical resources and collaborative work.

[0028] In particular, the target image information acquisition module 320 is used to obtain the acquisition device information of the target image information from the medical image cloud database based on the patient's basic information. That is, in the cloud platform-based medical image information management system of the present application, based on the patient's basic information (such as name, ID number, medical record number, etc.), the system can accurately identify and locate the medical image information of a specific patient. In one example, the patient extracts the acquisition device information of his medical image by entering personal information on the platform as the acquisition device information of the target image information, providing support for subsequent data processing and analysis. It should be understood that obtaining the acquisition device information of the target image information from the medical image cloud database based on the patient's basic information is a prior art. For example, it can adopt the principle disclosed in Chinese patent CN117316392A to obtain the acquisition device information of the target image information. Of course, it can also adopt other principles to obtain information, and this is not limited to the present application.

[0029] In particular, the pre-selected medical institution image information acquisition module 330 is used to obtain the acquisition information of the pre-selected medical institution's images, wherein the acquisition information includes acquisition equipment information, software version, imaging protocol and image processing technology. In the technical solution of the present application, the pre-selected medical institution is a medical institution connected to the medical image cloud database. Here, the acquisition equipment information, software version, imaging protocol and image processing technology reflect the equipment and technical status of the pre-selected medical institution. Through the cross-domain comprehensive analysis of the above-mentioned pre-selected medical institution image information, the comprehensiveness of the image data analysis of the pre-selected medical institution can be ensured, providing a comprehensive reference for the subsequent compatibility evaluation.

[0030] In particular, the semantic feature extraction module 340 is used to perform semantic embedding coding on the acquisition device information of the target image information and the acquisition information of the pre-selected medical institution images to obtain a sequence of semantic embedding coding vectors of the target image acquisition device information and semantic embedding coding vectors of the pre-selected medical institution image information. Considering that in medical image information management, different medical institutions may have different data formats, equipment models, software versions, etc., these differences make it difficult to directly compare and share data. Traditional data processing methods are usually limited to a single data domain, making it difficult to achieve cross-institutional data sharing and interoperability. In particular, through semantic embedding coding, it is possible to achieve a unified representation of data from different sources across different data domains, providing a basis for cross-domain queries. Therefore, in the technical solution of the present application, the acquisition device information of the target image information and the acquisition information of the pre-selected medical institution images are input into a semantic encoder based on the Bert model to obtain a sequence of semantic embedding coding vectors of the target image acquisition device information and semantic embedding coding vectors of the pre-selected medical institution image information. In particular, the Bert model is a deep learning model based on the Transformer architecture that excels at capturing implicit relationships and contextual information in text. By inputting the collection equipment information and the collection information of pre-selected medical institutions into the Bert model, the complex implicit relationships and contextual information between these information can be effectively captured, providing rich semantic representation. It can also achieve data standardization and unified representation between different data domains, providing strong support for cross-domain query and compatibility evaluation.

[0031] In particular, the query coding module 350 is used to query code the sequence of the target image acquisition device information semantic embedding coding vector and the pre-selected medical institution image information semantic embedding coding vector to obtain target image query response enhanced coding features. It is worth mentioning that in order to further improve the accuracy of the evaluation, the present application proposes a heterogeneous feature enhanced query coding method based on significant feature guidance, which is guided by the significant features in the sequence of the pre-selected medical institution image information semantic embedding coding vector, and strengthens the feature part of the target image acquisition device information semantic embedding coding vector that is significantly related to the pre-selected medical institution image information, thereby optimizing the query process and improving the efficiency and accuracy of query coding. In a specific example of the present application, such as Figure 3As shown, the query coding module 350 includes: a significant feature extraction unit 351, which is used to extract the significant features of the sequence of semantic embedded coding vectors of the pre-selected medical institution image information to obtain a semantic-guided benchmark; a heterogeneous feature enhancement query coding unit 352, which is used to perform heterogeneous feature enhancement query coding on the sequence of the target image acquisition device information semantic embedded coding vector and the pre-selected medical institution image information semantic embedded coding vector based on the semantic-guided benchmark to obtain a target image query response enhancement coding vector as the target image query response enhancement coding feature.

[0032] Specifically, the salient feature extraction unit 351 is used to extract the salient features of the sequence of semantic embedding coding vectors of the pre-selected medical institution image information to obtain a semantically guided benchmark. In a specific example of the present application, the salient feature extraction unit includes: a key matrix construction subunit, which is used to construct a key matrix based on the sequence of semantic embedding coding vectors of the pre-selected medical institution image information; a maximum value extraction subunit, which is used to extract the maximum value of each key vector in the key matrix to obtain a key matrix salient feature vector as a semantically guided benchmark. Specifically, first, in order to capture the relationship between each pre-selected medical institution image information semantic embedding coding vector in the sequence of semantic embedding coding vectors of the pre-selected medical institution image information so as to better represent the complex patterns in the data, the key embedding matrix is ​​used to linearly transform each pre-selected medical institution image information semantic embedding coding vector in the sequence of semantic embedding coding vectors of the pre-selected medical institution image information to map it to a new latent feature space, making it more suitable for the subsequent query encoding process. Next, the sequence of semantic embedding coding vectors of the pre-selected medical institution image information after the linear transformation is arranged in order as a key matrix for parallel processing.

[0033] More specifically, the key matrix construction subunit includes: a linear embedding coding secondary subunit, which is used to use the key embedding matrix to perform linear embedding coding on each pre-selected medical institution image information semantic embedding coding vector in the sequence of the pre-selected medical institution image information semantic embedding coding vector to obtain a sequence of pre-selected medical institution image information semantic embedding coding vectors after linear transformation; a matrix arrangement secondary subunit, which is used to use the pre-selected medical institution image information semantic embedding coding vector after linear transformation as the key vector and perform matrix arrangement on the sequence of pre-selected medical institution image information semantic embedding coding vectors after linear transformation to obtain the key matrix. Specifically, the process can be expressed as follows:

[0034] K={k1,k2,...,k n}

[0035] M o ={k1′;k2′;...;k n ′}

[0036] k i ′=f(k i ,W k )=k i W k +b k

[0037] Where K represents the sequence of semantic embedding coding vectors of the pre-selected medical institution image information, k1, k2, k i and k n are the first, second, i-th and n-th pre-selected medical institution image information semantic embedding coding vectors in the sequence of the pre-selected medical institution image information semantic embedding coding vectors, respectively. The value of n is the number of the pre-selected medical institution image information semantic embedding coding vectors. W k denote the key embedding matrix, b k Denote the key embedding bias vector, M o represents the key matrix, k1′, k2′, k i ′ and k n ′ are the first, second, i-th and n-th key vectors in the key matrix, that is, the semantic embedding coding vector of the pre-selected medical institution image information after the linear transformation.

[0038] Among them, the linear embedding coding secondary sub-unit is used to: multiply each pre-selected medical institution image information semantic embedding coding vector in the sequence of the pre-selected medical institution image information semantic embedding coding vector by the key embedding matrix and then add them positionally with the key embedding bias vector to obtain the sequence of the pre-selected medical institution image information semantic embedding coding vector after the linear transformation.

[0039] Then, the maximum value of each key vector is selected from the key matrix to obtain the key matrix salient feature vector. The key matrix salient feature vector is then used as a semantic guidance benchmark, representing the most prominent or representative part of the sequence of semantic embedding encoding vectors of the pre-selected medical institution image information after the linear transformation. Introducing the key matrix salient feature vector as an additional guidance signal helps reduce noise interference, strengthens the model's learning of important patterns, and also helps the attention mechanism focus more on the key parts of the data, thereby improving the model's attention to important details.

[0040] Specifically, the calculation process of the maximum value extraction subunit can be expressed as follows:

[0041] v tip ={max(k1′);max(k2′);...;max(k n ′)}

[0042] Among them, max(·) represents the maximum value function, v tip Represents the semantically oriented reference.

[0043] Specifically, the heterogeneous feature enhancement query encoding unit 352 is used to perform heterogeneous feature enhancement query encoding on the target image acquisition device information semantic embedding encoding vector and the pre-selected medical institution image information semantic embedding encoding vector based on the semantic guidance benchmark to obtain a target image query response enhancement encoding vector as the target image query response enhancement encoding feature. In a specific example of the present application, the heterogeneous feature enhancement query encoding unit includes: a target image acquisition device information linear embedding encoding subunit, used to use a query embedding matrix and a value embedding matrix to perform linear embedding encoding on the target image acquisition device information semantic embedding encoding vector to obtain a value vector and a query vector; a heterogeneous feature enhancement query encoding subunit, used to input the query vector, the value vector, the pre-selected medical institution image information semantic embedding encoding vector after each linear transformation in the key matrix, and the semantic guidance benchmark into a feature-guided cross-domain attention conversion layer to obtain a sequence of heterogeneous feature enhancement query encoding vectors; a query response enhancement encoding subunit, used to calculate the positional mean vector of the sequence of heterogeneous feature enhancement query encoding vectors to obtain the query response enhancement encoding vector as the target image query response encoding feature. Here, a query embedding matrix and a value embedding matrix are applied to the semantic embedding encoding vector of the target image acquisition device information to generate corresponding query and value vectors, respectively. A specially designed heterogeneous transformer structure is then used to interactively encode the query vector, the value vector, the key vectors in the key matrix, and the semantically guided benchmark. By introducing additional semantically guided benchmarks into the traditional attention mechanism, the weight distribution strategy is automatically adjusted to achieve effective optimization of the semantic embedding encoding vector of the target image acquisition device information. Finally, the sequence of heterogeneous feature-enhanced query encoding vectors is averaged by position, ultimately yielding a single query response enhanced encoding vector. This effectively mines the distribution of feature correlations between the target image information and the image information of each preselected medical institution.

[0044] Specifically, the target image acquisition device information linear embedding coding subunit is used to: multiply the semantic embedding coding vector of the target image acquisition device information by the query embedding matrix and then add it to the query embedding bias vector by position to obtain the query vector; multiply the semantic embedding coding vector of the target image acquisition device information by the value embedding matrix and then add it to the value embedding bias vector by position to obtain the value vector. Specifically, this process can be expressed as follows:

[0045] v q =x1W q +bq

[0046] v v =x1W v +b v

[0047] Wherein, x1 represents the semantic embedding coding vector of the target image acquisition device information, W q and W v denote the query embedding matrix and the value embedding matrix respectively, b q and b v They represent the query embedding bias vector and the value embedding bias vector, v q and v v represent the query vector and value vector respectively.

[0048] More specifically, the heterogeneous feature enhanced query encoding subunit is configured to: multiply the query vector by the transposed vector of the key vector and then divide the result by the two-norm of the semantically guided benchmark to obtain an attention score matrix; pass the attention score matrix through a softmax function and then multiply it by the semantically guided benchmark to obtain a template hint optimized attention weight vector; calculate the positional point multiplication between the template hint optimized attention weight vector and the value vector to obtain the heterogeneous feature enhanced query encoding vector. Specifically, the process can be expressed as follows:

[0049]

[0050] in,(·) T represents the transpose of the vector, ||·||2 represents the two-norm of the vector, and softmax is the normalized exponential function. Represents matrix multiplication operation, ⊙ represents positional multiplication, v pi represents the i-th heterogeneous feature enhanced query code vector in the sequence of heterogeneous feature enhanced query code vectors.

[0051] Specifically, the operation process of the query response enhancement coding subunit can be expressed as follows:

[0052]

[0053] Among them, v p represents the heterogeneous feature enhanced query encoding vector.

[0054] In particular, the compatibility assessment module 360 ​​is configured to assess the compatibility of the preselected medical institution based on the target image query response enhanced coding features. In a specific example of the present application, the target image query response enhanced coding vector is input into a decoder-based compatibility assessment module to obtain a compatibility assessment value. That is, by fully learning the target image query response information contained in the target image query response enhanced coding vector, the target image query response enhanced coding vector is decoded and mapped to the compatibility space of the preselected medical institution, thereby generating a compatibility assessment value.

[0055] In the technical solution of the present application, the target image acquisition device information semantic embedding coding vector represents the semantic embedding coding features of the target image acquisition device information, and the acquisition device information semantic embedding coding vector, software version semantic embedding coding vector, imaging protocol semantic embedding coding vector and image processing technology semantic embedding coding vector respectively represent the acquisition device information semantic embedding coding features, software version semantic embedding coding features, imaging protocol semantic embedding coding features and image processing technology semantic embedding coding features of the pre-selected medical institution. In this way, when the target image acquisition device information semantic embedding coding vector and the sequence of feature vectors composed of the acquisition device information semantic embedding coding vector, the software version semantic embedding coding vector, the imaging protocol semantic embedding coding vector and the image processing technology semantic embedding coding vector are input into the query coding optimization module with the key matrix significant features as the prompt template, the sparsity of the prompt template of the sequence of feature vectors superimposed on the correlation difference between the target image acquisition device information semantic embedding coding vector and each semantic embedding coding vector in the sequence of feature vectors will cause the feature manifold of the target image query response optimization coding vector to have expression sparsity and uneven distribution in the high-dimensional feature space, thereby affecting the accuracy of the compatibility evaluation value obtained by the decoder-based compatibility evaluation module.

[0056] Based on this, in a preferred embodiment, the target image query response optimized coding vector is input into a decoder-based compatibility evaluation module to obtain a compatibility evaluation value, including:

[0057] Performing feature modulation on the target image query response optimized coding vector relative to a target decoding domain to obtain an adapted target image query response optimized coding vector;

[0058] The adapted target image query response optimized coding vector is input into the decoder-based compatibility evaluation module to obtain the compatibility evaluation value.

[0059] Specifically, in this preferred embodiment, the process of performing feature modulation on the target image query response optimized coding vector relative to the target decoding domain to obtain an adapted target image query response optimized coding vector includes the following steps:

[0060] The global topological relationship mapping of the target image query response optimization coding vector is calculated to obtain the target image query response optimization global topological relationship matrix, which is expressed as:

[0061]

[0062] Among them, v i and v j They represent the i-th and j-th elements in the target image query response optimization encoding vector, D1 represents the global topological relationship matrix of the first target image, represents the (i, j)th element in the global topological relationship matrix of the first target image, D2 represents the global topological relationship matrix of the second target image, Represents the (i, j)th element in the global topological relationship matrix of the second target image, Represents matrix multiplication operation, and M represents the global topological relationship matrix of target image query response optimization.

[0063] Extract the decoding weight matrix M from the decoder-based compatibility assessment module d ;

[0064] The target image query response optimization global topology relationship matrix and the decoding weight matrix are fine-grainedly associated to obtain a target image query response global topology-decoding joint matrix, which is expressed as:

[0065] M t =Sigmoid(W t (αM+βM d ))

[0066] Among them, α represents the first weight hyperparameter, β represents the second weight hyperparameter, M d Denotes the decoding weight matrix, W t Represents the pre-training weight matrix, Sigmoid represents the S-type activation function, M t Represents the joint topology-decoding matrix of the target image query response.

[0067] The target image query response optimized encoding vector is recursively self-consistently decoded based on the target image query response global topology-decoding joint matrix to obtain a target image query recursive decoding optimization vector, which is expressed as:

[0068]

[0069] Where V represents the target image query response optimized encoding vector, V t Represents the target image query recursive decoding optimization vector.

[0070] Based on the target image query recursive decoding optimization vector, a target image query recursive decoding optimization weight vector is constructed, which is expressed as:

[0071] V a =Softmax(V t )

[0072] Among them, Softmax represents the normalized exponential function, V a Represents the target image query recursive decoding optimization weight vector.

[0073] The target image query recursive decoding optimization weight vector is used to perform feature modulation on the target image query response optimization coding vector to obtain the adapted target image query response optimization coding vector, which is expressed as:

[0074] V f =V a ⊙V

[0075] Among them, ⊙ represents the point product by position, V f Represents the optimized encoding vector of the adapted target image query response.

[0076] Accordingly, in this preferred embodiment, the global topological relationship mapping of the target image query response optimized coding vector is semantically fused with the decoding weight matrix of the decoder, and then the target image query response optimized coding vector is optimized based on recursive self-consistent coding based on the information matrix of the fused target decoding information to improve the detail and structure of the feature expression, thereby improving the accuracy of the compatibility evaluation value obtained by the decoder-based compatibility evaluation module.

[0077] As described above, the cloud-based medical imaging information management system 300 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a cloud-based medical imaging information management algorithm. In one possible implementation, the cloud-based medical imaging information management system 300 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or hardware module. For example, the cloud-based medical imaging information management system 300 can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the cloud-based medical imaging information management system 300 can also be one of the many hardware modules of the wireless terminal.

[0078] Alternatively, in another example, the cloud-based medical imaging information management system 300 and the wireless terminal may also be separate devices, and the cloud-based medical imaging information management system 300 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0079] While various embodiments of the present disclosure have been described above, the above descriptions are illustrative, non-exhaustive, and not intended to be limiting of the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or improvements to existing technologies, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A cloud-based medical imaging information management system, characterized in that: include: A medical imaging information storage module is used to collect medical imaging information and store the medical imaging information in a medical imaging cloud database of a cloud platform; A target image information acquisition module is used to acquire acquisition device information of the target image information from the medical image cloud database based on the patient's basic information; A module for acquiring image information of a pre-selected medical institution, configured to acquire image acquisition information of the pre-selected medical institution, wherein the acquisition information includes acquisition device information, software version, imaging protocol, and image processing technology; a semantic feature extraction module, configured to perform semantic embedding coding on each data item in the acquisition device information of the target image information and the acquisition information of the pre-selected medical institution images to obtain a sequence of semantic embedding coding vectors of the target image acquisition device information and semantic embedding coding vectors of the pre-selected medical institution image information; A query coding module, used to perform query coding on the sequence of the target image acquisition device information semantic embedding coding vector and the pre-selected medical institution image information semantic embedding coding vector to obtain a target image query response enhanced coding feature, including: a significant feature extraction unit, used to extract the significant features of the sequence of the pre-selected medical institution image information semantic embedding coding vector to obtain a semantic-guided benchmark; a heterogeneous feature enhanced query coding unit, used to perform heterogeneous feature enhanced query coding on the sequence of the target image acquisition device information semantic embedding coding vector and the pre-selected medical institution image information semantic embedding coding vector based on the semantic-guided benchmark to obtain a target image query response enhanced coding vector as the target image query response enhanced coding feature; A compatibility evaluation module is used to evaluate the compatibility of the preselected medical institutions based on the enhanced coding features of the target image query response.

2. The cloud-based medical imaging information management system according to claim 1, characterized in that: The semantic feature extraction module is used to: The data items in the acquisition device information of the target image information and the acquisition information of the pre-selected medical institution images are input into a semantic encoder based on the Bert model to obtain a sequence of the semantic embedding coding vector of the target image acquisition device information and the semantic embedding coding vector of the pre-selected medical institution image information.

3. The cloud-based medical imaging information management system according to claim 2, characterized in that: The salient feature extraction unit includes: A key matrix construction subunit, configured to construct a key matrix based on a sequence of semantic embedding coding vectors of the preselected medical institution imaging information; The maximum value extraction subunit is used to extract the maximum value of each key vector in the key matrix to obtain a significant feature vector of the key matrix as a semantic guidance benchmark.

4. The cloud-based medical imaging information management system according to claim 3, characterized in that: The key matrix construction subunit includes: a linear embedding coding secondary subunit, configured to perform linear embedding coding on each of the pre-selected medical institution image information semantic embedding coding vectors in the sequence of the pre-selected medical institution image information semantic embedding coding vectors using a key embedding matrix to obtain a sequence of linearly transformed pre-selected medical institution image information semantic embedding coding vectors; The matrix arrangement secondary subunit is used to use the semantic embedding coding vector of the pre-selected medical institution image information after the linear transformation as the key vector, and to perform matrix arrangement on the sequence of the semantic embedding coding vector of the pre-selected medical institution image information after the linear transformation to obtain the key matrix.

5. The cloud-based medical imaging information management system according to claim 4, characterized in that: The linear embedded coding secondary sub-unit is used to: Each preselected medical institution image information semantic embedding coding vector in the sequence of the preselected medical institution image information semantic embedding coding vectors is multiplied by the key embedding matrix and then added positionally to the key embedding bias vector to obtain the sequence of the preselected medical institution image information semantic embedding coding vectors after the linear transformation.

6. The cloud-based medical imaging information management system according to claim 5, characterized in that: The heterogeneous feature enhanced query encoding unit includes: a target image acquisition device information linear embedding coding subunit, configured to perform linear embedding coding on the target image acquisition device information semantic embedding coding vector using a query embedding matrix and a value embedding matrix to obtain a value vector and a query vector; a heterogeneous feature enhanced query encoding subunit, configured to input the query vector, the value vector, the semantic embedding encoding vector of the preselected medical institution imaging information after each linear transformation in the key matrix, and the semantic guidance benchmark into a feature-guided cross-domain attention conversion layer to obtain a sequence of heterogeneous feature enhanced query encoding vectors; The query response enhancement coding subunit is used to calculate the positional mean vector of the sequence of the heterogeneous feature enhanced query coding vectors to obtain the query response enhancement coding vector as the target image query response coding feature.

7. The cloud-based medical imaging information management system according to claim 6, characterized in that: The target image acquisition device information linear embedding encoding subunit is used to: Multiplying the target image acquisition device information semantic embedding encoding vector by the query embedding matrix and then adding the result to the query embedding bias vector positionally to obtain the query vector; The target image acquisition device information semantic embedding coding vector is multiplied by the value embedding matrix and then added to the value embedding bias vector by position to obtain the value vector.

8. The cloud-based medical imaging information management system according to claim 7, characterized in that: The heterogeneous feature enhanced query encoding subunit is used to: Multiplying the query vector by the transposed vector of the key vector and dividing by the two-norm of the semantic-oriented benchmark to obtain an attention score matrix; Passing the attention score matrix through a softmax function and multiplying it by the semantically guided benchmark to obtain a template-cued optimized attention weight vector; The position-wise multiplication between the template hint optimized attention weight vector and the value vector is calculated to obtain the heterogeneous feature enhanced query encoding vector.

9. The cloud-based medical imaging information management system according to claim 8, characterized in that: The compatibility evaluation module is used to: input the target image query response enhanced coding vector into the decoder-based compatibility evaluation module to obtain a compatibility degree evaluation value.

Images