A method, device and storage medium for optimizing storage and reading of medical big data

Abstract

The present invention provides a storage and reading optimization method, device and storage medium for medical big data. The method can respond to the storage task of the terminal, sequentially obtain the identification information of each medical data file in the storage task, match the identification information with the local index information, and further obtain the characteristic information of the medical data file for matching when the matching degree meets the conditions, thereby deleting low-value files in the storage task, and can also perform secondary deletion of redundant files on the server side. The scheme proposed by the present invention can effectively avoid uploading low-value data by performing double matching of identification information and characteristic information of medical data files in the storage task before uploading, reduce the burden of network transmission and the processing load of the server, and at the same time, combined with the deletion of redundant files on the server side, effectively prevent the waste of server storage space, and has the effect of improving data management efficiency and optimizing storage space utilization.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical data management, and specifically relates to a storage and reading optimization method, device and storage medium for medical big data. Background Art

[0002] With the rapid development of medical information technology, medical data has exploded, including but not limited to patient medical records, medical images, physiological indicator monitoring data, etc. These data are of great value for medical research, clinical decision support, health management, and public health monitoring.

[0003] The huge amount of data has brought unprecedented challenges to the storage, management and utilization of medical data systems. Most existing medical data systems rely on centralized server architecture, which facilitates centralized access and analysis of data and enables medical institutions or research institutions to upload collected medical data to central servers for storage and management via the Internet.

[0004] However, the servers of existing medical data systems often contain a large amount of low-value data that is the same or similar to other files, resulting in data redundancy and waste of storage resources. In addition, uploading a large amount of data directly to the server will greatly increase the consumption of network bandwidth, bring a heavy burden to network transmission, and cause slow data transmission speed, which greatly affects the timeliness and efficiency of data upload. Summary of the invention

[0005] In order to overcome the defects of the above-mentioned prior art, the present invention proposes a storage and reading optimization method for medical big data, the method comprising:

[0006] In response to the first storage task of the terminal, sequentially acquiring identification information of the medical data files to be matched in the first storage task, so as to match the identification information with the local index information table of the terminal;

[0007] If the index information whose matching degree with the identification information satisfies the first preset condition is matched, the characteristic information of the medical data file is obtained, and the characteristic information is matched with the specific content corresponding to the index information; if the index information whose matching degree with the identification information satisfies the first preset condition is not matched, or the matching degree between the characteristic information and the specific content corresponding to the index information does not satisfy the second preset condition, the matching of the medical data file is completed;

[0008] If the matching degree between the characteristic information and the specific content corresponding to the index information meets a second preset condition, deleting the medical data file from the first storage task;

[0009] After completing the matching of all the medical data files in the first storage task, executing the first storage task, and updating the index information table based on the first storage task;

[0010] In response to the first storage tasks from the multiple terminals, the server receives the medical data files included in the first storage tasks, and deletes the redundant medical data files included in the first storage tasks from the multiple terminals.

[0011] Furthermore, the method further comprises:

[0012] In response to a first request task of a terminal, parsing the first request task to determine first type information and first quantity information, and searching a local database of the terminal based on the first type information;

[0013] If the number of medical data files retrieved from the local database meets the first number information, displaying the retrieved medical data files to the user;

[0014] If the number of medical data files retrieved from the local database does not satisfy the first quantity information, a second request task for requesting server files from the database of the server is generated and executed based on the first type information, the first quantity information and the retrieved quantity information of the medical data files, so that the terminal obtains the server files from the server, and merges the medical data files with the server files to display them to the user.

[0015] Preferably, each terminal is pre-divided into several groups, each group includes at least two terminals, and each terminal belongs to at most one group; the method further includes:

[0016] After completing the matching of all the medical data files in the first storage task, generating a second storage task for storing the medical data files included in the first storage task in a local database of the terminal;

[0017] Obtaining the space capacity occupied by the medical data files stored in the local database of the terminal;

[0018] If the space capacity occupied by the stored medical data files does not exceed the first preset capacity value, executing the second storage task on the terminal;

[0019] If the space capacity occupied by the stored medical data files exceeds the first preset capacity value, the second storage task is executed at another terminal belonging to the same group as the terminal, so that when responding to the first request task of any terminal, the local databases of all terminals belonging to the same group as the terminal are retrieved through the first type of information.

[0020] Furthermore, the method further comprises:

[0021] After executing the second storage task, obtaining the space capacity occupied by the medical data file stored in the local database of the terminal;

[0022] If the space capacity occupied by the stored medical data files exceeds a second preset capacity value, obtaining identification information of all the medical data files in the local database, and dividing all the medical data files into a plurality of categories based on the identification information;

[0023] A similar file group is generated based on at least two medical data files in the same category whose matching degree of the identification information meets a first preset condition, and feature information of each medical data file in each similar file group is compared respectively, and the similar file group of medical data files having at least one set of feature information whose matching degree meets a second preset condition is displayed to the user to prompt the user to make corresponding deletions.

[0024] Preferably, the method further comprises:

[0025] After prompting the user to perform the corresponding deletion, obtaining the user's decision information on whether to delete the medical data files in the similar file groups of each category, and using the decision information as a training sample for a deep learning model;

[0026] The second preset condition is updated using the trained deep learning model, so that after a similar file group is generated based on at least two of the medical data files whose matching degree of the identification information in the same category meets the first preset condition, the step of "respectively comparing the characteristic information of each of the medical data files in each of the similar file groups" is performed based on the updated second preset condition.

[0027] Accordingly, the “if the index information whose matching degree with the identification information satisfies the first preset condition is matched, acquiring the characteristic information of the medical data file, and matching the characteristic information with the specific content corresponding to the index information” includes:

[0028] If index information is matched whose matching degree with the identification information satisfies the first preset condition, the category to which the medical data file belongs is determined and characteristic information of the medical data file is obtained, so as to match the characteristic information with the specific content corresponding to the index information based on the updated second preset condition.

[0029] Furthermore, the method further comprises:

[0030] If the space capacity occupied by the stored medical data files exceeds a third preset capacity value, the storage time of each medical data file in the local database is obtained, and the medical data files are deleted in sequence based on the storage time until the space capacity occupied by the stored medical data files does not exceed a fourth preset capacity value; the third preset capacity value is not less than the second preset capacity value and the fourth preset capacity value.

[0031] Preferably, each of the characteristic information corresponds to a different content part in the medical data file, and the “if the matching degree between the characteristic information and the specific content corresponding to the index information satisfies the second preset condition, deleting the medical data file from the first storage task” includes:

[0032] If the proportion of the characteristic information whose matching degree with the specific content corresponding to the index information satisfies the third preset condition in all characteristic information is not less than the preset proportion value, it is determined that the matching degree between the characteristic information and the specific content corresponding to the index information satisfies the second preset condition, and the medical data file is deleted from the first storage task;

[0033] An update file is generated based on the content portion corresponding to the feature information whose matching degree with the specific content corresponding to the index information does not satisfy the third preset condition, so as to add the update file to the first storage task.

[0034] The present invention also proposes a storage and reading optimization device for medical big data, the device comprising:

[0035] A response module, configured to respond to the first storage task of the terminal, and sequentially obtain identification information of the medical data files to be matched in the first storage task, so as to match the identification information with the local index information table of the terminal;

[0036] A matching module, configured to obtain characteristic information of the medical data file and match the characteristic information with specific content corresponding to the index information when matching index information with the identification information and satisfying a first preset condition; and to complete matching of the medical data file if matching index information with the identification information and satisfying the first preset condition is not matched, or matching between the characteristic information and specific content corresponding to the index information and does not satisfy a second preset condition;

[0037] a deleting module, configured to delete the medical data file from the first storage task when the matching degree between the characteristic information and the specific content corresponding to the index information satisfies a second preset condition;

[0038] An execution module, configured to execute the first storage task after completing the matching of all the medical data files in the first storage task, and update the index information table based on the first storage task;

[0039] The processing module is used to respond to the first storage task from multiple terminals, enable the server to receive the medical data files included in the first storage task, and delete the redundant medical data files included in the first storage task from multiple terminals on the server.

[0040] The present invention also proposes a computer-readable storage medium, which stores executable instructions for implementing the storage and reading optimization method of medical big data as described above when executed by a processor.

[0041] The present invention has at least the following beneficial effects:

[0042] The solution proposed in the present invention can effectively identify new files with a high similarity to existing files on the server and cancel the upload by performing feature matching and content matching on medical data files before uploading, thereby avoiding the uploading and storage of low-value data. This is particularly important for the storage management of massive medical data, and can significantly reduce the waste of storage space and storage costs. It is effectively combined with the removal of redundant data on the server side, which helps to reasonably allocate and utilize server resources, while reducing the need for manual intervention, improving the efficiency and accuracy of data management, and providing a more efficient, secure and economical solution for the storage and management of medical big data.

[0043] Furthermore, this solution can significantly improve data access speed and reduce network delay and bandwidth consumption by retrieving data in the local database first. When there is enough data in the local database, users can obtain it immediately without waiting for the server to respond. When the amount of data is insufficient, the server file can be combined to meet user needs. Through terminal grouping and data sharing, the storage space of the terminals in the group can be effectively utilized to avoid the problem of insufficient storage space of a single terminal. When the storage space of a terminal is close to the preset upper limit, the data can be migrated to other terminals in the same group, which not only ensures the availability of data but also improves storage efficiency. At the same time, when responding to download requests, data can be retrieved in a wider range, which improves the success rate of data retrieval.

[0044] Preferably, this solution can also generate similar file groups by classifying files to help users quickly identify and clean up redundant data in the terminal, thereby intelligently managing local database space and avoiding storage space waste. On this basis, the deep learning model is used to dynamically adjust the proportion value applied to each type of file, which can more accurately identify similar files of various types, improve the accuracy of feature information comparison, enable the system to better adapt to different types of medical data files, and improve the intelligence level and adaptive ability of the system.

[0045] In addition, this solution can more finely identify the differences between medical data files and index information by matching each feature information separately. When there is a partial match, only the updated file generated based on the unmatched feature information can be uploaded instead of the entire file, which can significantly reduce the amount of data transmission, improve upload efficiency, and ensure the integrity and consistency of the data, avoiding data duplication and redundancy. When the local database storage space is extremely tight, the deletion strategy based on storage time can be used to delete older data files first to make room for new data files, ensuring the timeliness of the data.

[0046] Therefore, the present invention provides a storage and reading optimization method, device and storage medium for medical big data. The scheme proposed in the present invention can effectively avoid uploading low-value data by performing double matching of identification information and feature information of medical data files in storage tasks before uploading, thereby reducing the burden of network transmission and the processing load of the server. At the same time, combined with the deletion of redundant files on the server side, it effectively prevents the waste of server storage space, and has the effect of improving data management efficiency and optimizing storage space utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0048] Figure 1 A schematic diagram of a method flow of a method for optimizing storage and reading of medical big data provided in Example 1;

[0049] Figure 2 A flow chart of a method for requesting data for a terminal;

[0050] Figure 3 A flow chart of a method for storing data in groups for multiple terminals;

[0051] Figure 4 This is an example diagram for dividing multiple terminal groups;

[0052] Figure 5 A flowchart of a method for cleaning terminal local storage and optimizing matching for classified files;

[0053] Figure 6 A schematic diagram of the module structure of the device for the storage and reading optimization method of medical big data provided in Example 2.

[0054] Reference numerals

[0055] 10-response module; 20-matching module; 30-deletion module; 40-update module; 50-execution module; 60-processing module; 71-retrieval module; 72-display module; 73-integration module; 81-generation module; 82-acquisition module; 83-storage module; 91-query module; 92-division module; 93-prompt module; 94-training module; 95-correction module. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Hereinafter, various embodiments of the present invention will be described more fully. The present invention may have various embodiments, and adjustments and changes may be made therein. However, it should be understood that there is no intention to limit the various embodiments of the present invention to the specific embodiments disclosed herein, but the present invention should be understood to cover all adjustments, equivalents and / or alternatives falling within the spirit and scope of the various embodiments of the present invention.

[0058] Hereinafter, the term "include" or "may include" used in various embodiments of the present invention indicates the presence of disclosed functions, operations or elements, and does not limit the addition of one or more functions, operations or elements. In addition, as used in various embodiments of the present invention, the terms "include", "have" and their cognates are intended only to indicate specific features, numbers, steps, operations, elements, components or combinations of the foregoing items, and should not be understood as first excluding the presence of one or more other features, numbers, steps, operations, elements, components or combinations of the foregoing items or the possibility of adding one or more features, numbers, steps, operations, elements, components or combinations of the foregoing items.

[0059] In various embodiments of the present invention, the expression "or" or "at least one of A or / and B" includes any combination or all combinations of the words listed at the same time. For example, the expression "A or B" or "at least one of A or / and B" may include A, may include B, or may include both A and B.

[0060] The expressions (such as "first", "second", etc.) used in various embodiments of the present invention may modify the various constituent elements in various embodiments, but may not limit the corresponding constituent elements. For example, the above expressions do not limit the order and / or importance of the elements. The above expressions are only used for the purpose of distinguishing an element from other elements. For example, the first user device and the second user device indicate different user devices, although both are user devices. For example, without departing from the scope of various embodiments of the present invention, the first element may be referred to as the second element, and similarly, the second element may also be referred to as the first element.

[0061] It should be noted that in the present invention, unless otherwise clearly specified and defined, the terms such as "installation", "connection", "fixation" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0062] In the present invention, those skilled in the art need to understand that the terms indicating orientation or positional relationship herein are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0063] The terms used in various embodiments of the present invention are only used for the purpose of describing specific embodiments and are not intended to limit various embodiments of the present invention. As used herein, the singular form is intended to also include the plural form, unless the context clearly indicates otherwise. Unless otherwise limited, all terms used here (including technical terms and scientific terms) have the same meaning as the meaning generally understood by those of ordinary skill in the art to which the various embodiments of the present invention belong. The terms (such as the terms defined in the dictionary generally used) will be interpreted as having the same meaning as the contextual meaning in the relevant technical field and will not be interpreted as having an idealized meaning or an overly formal meaning, unless clearly defined in various embodiments of the present invention.

[0064] Example 1

[0065] This embodiment proposes a method for optimizing the storage and reading of medical big data. Figure 1 , the method comprising:

[0066] S100: In response to a first storage task of a terminal, sequentially obtain identification information of medical data files to be matched in the first storage task to match the identification information with a local index information table of the terminal.

[0067] Specifically, the identification information can be the title information, summary information or metadata of the medical data file, such as file type, keywords, etc. The identification information can be used to help identify and match whether there are other file identifiers similar to the medical data file in the index information table, thereby determining whether further comparison of the file content is needed.

[0068] S200: If the index information whose matching degree with the identification information satisfies the first preset condition is matched, characteristic information of the medical data file is obtained, and the characteristic information is matched with specific content corresponding to the index information.

[0069] In this embodiment, the index information table includes index information of files in the local database of the terminal and historical files uploaded from the terminal to the server. When the similarity between the identification information and any index information reaches a preset standard value, it can be confirmed that the matching degree between the index information and the identification information meets the first preset condition;

[0070] If the index information whose matching degree with the identification information satisfies the first preset condition is not matched, or the matching degree between the characteristic information and the specific content corresponding to the index information does not meet the second preset condition, the matching of the medical data file is completed.

[0071] It should be noted that the number of index information matched in step S100 may be several, and therefore the matching of the feature information with the specific content of the index information in step S200 may be a one-to-one or one-to-many match.

[0072] Specifically, the characteristic information may include specific data or information parts corresponding to various important parts in the medical data file, such as chapter names, report records, experimental results, etc.

[0073] S300: If the matching degree between the characteristic information and the specific content corresponding to the index information meets the second preset condition, the medical data file is deleted from the first storage task.

[0074] Preferably, each characteristic information corresponds to a different content part in the medical data file. When the matching degree of the specific content corresponding to the index information satisfies the third preset condition and the proportion of the characteristic information in all the characteristic information is not less than the preset proportion value, it can be determined that the matching degree of the characteristic information and the specific content corresponding to the index information satisfies the second preset condition.

[0075] S400: Generate an update file based on the content portion corresponding to the feature information whose matching degree with the specific content corresponding to the index information does not satisfy the third preset condition, so as to add the update file to the first storage task.

[0076] S500: After completing the matching of all medical data files in the first storage task, execute the first storage task, and update the index information table based on the first storage task.

[0077] S600: In response to a first storage task from a plurality of terminals, enabling a server to receive medical data files included in the first storage task, and deleting redundant medical data files included in the first storage task from a plurality of terminals at the server.

[0078] It should be noted that the server may receive the same or similar medical data files from multiple terminals, and thus generate redundant medical data files. Optionally, step S600 may match the medical data files in the first storage task with the existing server files in the server database by comparing the original byte data, data blocks and / or hash values;

[0079] Exemplarily, the method proposed in this embodiment can divide the medical data file in the first storage task into data blocks of fixed size or variable size. Each data block will generate a unique identifier, and a hash value used to identify the uniqueness of the block content is calculated for each data block. And by checking whether the hash value already exists in the index table of the server, it is confirmed whether the data block has been stored, thereby adding the hash value of the non-duplicate data block to the index table and deleting the duplicate data blocks, and finally integrating the non-duplicate data blocks to form a new server file.

[0080] Further, see Figure 2 , the method further comprises:

[0081] S710: In response to a first request task of the terminal, parse the first request task to determine first type information and first quantity information, and search a local database of the terminal based on the first type information.

[0082] Specifically, the first type information may include several categories, and the first quantity information may include the required quantity of files of each category.

[0083] S720: If the number of medical data files retrieved from the local database meets the first number information, the retrieved medical data files are displayed to the user.

[0084] S730: If the number of medical data files retrieved from the local database does not satisfy the first quantity information, a second request task for requesting the server file from the database of the server is generated and executed based on the first type information, the first quantity information and the quantity information of the retrieved medical data files, so that the terminal obtains the server file from the server, and merges the medical data file with the server file for display to the user.

[0085] Specifically, the manner of presenting the retrieved medical data files to the user and merging and organizing the medical data files with the server files for presentation to the user may include but is not limited to graphical presentation, classification presentation, and the like.

[0086] Preferably, each terminal is pre-divided into several groups, each group includes at least two terminals, and each terminal belongs to at most one group, see Figure 3-4 , the method further comprises:

[0087] S810: After completing the matching of all medical data files in the first storage task, a second storage task is generated to store the medical data files included in the first storage task into a local database of the terminal.

[0088] In a specific embodiment, the basis for dividing the terminals into groups may include but is not limited to terminal functions, departments where the terminals are located, terminal authority levels, etc. For example, when the terminal functions are used as the basis for dividing the terminals into groups, multiple groups such as clinical terminals, management terminals, and logistics terminals can be divided. When the departments where the terminals are located are used as the basis for dividing the terminals into groups, multiple groups such as emergency department terminals, internal medicine terminals, surgical terminals, laboratory terminals, and pharmacy terminals can be divided.

[0089] S820: Obtain the space capacity occupied by the medical data files stored in the local database of the terminal.

[0090] S830: If the space occupied by the stored medical data files does not exceed the first preset capacity value, execute the second storage task at the terminal; otherwise, execute the second storage task at another terminal belonging to the same group as the terminal.

[0091] Therefore, when responding to the first request task of any terminal, the method proposed in this embodiment can search the local databases of all terminals belonging to the same group as the terminal through the first type of information.

[0092] Specifically, when the terminal executes the second storage task, the index information table is also updated based on the second storage task.

[0093] As mentioned above, multiple terminals connected to the server can be divided into several groups. The medical data files uploaded by the same terminal or all terminals in the same group will have a higher category similarity, and the user using the terminal has a higher demand for the medical data files uploaded by the terminal and other terminals belonging to the same group as the terminal. Therefore, the method proposed in this embodiment can first search from the local database of the terminal and other terminals belonging to the same group as the terminal when the user requests to obtain data;

[0094] Since terminals belonging to the same group can use the same local area network, the speed at which the terminals obtain data from the local database and other terminals in the same local area network is much higher than the speed at which they download data from the server. Therefore, the method proposed in this embodiment can effectively save time and server resources.

[0095] Further, see Figure 5 , the method further comprises:

[0096] S910: After executing the second storage task, obtain the space capacity occupied by the medical data files stored in the local database of the terminal.

[0097] S920: If the space capacity occupied by the stored medical data files exceeds the second preset capacity value, obtain identification information of all medical data files in the local database, and divide all medical data files into several categories based on the identification information.

[0098] Specifically, the method proposed in this embodiment can use the metadata or content analysis technology of the file to determine the category of the medical data file, such as image data, text records, experimental results, etc.

[0099] In this embodiment, the second preset capacity value is not greater than the first preset capacity value.

[0100] S930: Generate a similar file group based on at least two medical data files in the same category whose matching degree of identification information meets a first preset condition, compare the feature information of each medical data file in each similar file group, and display the similar file group of medical data files with at least one set of feature information whose matching degree meets a second preset condition to the user, so as to prompt the user to delete them accordingly.

[0101] In this embodiment, in step S930, a similar file group of medical data files having at least one set of feature information matching degree satisfying the second preset condition is displayed to the user, so as to prompt the user to delete them. Specifically, each medical data file having feature information with a repetition ratio exceeding a preset ratio value in each file and each medical data file having feature information with a repetition ratio exceeding a preset ratio value in each file is displayed to the user, so as to prompt the user to delete at least one file in each group.

[0102] For example, when the preset ratio is 90%, if a similar file group includes three medical data files A, B, and C, wherein the duplication ratio of the characteristic information in file A is 91.5%, the duplication ratio of the characteristic information in file B is 75.6%, and the duplication ratio of the characteristic information in file C is 90.3%, then the method proposed in this embodiment will display the three medical data files A, B, and C in the similar file group and the similarities between them to the user, and prompt the user to delete one of file A and file B;

[0103] At the same time, in addition to deleting one of file A and file B, the user can also view file C to decide whether to delete file C. If the user believes that the content contained in file A and / or file B is sufficient to overwrite the content contained in file C, the user can also choose to delete file C and choose to keep file A and file B at the same time, or delete one of file A and file B.

[0104] S940: Obtain the user's decision information on whether to delete the medical data files in each category of similar file groups, and use the decision information as a training sample for the deep learning model.

[0105] Specifically, the method proposed in this embodiment can be combined with a deep learning model to predict the value that needs to be reached for the similarity of similar medical data files in different categories, thereby updating the second preset condition based on historical data analysis and similar features of the predicted files through the training model, and optimizing the adjustment rules according to the model output.

[0106] S950: Update the second preset condition using the trained deep learning model.

[0107] Therefore, the method proposed in this embodiment can realize the targeted correction of the second preset condition used for medical data files of different categories. After generating a similar file group based on at least two medical data files whose matching degree of identification information in the same category meets the first preset condition, the step of "respectively comparing the characteristic information of each medical data file in each similar file group" described in step S730 is executed through the updated second preset condition.

[0108] Preferably, the method proposed in this embodiment can also set different initial preset ratio values ​​for each different category of medical data files, for example, setting an initial preset ratio value of 90% for image data, setting an initial preset ratio value of 80% for text data, and setting an initial preset ratio value of 85% for experimental data.

[0109] Correspondingly, after the second preset condition is modified, the category to which the medical data file belongs is determined in step S200, so that the feature information is matched with the specific content corresponding to the index information based on the updated second preset condition.

[0110] Furthermore, after the deep learning model completes training, the decision information generated by step S930 will still be input into the deep learning model as a training sample, thereby achieving continuous optimization and adaptive adjustment of the deep learning model. The deep learning model can continuously adjust its parameters and classification strategies based on new decision information to improve the accuracy of recognition and classification of similar medical data files. As a result, over time, the model can more accurately classify and process medical data files, thereby providing more effective support in storage space management.

[0111] In addition, as the deep learning model is continuously optimized, the system will be able to provide smarter suggestions and automated processing solutions, thereby reducing the user's workload in managing medical data files and improving the system's automation and intelligence levels.

[0112] Furthermore, the method further comprises:

[0113] If the space capacity occupied by the stored medical data files exceeds the third preset capacity value, the storage time of each medical data file in the local database is obtained, and the medical data files are deleted in sequence based on the storage time until the space capacity occupied by the stored medical data files does not exceed the fourth preset capacity value.

[0114] In this embodiment, the "deleting medical data files in sequence based on storage time" is achieved by deleting the medical data files in sequence starting from the earliest stored medical data file based on the order of storage time of each medical data file; specifically, the third preset capacity value is not less than the second preset capacity value and the fourth preset capacity value.

[0115] Example 2

[0116] This embodiment proposes a storage and reading optimization device for medical big data, which is used to implement the method proposed in Example 1. Figure 6 , the device comprises:

[0117] A response module 10, for responding to the first storage task of the terminal, sequentially acquiring identification information of the medical data files to be matched in the first storage task, so as to match the identification information with the local index information table of the terminal;

[0118] A matching module 20, configured to obtain characteristic information of the medical data file and match the characteristic information with specific content corresponding to the index information when matching the index information with the identification information to meet the first preset condition;

[0119] A deletion module 30, configured to delete the medical data file from the first storage task when the matching degree between the characteristic information and the specific content corresponding to the index information meets a second preset condition;

[0120] An updating module 40, configured to generate an update file based on the content portion corresponding to the feature information whose matching degree with the specific content corresponding to the index information does not satisfy the third preset condition, so as to add the update file to the first storage task;

[0121] An execution module 50, configured to execute the first storage task after completing the matching of all medical data files in the first storage task, and update the index information table based on the first storage task;

[0122] The processing module 60 is used to respond to the first storage task from multiple terminals, enable the server to receive the medical data files included in the first storage task, and delete the redundant medical data files included in the first storage task from multiple terminals on the server.

[0123] Specifically, the identification information may be the title information, summary information or metadata of the medical data file, such as the file type, keywords, etc. The identification information may be used to help identify and match whether there are other file identifiers similar to the medical data file in the index information table, so as to determine whether further comparison of the file contents is required;

[0124] In this embodiment, the index information table includes index information of files in the local database of the terminal and historical files uploaded from the terminal to the server. When the similarity between the identification information and any index information reaches a preset standard value, it can be confirmed that the matching degree between the index information and the identification information meets the first preset condition.

[0125] The characteristic information may include specific data or information corresponding to each important part of the medical data file, such as chapter name, report record, experimental results, etc.;

[0126] Preferably, each characteristic information corresponds to a different content part in the medical data file. When the matching degree of the specific content corresponding to the index information satisfies the third preset condition and the proportion of the characteristic information in all the characteristic information is not less than the preset proportion value, it can be determined that the matching degree of the characteristic information and the specific content corresponding to the index information satisfies the second preset condition.

[0127] If the matching module 20 does not match the index information whose matching degree with the identification information meets the first preset condition, or the matching degree between the characteristic information and the specific content corresponding to the index information does not meet the second preset condition, the matching of the medical data file is completed.

[0128] Optionally, the processing module 60 may match the medical data file in the first storage task with the existing server files in the server database by comparing raw byte data, data blocks and / or hash values.

[0129] Furthermore, the device also includes:

[0130] A retrieval module 71, configured to respond to a first request task of a terminal, parse the first request task to determine first type information and first quantity information, and search a local database of the terminal based on the first type information;

[0131] A display module 72, configured to display the retrieved medical data files to the user when the number of medical data files retrieved from the local database satisfies the first number information;

[0132] an integration module 73, configured to generate and execute a second request task for requesting a server file from a database of a server based on the first type information, the first quantity information and the quantity information of the retrieved medical data files when the quantity of the medical data files retrieved from the local database does not satisfy the first quantity information, so that the terminal obtains the server file from the server, and merges and organizes the medical data file with the server file to display to the user;

[0133] A generating module 81, for generating a second storage task for storing the medical data files included in the first storage task in a local database of the terminal after completing the matching of all medical data files in the first storage task;

[0134] An acquisition module 82 is used to acquire the space capacity occupied by the medical data files stored in the local database of the terminal;

[0135] The storage module 83 is used to execute the second storage task at the terminal when the space capacity occupied by the stored medical data files does not exceed the first preset capacity value; otherwise, the second storage task is executed at another terminal belonging to the same group as the terminal, so that when responding to the first request task of any terminal, the local databases of all terminals belonging to the same group as the terminal can be retrieved through the first type of information.

[0136] Specifically, the first type of information may include several categories, and the first quantity information may include the required quantity of files of each category; the display module 72 displays the retrieved medical data files to the user and the integration module 73 merges and organizes the medical data files with the server files for display to the user, which may include but is not limited to graphical presentation, classification presentation, etc.

[0137] In this embodiment, any group includes at least two terminals, and any terminal belongs to at most one group, that is, any terminal can belong to a specific group or not belong to any group.

[0138] In a specific embodiment, the basis for dividing the terminals into groups may include but is not limited to terminal functions, departments where the terminals are located, terminal authority levels, etc. For example, when the terminal functions are used as the basis for dividing the terminals into groups, multiple groups such as clinical terminals, management terminals, and logistics terminals can be divided. When the departments where the terminals are located are used as the basis for dividing the terminals into groups, multiple groups such as emergency department terminals, internal medicine terminals, surgical terminals, laboratory terminals, and pharmacy terminals can be divided.

[0139] Specifically, when the terminal executes the second storage task, the index information table is also updated based on the second storage task.

[0140] Furthermore, the device also includes:

[0141] A query module 91, used for obtaining the space capacity occupied by the medical data files stored in the local database of the terminal after the second storage task is executed;

[0142] A division module 92, configured to obtain identification information of all medical data files in the local database and divide all medical data files into a plurality of categories based on the identification information when the space capacity occupied by the stored medical data files exceeds a second preset capacity value;

[0143] The prompt module 93 is used to generate a similar file group based on at least two medical data files in the same category whose matching degree of identification information meets a first preset condition, compare the characteristic information of each medical data file in each similar file group, and display the similar file group of medical data files having at least one set of characteristic information whose matching degree meets a second preset condition to the user, so as to prompt the user to delete the medical data files accordingly;

[0144] A training module 94 is used to obtain the user's decision information on whether to delete the medical data files in each category of similar file groups, and use the decision information as a training sample for the deep learning model;

[0145] The correction module 95 is used to update the second preset condition using the trained deep learning model.

[0146] Specifically, the classification module 92 may use metadata or content analysis techniques of the files to determine the categories of medical data files, such as imaging data, text records, experimental results, etc.;

[0147] The training module 94 can combine the deep learning model to predict the value that the similarity of similar medical data files in different categories needs to reach, so as to automatically update the second preset condition based on the historical data analysis and similar features of the predicted files through the training model, and optimize the adjustment rules according to the model output;

[0148] In this embodiment, the second preset capacity value is not greater than the first preset capacity value.

[0149] It should be noted that the prompt module 93 displays to the user a similar file group of medical data files in which there is at least one set of feature information whose matching degree meets the second preset condition, so as to prompt the user to delete them. Specifically, each medical data file in which the repetition ratio of feature information in each file exceeds a preset ratio value and each medical data file in which the repetition ratio of feature information in each file exceeds a preset ratio value are displayed to the user, so as to prompt the user to delete at least one file in each group.

[0150] At the same time, by updating the second preset condition through the correction module 95, the second preset condition used for different categories of medical data files can be corrected in a targeted manner, and then the re-determined second preset condition is applied to the preset function performed by the matching module 20 and the prompt module 93.

[0151] Example 3

[0152] This embodiment proposes a computer-readable storage medium, which stores executable instructions for implementing the method proposed in Embodiment 1 when executed by a processor.

[0153] In summary, the present invention provides a storage and reading optimization method, device and storage medium for medical big data. The scheme proposed in the present invention can effectively avoid uploading low-value data by performing double matching of identification information and feature information of medical data files in storage tasks before uploading, thereby reducing the burden of network transmission and the processing load of the server. At the same time, combined with the deletion of redundant files on the server, it effectively prevents the waste of server storage space, and has the effect of improving data management efficiency and optimizing storage space utilization.

[0154] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing storage and reading of medical big data, characterized in that: The method comprises: In response to the first storage task of the terminal, sequentially acquiring identification information of the medical data files to be matched in the first storage task, so as to match the identification information with the local index information table of the terminal; If the index information whose matching degree with the identification information satisfies the first preset condition is matched, the characteristic information of the medical data file is obtained, and the characteristic information is matched with the specific content corresponding to the index information; if the index information whose matching degree with the identification information satisfies the first preset condition is not matched, or the matching degree between the characteristic information and the specific content corresponding to the index information does not satisfy the second preset condition, the matching of the medical data file is completed; If the matching degree between the characteristic information and the specific content corresponding to the index information meets a second preset condition, deleting the medical data file from the first storage task; After completing the matching of all the medical data files in the first storage task, executing the first storage task, and updating the index information table based on the first storage task; In response to the first storage tasks from the multiple terminals, the server receives the medical data files included in the first storage tasks, and deletes the redundant medical data files included in the first storage tasks from the multiple terminals.

2. The method according to claim 1, characterized in that The method further comprises: In response to a first request task of a terminal, parsing the first request task to determine first type information and first quantity information, and searching a local database of the terminal based on the first type information; If the number of medical data files retrieved from the local database meets the first number information, displaying the retrieved medical data files to the user; If the number of medical data files retrieved from the local database does not satisfy the first quantity information, a second request task for requesting server files from the database of the server is generated and executed based on the first type information, the first quantity information and the retrieved quantity information of the medical data files, so that the terminal obtains the server files from the server, and merges the medical data files with the server files to display them to the user.

3. The method according to claim 2, characterized in that Each terminal is pre-divided into a plurality of groups, each group includes at least two terminals, and each terminal belongs to at most one group; the method further includes: After completing the matching of all the medical data files in the first storage task, generating a second storage task for storing the medical data files included in the first storage task in a local database of the terminal; Obtaining the space capacity occupied by the medical data files stored in the local database of the terminal; If the space capacity occupied by the stored medical data files does not exceed the first preset capacity value, executing the second storage task on the terminal; If the space capacity occupied by the stored medical data files exceeds the first preset capacity value, the second storage task is executed at another terminal belonging to the same group as the terminal, so that when responding to the first request task of any terminal, the local databases of all terminals belonging to the same group as the terminal are retrieved through the first type of information.

4. The method according to claim 3, characterized in that The method further comprises: After executing the second storage task, obtaining the space capacity occupied by the medical data file stored in the local database of the terminal; If the space capacity occupied by the stored medical data files exceeds a second preset capacity value, obtaining identification information of all the medical data files in the local database, and dividing all the medical data files into a plurality of categories based on the identification information; A similar file group is generated based on at least two medical data files in the same category whose matching degree of the identification information meets the first preset condition, and the characteristic information of each medical data file in each similar file group is compared respectively, and the similar file group of medical data files in which there is at least one set of characteristic information whose matching degree meets the second preset condition is displayed to the user to prompt the user to make corresponding deletions.

5. The method according to claim 4, characterized in that The method further comprises: After prompting the user to perform the corresponding deletion, obtaining the user's decision information on whether to delete the medical data files in the similar file groups of each category, and using the decision information as a training sample for a deep learning model; The second preset condition is updated using the trained deep learning model, so that after a similar file group is generated based on at least two of the medical data files whose matching degree of the identification information in the same category meets the first preset condition, the step of "respectively comparing the characteristic information of each of the medical data files in each of the similar file groups" is performed based on the updated second preset condition.

6. The method according to claim 5, characterized in that The "if the index information whose matching degree with the identification information satisfies the first preset condition is matched, acquiring the characteristic information of the medical data file, and matching the characteristic information with the specific content corresponding to the index information" includes: If index information is matched whose matching degree with the identification information satisfies the first preset condition, the category to which the medical data file belongs is determined and characteristic information of the medical data file is obtained, so as to match the characteristic information with the specific content corresponding to the index information based on the updated second preset condition.

7. The method according to claim 4, characterized in that The method further comprises: If the space capacity occupied by the stored medical data files exceeds a third preset capacity value, the storage time of each medical data file in the local database is obtained, and the medical data files are deleted in sequence based on the storage time until the space capacity occupied by the stored medical data files does not exceed a fourth preset capacity value; the third preset capacity value is not less than the second preset capacity value and the fourth preset capacity value.

8. The method according to claim 1, characterized in that Each of the characteristic information corresponds to a different content part in the medical data file, and the "if the matching degree between the characteristic information and the specific content corresponding to the index information meets the second preset condition, deleting the medical data file from the first storage task" includes: If the proportion of the characteristic information whose matching degree with the specific content corresponding to the index information satisfies the third preset condition in all characteristic information is not less than the preset proportion value, it is determined that the matching degree between the characteristic information and the specific content corresponding to the index information satisfies the second preset condition, and the medical data file is deleted from the first storage task; An update file is generated based on the content portion corresponding to the feature information whose matching degree with the specific content corresponding to the index information does not satisfy the third preset condition, so as to add the update file to the first storage task.

9. A device for optimizing storage and reading of medical big data, characterized in that: The device comprises: A response module, configured to respond to the first storage task of the terminal, and sequentially obtain identification information of the medical data files to be matched in the first storage task, so as to match the identification information with the local index information table of the terminal; A matching module, configured to obtain characteristic information of the medical data file and match the characteristic information with specific content corresponding to the index information when matching index information with the identification information and satisfying a first preset condition; and to complete matching of the medical data file if matching index information with the identification information and satisfying the first preset condition is not matched, or matching between the characteristic information and specific content corresponding to the index information and does not satisfy a second preset condition; a deleting module, configured to delete the medical data file from the first storage task when the matching degree between the characteristic information and the specific content corresponding to the index information satisfies a second preset condition; An execution module, configured to execute the first storage task after completing the matching of all the medical data files in the first storage task, and update the index information table based on the first storage task; The processing module is used to respond to the first storage task from multiple terminals, enable the server to receive the medical data files included in the first storage task, and delete the redundant medical data files included in the first storage task from multiple terminals on the server.

10. A computer-readable storage medium, characterized in that: Executable instructions are stored, and when executed by a processor, the instructions are used to implement the method according to any one of claims 1 to 8.

Images