Health data analysis method and device based on artificial intelligence, equipment and medium

Abstract

The invention belongs to the technical field of artificial intelligence, and relates to a health data analysis method and device based on artificial intelligence, computer equipment and a storage medium. Cleaning the health data based on a cleaning strategy to obtain corresponding first health data; performing standardization processing on the first health data based on a conversion rule to obtain corresponding second health data; calling a health analysis model, and performing data analysis on the second health data based on the health analysis model to obtain corresponding physical examination quotation data; performing compliance verification on the physical examination quotation data based on the knowledge graph; and if the physical examination quotation data passes the compliance verification, returning the physical examination quotation data to the user. In addition, the physical examination quotation data can be stored in the block chain. The method can be applied to a physical examination quotation scene in the digital medical field, physical examination quotation processing of the user can be automatically and quickly completed, and the accuracy and reliability of the obtained physical examination quotation data are ensured.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology and can be applied to fields such as digital healthcare, particularly to artificial intelligence-based health data analysis methods, devices, computer equipment, and storage media. Background Art

[0002] In the healthcare industry, with the rapid development of technology, intelligent pricing systems, as an emerging concept, are gradually gaining significant attention, especially in the area of ​​health checkup benefits. This exploration of intelligent solutions aims to provide users with more personalized, accurate, and efficient pricing options for health checkup services through advanced information technology.

[0003] Although the application of intelligent pricing systems in medical examination benefits is still in its early stages, a number of forward-thinking and innovative enterprises and research institutions have emerged. They are committed to integrating cutting-edge technologies such as artificial intelligence, big data analysis, and natural language processing into the traditional medical examination pricing process in order to achieve intelligent, automated, and accurate pricing.

[0004] However, despite numerous attempts to apply intelligent technology to the pricing of medical examination benefits, current intelligent pricing systems on the market still face challenges in achieving both efficiency and reliability. Specifically, regarding efficiency, existing intelligent pricing systems often struggle to accurately identify users' personalized needs within a short timeframe, thus failing to generate pricing plans that meet user expectations. This not only impacts user experience but also limits the widespread application of intelligent pricing systems in the medical examination benefits sector. In terms of reliability, the accuracy and stability of intelligent pricing systems are fundamental to their widespread use. However, due to the complexity and diversity of medical data, as well as the complexity of pricing logic, existing systems often cannot guarantee the absolute accuracy and stability of pricing results. This, to some extent, affects users' trust and reliance on intelligent pricing systems.

[0005] In conclusion, while intelligent pricing systems have broad application prospects in the field of medical examination benefits, current systems on the market still have many shortcomings and challenges. Therefore, there is an urgent need for further in-depth research and development of more efficient and reliable intelligent pricing systems to meet the needs of the medical industry and a wide range of users. Summary of the Invention

[0006] The purpose of this application is to propose a health data analysis method, device, computer equipment, and storage medium based on artificial intelligence, so as to solve the technical problem that existing intelligent quotation systems cannot achieve high efficiency and reliability in quotation.

[0007] Firstly, an artificial intelligence-based health data analysis method is provided, including:

[0008] Obtain users' health data;

[0009] The health data is cleaned based on a preset cleaning strategy to obtain the corresponding first health data.

[0010] The first health data is standardized based on preset conversion rules to obtain the corresponding second health data;

[0011] A preset health analysis model is invoked, and the second health data is analyzed based on the health analysis model to obtain the corresponding physical examination quotation data; wherein, the health analysis model is a model obtained by training a preset machine learning model based on pre-constructed health data and adversarial sample data;

[0012] The medical examination quotation data is verified for compliance based on a pre-defined knowledge graph.

[0013] If the medical examination quotation data passes the compliance verification, the medical examination quotation data will be returned to the user.

[0014] Secondly, an artificial intelligence-based health data analysis device is provided, including:

[0015] The first acquisition module is used to acquire the user's health data;

[0016] The first processing module is used to clean the health data based on a preset cleaning strategy to obtain the corresponding first health data.

[0017] The second processing module is used to standardize the first health data based on a preset conversion rule to obtain the corresponding second health data.

[0018] The analysis module is used to call a preset health analysis model and perform data analysis on the second health data based on the health analysis model to obtain the corresponding physical examination quotation data; wherein, the health analysis model is a model obtained by training a preset machine learning model based on pre-constructed health data and adversarial sample data;

[0019] The verification module is used to perform compliance verification on the physical examination quotation data based on a preset knowledge graph;

[0020] The return module is used to return the medical examination quotation data to the user if the medical examination quotation data passes the compliance verification.

[0021] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described artificial intelligence-based health data analysis method.

[0022] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the aforementioned artificial intelligence-based health data analysis method.

[0023] In the aforementioned scheme implemented by the AI-based health data analysis method, device, computer equipment, and storage medium, the user's health data is acquired; then, the health data is cleaned based on a preset cleaning strategy to obtain corresponding first health data; and the first health data is standardized based on preset transformation rules to obtain corresponding second health data; subsequently, a preset health analysis model is invoked, and data analysis is performed on the second health data based on the health analysis model to obtain corresponding physical examination quotation data; wherein, the health analysis model is a model trained on a preset machine learning model based on pre-constructed health data and adversarial sample data; further, the physical examination quotation data is verified for compliance based on a preset knowledge graph; if the physical examination quotation data passes the compliance verification, the physical examination quotation data is returned to the user. This application uses a cleaning strategy to clean and process the acquired user health data to obtain first health data. Then, based on transformation rules, the first health data is standardized to obtain second health data. Subsequently, based on a health analysis model, the second health data is analyzed to obtain physical examination quotation data. Following this, a knowledge graph is used to perform compliance verification on the physical examination quotation data. After the compliance verification is passed, the physical examination quotation data is returned to the user. Thus, by processing user health data based on a health analysis model, this application can automatically and quickly complete the processing of physical examination quotations for users, improving the processing efficiency and ensuring the accuracy, reliability, and compliance of the obtained physical examination quotation data. Attached Figure Description

[0024] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is an exemplary system architecture diagram to which this application can be applied;

[0026] Figure 2 This is a flowchart of an embodiment of the AI-based health data analysis method according to this application;

[0027] Figure 3This is a schematic diagram of a structure of an embodiment of the AI-based health data analysis device according to this application;

[0028] Figure 4 This is a schematic diagram of the structure of one embodiment of the computer device according to this application. DETAILED DESCRIPTION

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0030] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0031] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0032] like Figure 1 As shown, system architecture 100 may include terminal device 101, network 102, and server 103. Terminal device 101 may be a laptop 1011, tablet 1012, or mobile phone 1013. Network 102 is used as a medium to provide a communication link between terminal device 101 and server 103. Network 102 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0033] Users can use terminal device 101 to interact with server 103 via network 102 to receive or send messages, etc. Various communication client applications can be installed on terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social media platform software, etc.

[0034] Terminal device 101 can be various electronic devices with a display screen and support web browsing. In addition to laptops 1011, tablets 1012, or mobile phones 1013, terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 player (Moving Picture Experts Group Audio Layer IV), a laptop computer, and a desktop computer, etc.

[0035] Server 103 can be a server that provides various services, such as a backend server that provides support for the pages displayed on terminal device 101.

[0036] It should be noted that the AI-based health data analysis method provided in this application is generally executed by a server / terminal device, and correspondingly, the AI-based health data analysis device is generally installed in the server / terminal device.

[0037] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0038] Continue to refer to Figure 2 This document illustrates a flowchart of an embodiment of the AI-based health data analysis method according to this application. The order of steps in the flowchart can be changed, and some steps can be omitted, depending on different needs. The AI-based health data analysis method provided in this application can be applied to any scenario requiring health data analysis, and thus can be applied to products in these scenarios, such as health data analysis in the digital healthcare field. The AI-based health data analysis method includes the following steps:

[0039] Step S201: Obtain the user's health data.

[0040] In this embodiment, the artificial intelligence-based health data analysis method runs on an electronic device (e.g., Figure 1The server / terminal device shown can acquire user health data via wired or wireless connection. It should be noted that the aforementioned wireless connection methods may include, but are not limited to, 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra-wide wireless band) connections, and other currently known or future known wireless connection methods. The executing entity of this application is a data analysis system, which may be simply referred to as the system. This application can be applied to business scenarios involving the pricing and processing of physical examination benefits in the digital healthcare field. The aforementioned health data may be the user's medical record data, which may include the user's age, gender, past medical history, etc.

[0041] The system supports simplified information collection processes, including: Intelligent form filling: Leveraging natural language processing (NLP) technology and machine learning algorithms, the system automatically infers and fills in the remaining necessary fields based on a small amount of information provided by the user. Example scenario: The user only needs to enter their name and ID number, and the system can automatically fill in most basic information, such as age and gender.

[0042] Voice / Image Recognition Interface: Integrates voice recognition API and OCR (Optical Character Recognition) technology, allowing users to quickly input relevant information via voice commands or uploaded images. Example Scenario: A user simply takes a photo of a lab report, and the system automatically extracts key information and fills it into the corresponding health checkup items.

[0043] Preset Template Selection: Provides preset physical examination package templates for common scenarios, allowing users to easily choose according to their own needs without having to configure them one by one. Example Scenario: Offers multiple packages such as "Routine Physical Examination" and "Women's Special Examination" for users to choose from; simply click to select the required items.

[0044] Progressive guidance: The system guides users through completing the information by asking progressively more relevant questions, displaying only the most relevant questions at a time to avoid information overload. Example scenario: When a user begins filling out a medical check-up appointment, the system first asks if the user has any past medical history, and then determines the next questions based on the answer.

[0045] Step S202: Clean the health data based on a preset cleaning strategy to obtain the corresponding first health data.

[0046] In this embodiment, the specific implementation process of cleaning the health data based on the preset cleaning strategy to obtain the corresponding first health data will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0047] Step S203: Standardize the first health data based on preset conversion rules to obtain the corresponding second health data.

[0048] In this embodiment, the specific implementation process of standardizing the first health data based on the preset conversion rules to obtain the corresponding second health data will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0049] Step S204: Invoke the preset health analysis model and perform data analysis on the second health data based on the health analysis model to obtain the corresponding physical examination quotation data; wherein, the health analysis model is a model obtained by training a preset machine learning model based on pre-constructed health data and adversarial sample data.

[0050] In this embodiment, the second health data can be input into the health analysis model to perform data analysis and obtain corresponding analysis results, namely a list of physical examination items. Then, based on the list of physical examination items and relevant cost standards (such as hospital pricing, regional differences, etc.), the estimated cost is estimated to obtain the physical examination quotation data. The calculation process of the physical examination quotation data can be automated using a cost estimation tool, providing a detailed cost breakdown. Furthermore, users can adjust the list of physical examination items and costs according to their actual needs and budget. The health analysis model is a model trained on a pre-built machine learning model using pre-constructed health data and adversarial sample data. The specific construction process of the health analysis model will be described in further detail in subsequent embodiments of this application and will not be elaborated upon here.

[0051] Step S205: Perform compliance verification on the physical examination quotation data based on a preset knowledge graph.

[0052] In this embodiment, the specific implementation process of verifying the compliance of the physical examination quotation data based on the preset knowledge graph will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0053] Step S206: If the medical examination quotation data passes the compliance verification, the medical examination quotation data is returned to the user.

[0054] In this embodiment, the specific implementation process of returning the physical examination quotation data to the user will be described in more detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0055] This application first acquires the user's health data; then, it cleans the health data based on a preset cleaning strategy to obtain corresponding first health data; and then, it standardizes the first health data based on preset transformation rules to obtain corresponding second health data; subsequently, it calls a preset health analysis model and performs data analysis on the second health data based on the health analysis model to obtain corresponding physical examination quotation data; wherein, the health analysis model is a model trained on a preset machine learning model based on pre-constructed health data and adversarial sample data; further, it performs compliance verification on the physical examination quotation data based on a preset knowledge graph; if the physical examination quotation data passes the compliance verification, the physical examination quotation data is returned to the user. This application uses a cleaning strategy to clean and process the acquired user health data to obtain first health data. Then, based on transformation rules, the first health data is standardized to obtain second health data. Subsequently, based on a health analysis model, the second health data is analyzed to obtain physical examination quotation data. Following this, a knowledge graph is used to perform compliance verification on the physical examination quotation data. After the compliance verification is passed, the physical examination quotation data is returned to the user. Thus, by processing user health data based on a health analysis model, this application can automatically and quickly complete the processing of physical examination quotations for users, improving the processing efficiency and ensuring the accuracy, reliability, and compliance of the obtained physical examination quotation data.

[0056] In some alternative implementations, prior to step S204, the electronic device may also perform the following steps:

[0057] Obtain pre-collected health data.

[0058] In this embodiment, a complete dataset containing user health status, physical examination items, costs, and user group characteristics (such as age, gender, region, etc.) is collected in advance. The dataset is then cleaned, standardized, and feature-engineered to ensure data quality and consistency, thereby obtaining the corresponding health data.

[0059] The machine learning model is trained based on the health data to obtain the corresponding base model.

[0060] In this embodiment, a suitable machine learning algorithm, such as decision tree, random forest, or neural network, can be selected as the aforementioned machine learning model based on the complexity of the problem and the characteristics of the data. Then, the aforementioned health data is used to train the machine learning model, enabling it to accurately predict the user's health status and recommend physical examination items, thus obtaining a well-trained base model.

[0061] Generate adversarial sample data corresponding to the base model.

[0062] In this embodiment, adversarial sample data can be generated based on the prediction results of the aforementioned basic model using adversarial attack methods (such as FGSM (Fast Gradient Sign Method) and PGD (Properted Gradient Descent)). These adversarial sample data significantly impact the model's prediction results while retaining most of the features of the original data. Fast Gradient Sign Method (FGSM) is an algorithm for generating adversarial examples. The core of this method is to maximize the loss function by subjecting the gradient of the input image to a limited perturbation, given the model parameters, thereby causing the model to make incorrect predictions. PGD refers to a method in neural network adversarial attacks that enhances the effectiveness of adversarial examples through multiple iterations. Specifically, it iterates multiple times based on FGSM (Fast Gradient Sign Method) to ensure that the strongest perturbation direction is found in the nonlinear model.

[0063] Construct a corresponding adversarial model based on the adversarial samples.

[0064] In this embodiment, an adversarial model can be trained using the aforementioned adversarial samples. This adversarial model aims to identify and correct biases in the base model. The adversarial model can be a classifier used to distinguish between adversarial and normal samples, or a regressor used to adjust the prediction results of the base model.

[0065] Based on a preset joint training strategy, the adversarial model is used to train the base model to obtain a trained first model.

[0066] In this embodiment, the aforementioned joint training strategy refers to introducing ideas from GANs (Generative Adversarial Networks) to design an adversarial training mechanism. This allows the model to self-correct biases during the learning process; that is, two models compete with each other, one attempting to generate a fair price, while the other tries to identify biases. Example scenario: After multiple rounds of adversarial training, the final model can provide a more balanced price for health checkup packages across different groups of people.

[0067] The joint training strategy described above may include: Alternating training: During training, the parameters of the base model and the adversarial model are updated alternately. First, the base model is trained using normal samples and adversarial samples. Then, the prediction results of the adversarial model are used to adjust the output of the base model. Next, the updated base model is used to generate new adversarial samples, and the adversarial model is trained to better recognize these samples. Loss function design: A suitable loss function is designed to simultaneously consider the model's accuracy and fairness. For example, a regularization term can be added to the loss function to penalize unfair predictions made by the model for a specific group. The first trained model can be obtained by training the base model using the adversarial model according to the training steps corresponding to the joint training strategy described above.

[0068] The first model is optimized based on a preset fairness assessment strategy to obtain an optimized second model.

[0069] In this embodiment, the fairness assessment strategy includes using statistical methods (such as difference measures, fairness indicators, etc.) to evaluate the fairness performance of the model. These assessment indicators can help identify biases in the model and quantify their degree. For example, statistical methods can be used to analyze the distribution of various population groups to identify potential bias factors (such as age, gender, region, etc.). The first model can be subjected to fairness assessment processing according to the assessment steps corresponding to the fairness assessment strategy to obtain the corresponding assessment results. Then, based on the assessment results, the parameters of the base model and the adversarial model can be adjusted to optimize the accuracy and fairness of the model. This may include multiple iterations and fine-tuning to obtain an optimized second model.

[0070] The second model is used as the health analysis model.

[0071] In this embodiment, user feedback regarding the fairness of pricing and the predictive accuracy of the health analysis model can be collected. This feedback data can be used to further adjust the model's parameters and training strategies to better meet user needs. Additionally, the model's accuracy and fairness performance are periodically evaluated to ensure the health analysis model can adapt to new data and user requirements. Based on the evaluation results and changes in user needs, the health analysis model is updated and optimized. This may include adding new features, adjusting the model structure, or introducing new training algorithms.

[0072] In addition, a complete clinical validation process has been established to ensure the effectiveness and accuracy of the intelligent pricing solution, including:

[0073] 1) Double-blind experimental design

[0074] Implementation Details: Conduct a large-scale double-blind controlled trial to compare the effectiveness of the intelligent pricing system with traditional manual pricing, ensuring the reliability and accuracy of the new system. Example Scenario: Randomly select a certain number of users to participate in the experiment; some will use the intelligent pricing system, while the other will receive manual pricing. Afterwards, compare the satisfaction and service quality of the two groups.

[0075] 2) Expert Review Committee

[0076] Implementation Details: Establish a review committee composed of medical experts to regularly review the output of the intelligent pricing system and provide improvement suggestions. Example Scenario: Hold an expert review meeting every six months to discuss the system's performance in the most recent period and formulate an optimization plan for the next phase.

[0077] 3) Long-term follow-up research

[0078] Implementation Details: Initiate a long-term user health status tracking project to collect data on the effects of actual physical examinations, thereby validating the effectiveness of the intelligent pricing system. Example Scenario: Track participating users for one year, recording health improvements after each physical examination to evaluate the contribution of the intelligent pricing system.

[0079] 4) Continuous improvement mechanism

[0080] Implementation Details: The intelligent pricing model is continuously iterated and upgraded based on the latest medical research findings and user feedback to maintain its advanced nature and adaptability. Example Scenario: As new diagnostic and treatment technologies develop, the system promptly updates relevant health checkup recommendations and pricing strategies.

[0081] This application acquires pre-collected health data; trains a machine learning model based on the health data to obtain a corresponding base model; then generates adversarial sample data corresponding to the base model; constructs a corresponding adversarial model based on the adversarial samples; subsequently, based on a preset joint training strategy, trains the base model using the adversarial model to obtain a trained first model; subsequently, optimizes the first model based on a preset fairness evaluation strategy to obtain an optimized second model; finally, uses the second model as the health analysis model. This application achieves efficient and accurate construction of a health analysis model by using an adversarial training mechanism to train the machine learning model based on acquired health data, effectively improving the fairness and robustness of the generated health analysis model.

[0082] In some optional implementations of this embodiment, step S202 includes the following steps:

[0083] Invoke the preset cleaning tool.

[0084] In this embodiment, the cleaning tool is an automated cleaning tool based on a rule engine and a machine learning model, which has the function of automatically detecting and correcting errors, missing values ​​and outliers.

[0085] The first health data is cleaned using the cleaning tool to obtain the corresponding first processed data.

[0086] In this embodiment, the cleaning tool described above can be used to identify and process different types of erroneous data in the first health data by invoking corresponding preset rules, thereby ensuring the accuracy and consistency of the obtained first processed data. Example scenario: Importing a batch of medical record data from different hospitals, the cleaning tool automatically identifies and fills in some missing test result fields.

[0087] The first processed data is integrated and processed based on a preset data model to obtain the corresponding second processed data.

[0088] In this embodiment, the aforementioned data model is a pre-constructed general data model covering common medical terms, diagnostic codes, and test results, such as ICD codes and LOIC standards. This data model can accommodate data from different sources and formats, and all external data must be converted to this format before entering the system. By using the data model to integrate and process the first processed data, the first processed data can be converted into a unified data model format, thereby obtaining the corresponding second processed data. Example scenario: A cooperating hospital uploads a batch of test reports using a local coding system. The system converts them into an international standard format according to a mapping table. The data model can be updated and expanded periodically to adapt to new medical data needs.

[0089] The second processed data is used as the first health data.

[0090] This application utilizes a pre-defined cleaning tool to clean the first health data, obtaining corresponding first processed data. Then, based on a pre-defined data model, it integrates the first processed data to obtain corresponding second processed data. This second processed data is then used as the first health data. By using a cleaning tool to clean the first health data and then integrating it using a data model, this application achieves efficient and accurate cleaning of health data, ensuring the accuracy and standardization of the generated first health data.

[0091] In some alternative implementations, step S203 includes the following steps:

[0092] Call the preset mapping table.

[0093] In this embodiment, the aforementioned mapping table is a pre-built and maintained dynamically updated mapping table used to convert non-standard format data into a data format that conforms to internal standards. The mapping table also supports the rapid addition of new mapping relationships, ensuring a smooth transition between old and new data formats.

[0094] Based on the mapping table, the first health data is converted into a data format to obtain third data corresponding to a preset standard format.

[0095] In this embodiment, a data conversion tool can be used to perform corresponding format conversion on the first health data based on the relevant data content in the mapping table, thereby obtaining third data that conforms to the internal standard data format. Example scenario: A new cooperating clinic was added, and the data format it provided was slightly different from the existing standard. Seamless integration was achieved by updating the rule base.

[0096] The quality of the third data is assessed.

[0097] In this embodiment, a preset data quality scoring tool is used to assess the quality of the third data, thereby quantifying its quality and obtaining corresponding quality assessment results. Implementation details include: assigning a score to each data point, with lower scores indicating higher data quality; and marking low-quality data for subsequent review or correction. The quality assessment results include passing or failing the assessment. Furthermore, the aforementioned data quality scoring tool is built upon pre-defined data quality assessment standards, such as completeness, accuracy, consistency, and timeliness.

[0098] If the third data passes the quality assessment, then the third data will be used as the second health data.

[0099] This application utilizes a preset mapping table; then, based on the mapping table, it converts the first health data into a data format to obtain third data corresponding to a preset standard format; subsequently, it performs a quality assessment on the third data; if the third data passes the quality assessment, it is used as the second health data. This application, by using a mapping table to convert the first health data into a data format corresponding to a preset standard format, intelligently performs a quality assessment on the third data, and when the third data passes the quality assessment, uses the generated third data as the required second health data, effectively ensuring the accuracy and standardization of the obtained second health data.

[0100] In some alternative implementations, step S205 includes the following steps:

[0101] Invoke the pre-built knowledge graph.

[0102] In this embodiment, the construction process of the aforementioned knowledge graph includes: pre-collecting textual data of medical laws and regulations from around the world; then using knowledge graph technologies (such as Neo4j, RDF, etc.) to construct a regulatory knowledge graph containing information such as regulatory clauses, scope of application, and update dates. A regulatory update mechanism can be established to ensure the real-time accuracy of the information in the knowledge graph.

[0103] Compliance rules are extracted from the knowledge graph.

[0104] In this embodiment, the aforementioned compliance rules refer to the rules contained in the knowledge graph used to detect whether the medical examination quote complies with regulatory requirements.

[0105] The medical examination quote data is matched based on the aforementioned compliance rules to obtain the corresponding matching results.

[0106] In this embodiment, a preset compliance check engine can be used to match the aforementioned medical examination quote data with extracted compliance rules to detect whether the medical examination quote data complies with local regulatory requirements and generate corresponding matching results. The matching results include whether the match was successful or failed.

[0107] If the matching result is successful, the medical examination quotation data is determined to have passed the compliance verification; otherwise, the medical examination quotation data is determined to have failed the compliance verification.

[0108] In this embodiment, if the matching result is successful, the medical examination quotation data is determined to have passed the compliance verification; if the matching result is unsuccessful, the medical examination quotation data is determined to have failed the compliance verification.

[0109] The system also includes a legal consultation access function, allowing users to connect with a professional legal advisory team for assistance with a single click when encountering complex or uncertain situations. Example scenario: If a user has doubts about the legality of a certain medical examination item, they can obtain professional answers through the built-in legal consultation service.

[0110] Additionally, the system can automatically generate a privacy statement document applicable to the current user based on the privacy protection regulations of different regions. Example scenario: When a user registers, the system automatically generates a privacy statement that complies with local regulations based on their location. The user reads and agrees to this statement before continuing.

[0111] This application utilizes a pre-built knowledge graph to extract compliance rules. Subsequently, it matches the medical examination price data against these compliance rules to obtain matching results. If the matching result is successful, the medical examination price data is deemed to have passed compliance verification; otherwise, it is deemed to have failed compliance verification. This application improves the efficiency of compliance verification by calling the pre-built knowledge graph, extracting compliance rules, and then matching the medical examination price data against these rules to obtain matching results. Based on these matching results, it automatically and accurately completes the compliance verification process for the medical examination price data and generates corresponding compliance verification results, ensuring the accuracy of the obtained compliance verification results.

[0112] In some optional implementations of this embodiment, step S206 includes the following steps:

[0113] Obtain explanatory information corresponding to the medical examination quotation data.

[0114] In this embodiment, the explanatory information refers to the main factors affecting the price and their weights that are matched with the above-mentioned physical examination price data, obtained from the health analysis model.

[0115] A corresponding medical examination quotation report is generated based on the explanatory information and the medical examination quotation data.

[0116] In this embodiment, a corresponding medical examination quotation report can be generated by filling the aforementioned explanatory information and medical examination quotation data into the corresponding positions within a preset quotation report template. The content of the quotation report template is not specifically limited and can be set according to actual business needs.

[0117] Obtain preset target feedback channels.

[0118] In this embodiment, the selection of the above-mentioned target feedback channels is not specifically limited and can be determined according to the actual push needs. For example, it may include user interface display, visualization tool display (such as charts, line graphs, etc.), email push, message push, etc.

[0119] Based on the target feedback channel, the medical examination quotation report is sent to the user.

[0120] In this embodiment, the feedback process of sending the medical examination quotation report to the user can be performed by using a selected target feedback channel.

[0121] This application obtains explanatory information corresponding to the medical examination quotation data; then generates a corresponding medical examination quotation report based on the explanatory information and the medical examination quotation data; subsequently, it obtains a preset target feedback channel; and then sends the medical examination quotation report to the user based on the target feedback channel. This application obtains explanatory information corresponding to the medical examination quotation data and automatically and intelligently generates a corresponding medical examination quotation report based on the explanatory information and the medical examination quotation data, and then sends the medical examination quotation report to the user based on the obtained target feedback channel. This allows the user to easily understand the composition and rationality of the medical examination quotation by reviewing the medical examination quotation report, thereby improving the user experience.

[0122] In some optional implementations of this embodiment, after step S203, the electronic device may further perform the following steps:

[0123] Obtain the preset end-to-end encryption policy.

[0124] In this embodiment, the selection of the above end-to-end encryption strategy is not specifically limited. For example, strong encryption algorithms such as AES-256 can be used.

[0125] The second health data is encrypted based on the end-to-end encryption strategy to obtain the corresponding third health data.

[0126] In this embodiment, the second health data can be encrypted using AES-256 or other strong encryption algorithms, employing a public / private key pair to encrypt and decrypt the data. This ensures that only authorized users can access the data and generates encrypted third health data. The security of the encryption algorithm can be checked periodically and updated as needed.

[0127] Call the preset data center.

[0128] In this embodiment, the aforementioned data center is a pre-built storage medium responsible for storing and processing the health data of the collected users. For example, it may be a local database, a local disk, a cloud server, a blockchain, etc.

[0129] The third health data is stored in the data center.

[0130] In this embodiment, the location information of the data center can be determined, and then the third health data can be stored in the data center according to the location information, thereby completing the standardized storage processing of the third health data.

[0131] The system implements strict access controls for sensitive data, requiring users to provide multiple verification methods (such as passwords, fingerprints, and facial recognition) to enhance account security. Specific implementation details include setting at least two verification methods for each user account, such as SMS verification code + password, or fingerprint + facial recognition. Multiple verification steps are required each time a user logs in or performs a sensitive operation. Example scenario: When a user attempts to view a detailed medical examination report or modify their personal information, the system requires them to confirm their identity via SMS verification code first.

[0132] Additionally, consider using blockchain or other forms of DLT to record and manage ownership and access rights to health data, ensuring data integrity and immutability. Example scenario: Data sharing between healthcare institutions is conducted through a blockchain platform, with each data exchange generating an immutable transaction record.

[0133] In addition, a dedicated security team is established to be responsible for the daily monitoring of the system and to conduct a comprehensive security audit every quarter, including code review and penetration testing. Example scenario: After discovering a potential vulnerability, the security team immediately fixes it and updates relevant protective measures to prevent similar problems from recurring.

[0134] This application obtains a preset end-to-end encryption strategy; then, based on the end-to-end encryption strategy, it encrypts the second health data to obtain the corresponding third health data; subsequently, it calls a preset data center; and finally, it stores the third health data in the data center. After standardizing the first health data based on conversion rules to obtain the second health data, this application automatically and intelligently encrypts the second health data based on the end-to-end encryption strategy to obtain the corresponding third health data, thereby effectively ensuring the security of health data during transmission. Furthermore, it stores the third health data in the data center, thus ensuring the data security and stability of the third health data.

[0135] In some alternative implementations, the user information obtained is subject to user consent and complies with relevant laws and policies.

[0136] Furthermore, any software tools or components not belonging to our company that appear in the embodiments of this application are merely illustrative examples and do not represent actual use.

[0137] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0138] It should be emphasized that, to further ensure the privacy and security of the aforementioned medical examination quotation data, the medical examination quotation data can also be stored in a blockchain node.

[0139] The blockchain referred to in this application is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.

[0140] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0141] Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0142] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware through computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the methods described above. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0143] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0144] Further reference Figure 3 As a response to the above Figure 2 The implementation of the method shown in this application provides an embodiment of a health data analysis device based on artificial intelligence. This device embodiment is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0145] like Figure 3 As shown, the AI-based health data analysis device 300 described in this embodiment includes: a first acquisition module 301, a first processing module 302, a second processing module 303, an analysis module 304, a verification module 305, and a return module 306. Wherein:

[0146] The first acquisition module 301 is used to acquire the user's health data;

[0147] The first processing module 302 is used to clean the health data based on a preset cleaning strategy to obtain the corresponding first health data.

[0148] The second processing module 303 is used to standardize the first health data based on a preset conversion rule to obtain the corresponding second health data.

[0149] Analysis module 304 is used to call a preset health analysis model and perform data analysis on the second health data based on the health analysis model to obtain corresponding physical examination quotation data; wherein, the health analysis model is a model obtained by training a preset machine learning model based on pre-constructed health data and adversarial sample data;

[0150] Verification module 305 is used to perform compliance verification on the physical examination quotation data based on a preset knowledge graph;

[0151] The return module 306 is used to return the medical examination quotation data to the user if the medical examination quotation data passes the compliance verification.

[0152] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based health data analysis method in the aforementioned embodiments, and will not be repeated here.

[0153] In some optional implementations of this embodiment, the AI-based health data analysis device further includes:

[0154] The second acquisition module is used to acquire pre-collected health data;

[0155] The first training module is used to train the machine learning model based on the health data to obtain the corresponding basic model;

[0156] A generation module is used to generate adversarial sample data corresponding to the base model;

[0157] A construction module is used to build a corresponding adversarial model based on the adversarial samples;

[0158] The second training module is used to train the base model using the adversarial model based on a preset joint training strategy, so as to obtain a trained first model.

[0159] An optimization module is used to optimize the first model based on a preset fairness assessment strategy to obtain an optimized second model.

[0160] A determination module is used to use the second model as the health analysis model.

[0161] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based health data analysis method in the aforementioned embodiments, and will not be repeated here.

[0162] In some optional implementations of this embodiment, the first processing module 302 includes:

[0163] The first submodule is used to invoke the preset cleaning tools;

[0164] The cleaning submodule is used to perform data cleaning processing on the first health data based on the cleaning tool to obtain the corresponding first processed data.

[0165] The integration submodule is used to perform data integration processing on the first processed data based on a preset data model to obtain the corresponding second processed data.

[0166] The first determining submodule is used to use the second processed data as the first health data.

[0167] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based health data analysis method in the aforementioned embodiments, and will not be repeated here.

[0168] In some optional implementations of this embodiment, the second processing module 303 includes:

[0169] The second calling submodule is used to call the preset mapping table;

[0170] The conversion submodule is used to convert the first health data based on the mapping table to obtain third data corresponding to a preset standard format.

[0171] An evaluation submodule is used to perform a quality assessment on the third data;

[0172] The second determining submodule is used to use the third data as the second health data if the third data passes the quality assessment.

[0173] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based health data analysis method in the aforementioned embodiments, and will not be repeated here.

[0174] In some optional implementations of this embodiment, the verification module 305 includes:

[0175] The third invocation submodule is used to invoke the pre-built knowledge graph;

[0176] An extraction submodule is used to extract compliance rules from the knowledge graph;

[0177] The matching submodule is used to match the medical examination quotation data based on the compliance rules to obtain the corresponding matching results;

[0178] The determination submodule is used to determine that the medical examination quotation data passes the compliance verification if the matching result is a successful match, and otherwise determine that the medical examination quotation data fails the compliance verification.

[0179] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based health data analysis method in the aforementioned embodiments, and will not be repeated here.

[0180] In some optional implementations of this embodiment, the return module 306 includes:

[0181] The first acquisition submodule is used to acquire explanatory information corresponding to the physical examination quotation data;

[0182] A generation submodule is used to generate a corresponding medical examination quotation report based on the explanation information and the medical examination quotation data;

[0183] The second acquisition submodule is used to acquire preset target feedback channels;

[0184] The sending submodule is used to send the physical examination quotation report to the user based on the target feedback channel.

[0185] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based health data analysis method in the aforementioned embodiments, and will not be repeated here.

[0186] In some optional implementations of this embodiment, the AI-based health data analysis device further includes:

[0187] The third acquisition module is used to acquire the preset end-to-end encryption strategy;

[0188] An encryption module is used to encrypt the second health data based on the end-to-end encryption strategy to obtain the corresponding third health data;

[0189] The calling module is used to call the preset data center;

[0190] A storage module is used to store the third health data in the data center.

[0191] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based health data analysis method in the aforementioned embodiments, and will not be repeated here.

[0192] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0193] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected via a system bus. It should be noted that only the computer device 4 with components 41-43 is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0194] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0195] The memory 41 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 4, such as the hard disk or memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 4. Of course, the memory 41 may also include both the internal storage unit and its external storage device of the computer device 4. In this embodiment, the memory 41 is typically used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions for health data analysis methods based on artificial intelligence. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or will be output.

[0196] In some embodiments, the processor 42 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to execute computer-readable instructions stored in the memory 41 or to process data, for example, to execute computer-readable instructions for the AI-based health data analysis method.

[0197] The network interface 43 may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 4 and other electronic devices.

[0198] Compared with the prior art, the embodiments of this application have the following main advantages:

[0199] In this embodiment, the application cleanses the acquired user's health data using a cleaning strategy to obtain first health data. Then, it standardizes the first health data based on transformation rules to obtain second health data. Subsequently, it analyzes the second health data using a health analysis model to obtain physical examination quotation data. Following this, it performs compliance verification on the physical examination quotation data using a knowledge graph. After the physical examination quotation data passes the compliance verification, it is returned to the user. Thus, by processing user health data using a health analysis model, this application can automatically and quickly complete the processing of physical examination quotations for users, improving the processing efficiency of physical examination quotation processing and ensuring the accuracy, reliability, and compliance of the obtained physical examination quotation data.

[0200] This application also provides another embodiment, namely, providing a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor to cause the at least one processor to perform the steps of the artificial intelligence-based health data analysis method described above.

[0201] Compared with the prior art, the embodiments of this application have the following main advantages:

[0202] In this embodiment, the application cleanses the acquired user's health data using a cleaning strategy to obtain first health data. Then, it standardizes the first health data based on transformation rules to obtain second health data. Subsequently, it analyzes the second health data using a health analysis model to obtain physical examination quotation data. Following this, it performs compliance verification on the physical examination quotation data using a knowledge graph. After the physical examination quotation data passes the compliance verification, it is returned to the user. Thus, by processing user health data using a health analysis model, this application can automatically and quickly complete the processing of physical examination quotations for users, improving the processing efficiency of physical examination quotation processing and ensuring the accuracy, reliability, and compliance of the obtained physical examination quotation data.

[0203] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0204] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A health data analysis method based on artificial intelligence, characterized in that, Includes the following steps: Obtain users' health data; The health data is cleaned based on a preset cleaning strategy to obtain the corresponding first health data. The first health data is standardized based on preset conversion rules to obtain the corresponding second health data; A preset health analysis model is invoked, and the second health data is analyzed based on the health analysis model to obtain the corresponding physical examination quotation data; wherein, the health analysis model is a model obtained by training a preset machine learning model based on pre-constructed health data and adversarial sample data; The medical examination quotation data is verified for compliance based on a pre-defined knowledge graph. If the medical examination quotation data passes the compliance verification, the medical examination quotation data will be returned to the user.

2. The health data analysis method based on artificial intelligence according to claim 1, characterized in that, Before the step of invoking the preset health analysis model, the following is also included: Acquire pre-collected health data; The machine learning model is trained based on the health data to obtain the corresponding base model; Generate adversarial sample data corresponding to the base model; Construct a corresponding adversarial model based on the aforementioned adversarial samples; Based on a preset joint training strategy, the adversarial model is used to train the base model to obtain a trained first model. The first model is optimized based on a preset fairness assessment strategy to obtain an optimized second model; The second model is used as the health analysis model.

3. The health data analysis method based on artificial intelligence according to claim 1, characterized in that, The step of cleaning the health data based on a preset cleaning strategy to obtain the corresponding first health data specifically includes: Invoke the preset cleaning tools; The first health data is cleaned using the cleaning tool to obtain the corresponding first processed data. Based on a preset data model, the first processed data is integrated and processed to obtain the corresponding second processed data; The second processed data is used as the first health data.

4. The health data analysis method based on artificial intelligence according to claim 1, characterized in that, The step of standardizing the first health data based on preset conversion rules to obtain the corresponding second health data specifically includes: Call the preset mapping table; Based on the mapping table, the first health data is converted into a data format to obtain third data corresponding to a preset standard format. The quality of the third data is assessed. If the third data passes the quality assessment, then the third data will be used as the second health data.

5. The health data analysis method based on artificial intelligence according to claim 1, characterized in that, The step of performing compliance verification on the medical examination quotation data based on a preset knowledge graph specifically includes: Invoke the pre-built knowledge graph; Compliance rules are extracted from the knowledge graph. The medical examination quote data is matched based on the compliance rules to obtain the corresponding matching results; If the matching result is successful, the medical examination quotation data is determined to have passed the compliance verification; otherwise, the medical examination quotation data is determined to have failed the compliance verification.

6. The health data analysis method based on artificial intelligence according to claim 1, characterized in that, The step of returning the medical examination quote data to the user specifically includes: Obtain explanatory information corresponding to the medical examination quotation data; A corresponding medical examination quotation report is generated based on the explanatory information and the medical examination quotation data; Obtain the preset target feedback channels; Based on the target feedback channel, the medical examination quotation report is sent to the user.

7. The health data analysis method based on artificial intelligence according to claim 1, characterized in that, After the step of standardizing the first health data based on a preset conversion rule to obtain the corresponding second health data, the method further includes: Obtain the preset end-to-end encryption policy; The second health data is encrypted based on the end-to-end encryption strategy to obtain the corresponding third health data. Call the preset data center; The third health data is stored in the data center.

8. A health data analysis device based on artificial intelligence, characterized in that, include: The first acquisition module is used to acquire the user's health data; The first processing module is used to clean the health data based on a preset cleaning strategy to obtain the corresponding first health data. The second processing module is used to standardize the first health data based on a preset conversion rule to obtain the corresponding second health data. The analysis module is used to call a preset health analysis model and perform data analysis on the second health data based on the health analysis model to obtain the corresponding physical examination quotation data; wherein, the health analysis model is a model obtained by training a preset machine learning model based on pre-constructed health data and adversarial sample data; The verification module is used to perform compliance verification on the physical examination quotation data based on a preset knowledge graph; The return module is used to return the medical examination quotation data to the user if the medical examination quotation data passes the compliance verification.

9. A computer device, characterized in that, The method includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the artificial intelligence-based health data analysis method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the artificial intelligence-based health data analysis method as described in any one of claims 1 to 7.

Images