Nursing data management optimization system for tumor patients

Abstract

The invention relates to the technical field of nursing data management, and discloses a nursing data management optimization system for tumor patients, and the system comprises a nursing data integration module which is used for collecting and integrating the nursing data of the tumor patients; the dynamic data standardization module is used for carrying out real-time standardization processing on the integrated multi-modal data; the secure storage module is used for storing the standardized nursing data in a cloud distributed database; the analysis and prediction module is used for carrying out analysis and trend prediction on the nursing data; the intelligent decision support module is used for providing intelligent decision support for nursing personnel according to analysis and prediction results; and the interactive user interface module is used for supporting input of nursing data, query of patient information and display of intelligent decision suggestions. The efficiency and quality of nursing data management are improved, a scientific basis is provided for nursing of tumor patients, and the nursing effect and the rehabilitation rate of the patients can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of nursing data management, and particularly to an optimized nursing data management system for cancer patients. Background Art

[0002] With the development of medical technology, the sources of nursing data have become increasingly diverse, including hospital information systems, wearable devices, nursing record sheets, and patients' self-reported voice recordings. These data include not only structured data but also a large amount of unstructured data, such as text, images, and voices. However, existing nursing data management systems have deficiencies in data integration, standardization processing, secure storage, analysis and prediction, and intelligent decision support. For example, the standardization processing of data often relies on fixed rules and is difficult to adapt to changes in data sources; the security and scalability of data storage are insufficient; the analysis and prediction functions are relatively single, lacking the ability to generate personalized nursing plans; and the real-time performance and adaptability of the intelligent decision support module also need to be improved.

[0003] In addition, there are also many problems in the user interaction of existing nursing data management systems. When nursing staff enter data, they usually need to input manually, which is inefficient and error-prone. When querying data, the system often lacks efficient search and filtering functions, making it difficult for nursing staff to quickly find the required information. When presenting analysis results, there is a lack of intuitive visualization tools and voice feedback functions, making it difficult for nursing staff to quickly understand and apply the analysis results. Therefore, the present invention proposes an optimized nursing data management system for cancer patients.

[0004] In summary, existing nursing data management systems have many deficiencies in data integration, standardization processing, secure storage, analysis and prediction, intelligent decision support, and user interaction, and cannot meet the needs of modern medical nursing. Therefore, the present invention proposes an optimized nursing data management system for cancer patients, aiming to solve the above problems and improve the efficiency and intelligence level of nursing data management. Summary of the Invention

[0005] The purpose of the present invention is to propose an optimized nursing data management system for cancer patients to solve the problems of low efficiency and insufficient intelligence level in existing nursing data management.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions: An optimized nursing data management system for cancer patients, comprising: A nursing data integration module, used to collect and integrate the nursing data of cancer patients, and by classifying and converting unstructured data into structured data, unify the data from different sources into a standard format to achieve the integration of multi-modal data; A dynamic data normalization module for real-time normalization of integrated multimodal data, generating normalization rules through a common data model CDM and machine learning algorithms, and capable of automatically adjusting the normalization rules according to user feedback and system operation data; A secure storage module for storing the normalized nursing data in a cloud distributed database, and realizing the storage, management and recovery of data through a distributed architecture, encryption technology, access control mechanism and backup technology; An analysis and prediction module for analyzing and predicting trends in nursing data, extracting potential patterns and correlations through deep learning algorithms and data analysis algorithms, evaluating the patient's emotional state, quality of life level and nursing operation effect, predicting the rehabilitation trend and generating a personalized nursing plan; An intelligent decision support module for providing intelligent decision support for nursing staff according to the analysis and prediction results, capable of dynamically adjusting decision suggestions according to the feedback of nursing staff, real-time monitoring the execution effect of the nursing plan, analyzing potential risk factors and sending out warning signals; An interactive user interface module for supporting nursing staff to input nursing data through voice recognition and gesture operations, query patient information and nursing data, and intuitively display the data analysis results and intelligent decision suggestions.

[0007] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention provides a nursing data integration module, which can effectively integrate multimodal data by classifying and converting unstructured data into structured data, and solves the problems of diverse data sources and complex data types in the existing system.

[0008] The present invention provides a dynamic data normalization module, which automatically generates normalization rules through a common data model CDM and machine learning algorithms, and dynamically adjusts the rules according to user feedback and system operation data, improving the flexibility and adaptability of data normalization processing.

[0009] The present invention provides a secure storage module, which ensures the security and scalability of data through a distributed architecture, encryption technology, access control mechanism and backup technology, and solves the problems of insufficient data storage security and scalability in the existing system.

[0010] The present invention provides an analysis and prediction module, which extracts potential patterns and correlations through deep learning algorithms and data analysis algorithms, can evaluate the patient's emotional state, quality of life level and nursing operation effect, and predict the rehabilitation trend to generate a personalized nursing plan, improving the scientific and personalized level of nursing decision-making.

[0011] The present invention provides an intelligent decision-making support module, which issues warning signals by dynamically adjusting decision-making suggestions, real-time monitoring the execution effect of the nursing plan, and analyzing potential risk factors, thereby improving the intelligence level and real-time performance of nursing decision-making.

[0012] The present invention provides an interactive user interface module, which realizes rapid data entry through voice recognition and gesture operations, supports fuzzy search, multi-condition filtering, and data visualization display, thereby improving the work efficiency of nursing staff and the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0014] Figure 1 It is a schematic diagram of the system composition provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0015] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following will, in conjunction with the accompanying drawings and preferred embodiments, detail the specific implementation manners, structures, features, and effects of a nursing data management optimization system for cancer patients proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0016] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0017] The following embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention.

[0018] The following will specifically describe the specific solution of a nursing data management optimization system for cancer patients provided by the present invention with reference to the accompanying drawings.

[0019] Embodiment

[0020] In an embodiment of the present invention, a nursing data management optimization system for cancer patients is provided, including: A nursing data integration module, which is used to collect and integrate the nursing data of cancer patients, and by classifying and converting unstructured data into structured data, unify the data from different sources into a standard format, and achieve the integration of multi-modal data; A dynamic data normalization module for real-time normalization of integrated multi-modal data, generating normalization rules through a common data model CDM and machine learning algorithms, and capable of automatically adjusting the normalization rules according to user feedback and system operation data; A secure storage module for storing the normalized nursing data in a cloud distributed database, achieving data storage, management, and recovery through a distributed architecture, encryption technology, access control mechanism, and backup technology; An analysis and prediction module for analyzing and predicting trends in nursing data, extracting potential patterns and correlations through deep learning algorithms and data analysis algorithms, evaluating the patient's emotional state, quality of life level, and nursing operation effect, predicting the rehabilitation trend, and generating personalized nursing plans; An intelligent decision support module for providing intelligent decision support for nursing staff based on the analysis and prediction results, capable of dynamically adjusting decision suggestions according to the feedback of nursing staff, real-time monitoring the execution effect of the nursing plan, analyzing potential risk factors, and sending warning signals; An interactive user interface module for supporting nursing staff to input nursing data through voice recognition and gesture operations, query patient information and nursing data, and intuitively display data analysis results and intelligent decision suggestions.

[0021] Please refer to Figure 1 which is a schematic diagram of the system composition provided by the embodiments of the present invention.

[0022] I. Nursing data integration module The nursing data integration module includes a nursing data acquisition unit, a data classification and transformation unit, and a nursing data fusion unit; The nursing data acquisition unit real-time collects nursing data from multiple data sources through a multi-source data access interface. The multi-source data access interface includes API interfaces, Wi-Fi communication protocols, file transfer protocols, and Web service interfaces. The data sources include hospital information systems, wearable devices, nursing record forms, and patient self-reported voice records. The nursing data includes patient basic information, nursing operation records, vital sign data, psychological assessment data, and patient self-reported data; The data classification and transformation unit classifies nursing data into structured data and unstructured data through a data type recognition algorithm, uses natural language processing technology to convert unstructured text data into structured data, and uses image recognition technology to convert unstructured image data into structured data. The unstructured data includes unstructured text data and unstructured image data; The nursing data fusion unit fuses the transformed unstructured data with the original structured data, and through data mapping and format conversion technology, unifies data from different sources into a standard format to achieve the fusion of multi-modal data.

[0023] It should be noted that the API interface is used to obtain structured data from the Hospital Information System (HIS), including medical record information, diagnosis results, and treatment records, and ensures the accurate transmission and reception of data through the standard HL7 protocol or the FHIR (Fast Healthcare Interoperability Resources) interface.

[0024] The Wi-Fi communication protocol is used to obtain vital sign data from wearable devices, including heart rate, blood pressure, body temperature, and blood oxygen saturation, and enables seamless connection and data transmission between the device and the system through the Wi-Fi wireless communication protocol.

[0025] The File Transfer Protocol (FTP) is used to receive data in the form of nursing record sheets and inspection reports.

[0026] The Web service interface is used to receive patients' self-reported voice records and psychological assessment data submitted through the web page.

[0027] The Hospital Information System (HIS) provides structured data on patients' medical record information, diagnosis results, and treatment records.

[0028] Wearable devices provide patients' vital sign data, including heart rate, blood pressure, body temperature, and blood oxygen saturation.

[0029] The nursing record sheet provides nursing operation records, including nursing time, nursing content, and nursing effect.

[0030] Patients' self-reported voice records provide patients' psychological states and self-feelings through voice records or text records.

[0031] Patients' basic information includes name, gender, age, and medical history.

[0032] Nursing operation records include nursing time, nursing content, and nursing staff.

[0033] Vital sign data includes heart rate, blood pressure, body temperature, and blood oxygen saturation.

[0034] Psychological assessment data includes the degree of anxiety and the degree of depression.

[0035] Patients' self-reported data includes patients' pain feelings and discomfort descriptions.

[0036] Natural language processing technology is a technology that converts human language (natural language) into a format that can be understood by computers, with the goal of enabling computers to understand, generate, and process human language, thereby achieving natural interaction between humans and machines.

[0037] Unstructured text data: including the text descriptions in the nursing record sheets and the text transcriptions of the patient's self-reported voice recordings. This unit uses natural language processing (NLP) technology to parse the unstructured text data, extract key information, and convert it into structured data. Key information of nursing operations (operation time, operation content) is extracted through text analysis and stored as structured fields.

[0038] Image recognition technology is a technology that analyzes and processes image data through computer algorithms to identify objects, scenes, or features in the image. The goal is to enable the computer to automatically recognize and understand the image content, thereby realizing the functions of image classification, detection, and segmentation.

[0039] Unstructured imaging data: including images of the patient's wounds and imaging in the examination reports. Using image recognition technology, the unstructured imaging data is analyzed, key features are extracted, and converted into structured data. Information such as the size and shape of the wound is extracted through image recognition and stored as structured fields.

[0040] Data mapping: Data mapping is a process of corresponding and converting data fields from different sources to ensure data consistency and comparability. The basic patient information fields in the hospital information system are mapped to the patient identification fields in the wearable device.

[0041] Format conversion: Convert data in different formats into a unified standard format for subsequent processing and analysis. Convert text data into JSON format and convert imaging data into a unified image format.

[0042] Data fusion: Integrate structured data and the converted unstructured data to form a complete nursing data set. Integrate the patient's medical record information, vital sign data, and nursing operation records to form a complete patient nursing data record.

[0043] II. Dynamic Data Standardization Module The dynamic data standardization module includes a standardization rule generation unit and a rule update unit; The standardization rule generation unit is used to generate standardization rules to standardize the data. Define the structure and content of the data according to the common data model CDM, and automatically generate standardization rules using machine learning algorithms. The standardization rules include data format specifications, field mapping rules, data type conversion rules, data integrity constraints, and data consistency constraints; A rule update unit for optimizing the rule generation process, dynamically updating the standardization rules, collecting user feedback and system operation data through a feedback loop mechanism, optimizing the standardization rule generation process, and dynamically updating the standardization rules using an incremental learning algorithm. The feedback loop mechanism includes user feedback collection, system operation data monitoring, and rule optimization feedback. The incremental learning algorithm includes online learning and incremental learning.

[0044] It should be noted that the specific steps for automatically generating standardization rules according to the common data model CDM to define the structure and content of the data and using machine learning algorithms are as follows: Step 1: Define the common data model (CDM) Determine data entities: Identify the key entities in the nursing data, including patients, nursing operations, and vital signs.

[0045] Define the data structure: Define the data structure for each entity, including field names, data types, and field lengths.

[0046] Establish a relationship model: Define the relationships between entities, including the association between patients and nursing operations, and the association between nursing operations and vital signs.

[0047] Formulate data specifications: Formulate data format specifications, integrity constraints, and consistency constraints for each field.

[0048] Step 2: Data collection and preprocessing Collect data: Collect nursing data from multiple data sources (hospital information systems, wearable devices, nursing records).

[0049] Data cleaning: Remove duplicate data, correct incorrect data, and fill in missing values.

[0050] Data annotation: Annotate the collected data, mark the source, type, and content of the data, and provide training data for machine learning algorithms.

[0051] Step 3: Automatically generate standardization rules using machine learning algorithms Feature extraction: Extract features from the preprocessed data, including the distribution of field values and the associations between fields.

[0052] Model training: Use machine learning algorithms (decision trees, random forests, neural networks) to train the annotated data and generate standardization rules.

[0053] Rule generation: Generate standardization rules based on the training results, including data format specifications, field mapping rules, data type conversion rules, data integrity constraints, and data consistency constraints.

[0054] The Common Data Model (CDM) is a standardized data model used to define the structure and content of data, including data entities: patients, care operations, vital signs; data structures: field names, data types, and field lengths for each entity; relationship models: relationships between entities; data specifications: format specifications, integrity constraints, and consistency constraints for fields.

[0055] Machine learning algorithms are technologies that train models through data to achieve automatic learning and prediction.

[0056] Data format specifications refer to the unified regulations on the format of data to ensure data consistency and readability, including date format: unified as "YYYY-MM-DD"; numerical format: unified with a fixed number of decimal places (retaining two decimal places); text format: unified with a specific encoding format (UTF-8).

[0057] Field mapping rules refer to mapping fields from different data sources to the corresponding fields in the Common Data Model (CDM), including: mapping the "Patient ID" field in the hospital information system to the "Patient Identification" field in the CDM; mapping the "User ID" field in the wearable device to the "Patient Identification" field in the CDM; mapping the "Care Time" field in the care record form to the "Operation Time" field in the CDM.

[0058] Data type conversion rules refer to converting data from one type to another to meet the requirements of the common data model, including string to numeric: converting a string-type number ("123") to an integer type; date format conversion: converting dates in different formats ("MM / DD / YYYY") to a unified format ("YYYY-MM-DD"); text encoding conversion: converting text in different encoding formats (GBK) to a unified encoding format (UTF-8).

[0059] Data integrity constraints refer to constraints on the integrity and validity of data to ensure data accuracy and reliability, including non-null constraints: certain fields (patient name, care time) cannot be null; range constraints: the values of certain fields must be within a specific range (body temperature between 35°C and 42°C); uniqueness constraints: the values of certain fields must be unique (patient ID).

[0060] Data consistency constraints refer to constraints on data consistency to ensure data consistency across different sources, including field value consistency: patient names in different data sources must be consistent; associated data consistency: the vital sign data of a patient must be consistent with the patient ID in the care operation record.

[0061] Collect user feedback and system operation data through a feedback loop mechanism to optimize the generation process of standardization rules. The specific steps for dynamically updating standardization rules using incremental learning algorithms are as follows: Step 1: User feedback collection User interface design: Design a feedback function in the interactive user interface module to allow caregivers to evaluate and provide feedback on the data standardization results.

[0062] Feedback collection: Collect feedback information from caregivers through the user interface, mark whether the data meets expectations, and propose suggestions for rule modification.

[0063] Step 2: System operation data monitoring Data monitoring: Real-time monitor the data generated during the system operation, including the accuracy of data conversion and the application effect of rules.

[0064] Anomaly detection: Detect whether errors occur during the data conversion process or whether the rules can correctly handle new data formats.

[0065] Step 3: Rule optimization feedback Feedback analysis: Analyze user feedback and system operation data to generate feedback information for rule optimization.

[0066] Optimization suggestions: Propose rule optimization suggestions based on the feedback information, including adjusting field mapping rules and modifying data format specifications.

[0067] Step 4: Rule update Incremental learning: Use incremental learning algorithms (online learning and incremental learning) to dynamically update standardization rules.

[0068] Rule adjustment: Adjust the standardization rules according to the optimization suggestions to ensure that the rules are always in the optimal state.

[0069] User feedback collection: Collect the evaluation and suggestions of caregivers on the data standardization results through the user interface to understand user needs and system deficiencies.

[0070] System operation data monitoring: Real-time monitor the data generated during the system operation, detect the accuracy of data conversion, the application effect of rules, and abnormal situations.

[0071] Rule optimization feedback: Generate optimization suggestions based on user feedback and system operation data, adjust the standardization rules to ensure that the rules are always in the optimal state.

[0072] Online learning: The model is updated immediately every time it receives new data, can quickly adapt to data changes, and is suitable for real-time data processing scenarios.

[0073] Incremental learning: The model is gradually updated after receiving a batch of new data. It can gradually adapt to new data patterns without retraining the entire model. It is suitable for scenarios where the amount of data is large and arrives gradually.

[0074] 3. Secure Storage Module The secure storage module includes a distributed storage unit, a data encryption unit, an access control unit and a data backup unit; Distributed storage unit, used to store standardized nursing data in a distributed manner, and to achieve data scalability through a distributed architecture, which includes a distributed file system, a distributed database, data sharding, a replication mechanism, and load balancing; The data encryption unit is used to encrypt the stored data, use the SHA-256 hash algorithm to hash the data, use asymmetric encryption technology to encrypt the data, and combine the blockchain technology to store the data in a chain to prevent the data from being tampered with; Access control unit, which is used to manage user access rights to data. It adopts a role-based access control mechanism to assign different access rights according to user roles. It adopts a multi-factor authentication mechanism to ensure that only authorized users can access and operate data through identity authentication and authorization mechanisms. The data backup unit is used to regularly back up the stored data, use regular snapshot technology to perform snapshot backup of the data, use off-site backup technology to store the backup data on servers in different geographical locations, and use multi-copy backup technology to achieve redundancy and recoverability of the backup data.

[0075] It should be noted that the distributed file system includes the Hadoop Distributed File System (HDFS), which is used to store large-scale file data and supports high-throughput data access.

[0076] Distributed databases: including Cassandra and MongoDB, which are used to store structured and semi-structured data and support high-concurrency data read and write operations.

[0077] Data sharding: Split data into multiple fragments and store them on different storage nodes to improve data reading and writing efficiency.

[0078] Replication mechanism: Multiple copies are created for each data fragment and stored in different physical locations to prevent data loss.

[0079] Load balancing: Through load balancing technology, data access requests are reasonably distributed to different storage nodes to ensure high availability and performance of the system.

[0080] SHA-256 hash algorithm: Hash the data to generate a unique hash value for data integrity verification.

[0081] Asymmetric encryption technology: including the RSA algorithm, which encrypts data to ensure the security of data during storage and transmission.

[0082] Blockchain technology: combined with the chain storage structure of the blockchain, data blocks are linked in an immutable way to further enhance the security and credibility of data.

[0083] Role-based access control mechanism: assigns different access permissions according to the roles of users (doctors, nurses, administrators) to ensure that users can only access data within their authorized scope.

[0084] Multi-factor authentication mechanism: combines multiple authentication methods (password, fingerprint, SMS verification code) to enhance the security of user identity verification.

[0085] Identity authentication and authorization mechanism: confirms the user's identity through authentication and checks whether the user has the permission to access specific data through the authorization mechanism.

[0086] Regular snapshot technology: regularly takes snapshot backups of the stored data to generate snapshot copies of the data for quick recovery in case of data loss or corruption.

[0087] Off-site backup technology: stores the backup data on servers in different geographical locations to prevent data loss caused by natural disasters, fires and other unexpected events.

[0088] Multi-copy backup technology: creates multiple copies of the backup data and stores them on different storage media or locations to further improve the redundancy and recoverability of the data.

[0089] IV. Analysis and Prediction Module The analysis and prediction module includes a data analysis unit, a trend prediction unit and a solution generation unit; The data analysis unit uses deep learning algorithms and data analysis algorithms to analyze nursing data. The deep learning algorithms include convolutional neural networks and long short-term memory networks. The convolutional neural network is used to analyze image data, and the long short-term memory network is used to analyze time series data. The data analysis algorithms include data mining techniques for extracting potential patterns and correlation relationships from nursing data to evaluate the patient's emotional state, quality of life level and nursing operation effect; The trend prediction unit is used to predict the patient's recovery trend based on historical data and analysis results. It adopts the ARIMA model and deep learning prediction model, combines the patient's individual characteristics and historical recovery data to generate a personalized recovery path and provides a visual display of the recovery trend; A solution generation unit, which is used to generate personalized nursing plan suggestions based on the analysis and prediction results. By adopting a rule-based inference engine and a machine learning recommendation system, it comprehensively considers the patient's emotions, quality of life, nursing operation effects, and rehabilitation trends to generate a nursing plan.

[0090] It should be noted that the Convolutional Neural Network (CNN): is used to analyze image data and extract the features of the lesion area from medical images. The steps for the Convolutional Neural Network to analyze image data include: 1. Data preprocessing: Image normalization: Normalize the image data to the same size and pixel value range (0 to 1); Data augmentation: Increase the diversity of training data through operations such as rotation, flipping, and cropping.

[0091] 2. Convolutional layer: Convolution operation: Slide the convolutional kernel on the image to extract local features; Activation function: Usually use the ReLU (Rectified Linear Unit) activation function to increase non-linearity.

[0092] 3. Pooling layer: Downsampling: Reduce the size of the feature map through max pooling or average pooling to retain important features.

[0093] 4. Fully connected layer: Feature flattening: Flatten the multi-dimensional feature map into a one-dimensional vector; Classification or regression: Use the fully connected layer for classification or regression tasks and output the final result.

[0094] 5. Model training: Loss function: Use the cross-entropy loss function (for classification tasks) or the mean squared error loss function (for regression tasks); Optimization algorithm: Use the gradient descent algorithm (Adam) to optimize the model parameters.

[0095] 6. Model evaluation: Validation set evaluation: Use the validation set to evaluate the model performance and adjust the hyperparameters; Test set evaluation: Use the test set to evaluate the final model performance.

[0096] Long Short-Term Memory Network (LSTM): is used to analyze time series data and extract time series features from the patient's vital sign data. The steps for the Long Short-Term Memory Network (LSTM) to analyze time series data include: 1. Data preprocessing: Time series normalization: Normalize the time series data to the same range (0 to 1); Sequence segmentation: Segment the time series into subsequences of a fixed length.

[0097] 2. LSTM layer: Input gate: Controls the degree to which the current input information enters the cell state; Forget gate: Controls the degree to which information in the cell state is forgotten; Output gate: Controls the degree to which information in the cell state is output.

[0098] 3. Fully Connected Layer: Feature Extraction: Extract the output features of the LSTM layer into a one-dimensional vector; Classification or Regression: Use the fully connected layer for classification or regression tasks and output the final result.

[0099] 4. Model Training: Loss Function: Use the cross-entropy loss function (for classification tasks) or the mean squared error loss function (for regression tasks); Optimization Algorithm: Use the gradient descent algorithm (Adam) to optimize the model parameters.

[0100] 5. Model Evaluation: Validation Set Evaluation: Use the validation set to evaluate the model performance and adjust the hyperparameters; Test Set Evaluation: Use the test set to evaluate the final model performance.

[0101] Data Mining Techniques: Used to extract potential patterns and association relationships from nursing data, identify patterns of patients' emotional states through clustering analysis, discover associations between nursing operations and patients' recovery effects through association rule mining. Specific steps of data mining techniques: 1. Data Preprocessing: Data Cleaning: Remove duplicate data, correct incorrect data, and fill in missing values; Data Normalization: Normalize the data to the same range (0 to 1).

[0102] 2. Feature Extraction: Feature Selection: Select features related to the target variable; Feature Generation: Generate new features through combination or transformation.

[0103] 3. Pattern Discovery: Clustering Analysis: Use the K-Means algorithm to divide the data into different clusters and discover the internal structure of the data; Association Rule Mining: Use the Apriori algorithm to discover association rules in the data.

[0104] 4. Association Relationship Analysis: Correlation Analysis: Calculate the correlation coefficients between features to discover potential association relationships; Causal Relationship Analysis: Use causal inference methods to analyze the causal relationships between features.

[0105] 5. Evaluation Metrics: Emotional State Evaluation: Evaluate the patient's emotional state by analyzing the patient's self-reported voice records and psychological assessment data; Quality of Life Evaluation: Evaluate the patient's quality of life level by analyzing the patient's living habits and activity ability data; Nursing Operation Effect Evaluation: Evaluate the effect of nursing operations by analyzing nursing operation records and vital sign data.

[0106] ARIMA Model: Used to model and predict time series data and predict the future changes in patients' vital signs.

[0107] Deep Learning Prediction Model: Transformer Model, used to process complex sequence data and improve the prediction accuracy.

[0108] Combined with patient individual characteristics: Consider the patient's age, gender, medical history, and individual characteristics of the treatment plan.

[0109] Combined with historical rehabilitation data: Utilize the patient's historical rehabilitation data, including past nursing operation records and vital sign changes, to generate a personalized rehabilitation path.

[0110] Rehabilitation trend chart: Display the patient's rehabilitation trend through visual charts, including line charts showing the change trend of vital signs and bar charts showing the effect of nursing operations.

[0111] The specific steps for generating a personalized rehabilitation path and providing a visual display of the rehabilitation trend by using the ARIMA model and deep learning prediction model, combined with patient individual characteristics and historical rehabilitation data, are as follows: 1. Data preprocessing: Time series normalization: Normalize the time series data to the same range (0 to 1); Sequence segmentation: Segment the time series into subsequences of a fixed length.

[0112] 2. ARIMA model: Model establishment: Select appropriate ARIMA model parameters (p, d, q); Model fitting: Fit the ARIMA model using historical data; Model prediction: Use the fitted model to predict future data.

[0113] 3. Deep learning prediction model: Model selection: Select an appropriate deep learning model (Transformer); Model training: Train the deep learning model using historical data; Model prediction: Use the trained model to predict future data.

[0114] 4. Generation of personalized rehabilitation path: Combined with individual characteristics: Consider the patient's age, gender, medical history, and individual characteristics of the treatment plan; Combined with historical rehabilitation data: Utilize the patient's historical rehabilitation data, including past nursing operation records and vital sign changes; Path generation: Generate a personalized rehabilitation path based on the prediction results.

[0115] 5. Visual display: Rehabilitation trend chart: Display the patient's rehabilitation trend through visual charts, including line charts showing the change trend of vital signs and bar charts showing the effect of nursing operations.

[0116] The specific steps for generating a nursing plan by using a rule-based inference engine and a machine learning recommendation system, comprehensively considering the patient's emotions, quality of life, nursing operation effect, and rehabilitation trend, are as follows: 1. Rule definition: Medical knowledge rules: Define a series of rules based on medical knowledge and nursing experience; Nursing experience rules: Define a series of rules based on the experience of nursing staff.

[0117] 2. Rule-based reasoning: Rule matching: Match corresponding rules according to the analysis results and prediction results; Reasoning generation: Generate nursing suggestions according to the matched rules.

[0118] 3. Machine learning recommendation system: Model training: Use historical nursing data to train a machine learning model to learn the relationship between nursing plans and patients' rehabilitation effects; Plan recommendation: According to the analysis and prediction results, recommend personalized nursing plans, including recommending specific nursing operations and drug treatment plans.

[0119] 4. Factors to be considered comprehensively: Emotional state: Adjust the psychological counseling content in the nursing plan according to the patient's emotional state; Quality of life level: Adjust the life guidance content in the nursing plan according to the patient's quality of life level; Nursing operation effect: Adjust the operation content and frequency in the nursing plan according to the effect of nursing operations; Rehabilitation trend: Adjust the rehabilitation plan in the nursing plan according to the patient's rehabilitation trend.

[0120] 5. Nursing plan generation: Plan integration: Integrate the rule-based reasoning results and machine learning recommendation results into a complete nursing plan; Plan optimization: Optimize the nursing plan according to the comprehensively considered factors to ensure its scientificity and effectiveness.

[0121] V. Intelligent decision-making support module The intelligent decision-making support module includes a decision-making advice unit, a feedback adjustment unit, and a risk warning unit; The decision-making advice unit is used to provide intelligent decision-making support for nursing staff according to the analysis and prediction results, and uses a rule-based decision tree and a machine learning classifier to generate psychological counseling decisions and nursing plan optimization suggestions; The feedback adjustment unit is used to automatically adjust the decision-making advice according to the feedback of the nursing staff, adopts a reinforcement learning algorithm and a feedback loop mechanism to optimize the decision-making process, can monitor the execution effect of the nursing plan in real time, dynamically adjust the advice content, and adapt to the patient's rehabilitation situation; The risk warning unit is used to send a warning signal to the nursing staff according to the patient's rehabilitation risk assessment results, adopts a real-time data stream processing framework and an anomaly detection algorithm for real-time monitoring and warning, and can analyze potential risk factors to warn of possible rehabilitation risks in advance.

[0122] It should be noted that the specific steps for generating psychological counseling decisions and nursing plan optimization suggestions using a rule-based decision tree include 1. Rule definition: Medical knowledge rules: Define a series of rules according to medical knowledge and nursing experience, such as "If the patient is in a low mood and the duration exceeds one week, it is recommended to increase the frequency of psychological counseling"; Nursing experience rules: Define a series of rules according to the experience of nursing staff, such as "If the patient's vital signs are abnormal, it is recommended to increase the monitoring frequency".

[0123] 2. Decision tree construction: Feature selection: Select features related to the target variable, including emotional state, quality of life level, and nursing operation effect; Tree generation: Use decision tree algorithms (ID3, C4.5, CART) to construct a decision tree model and generate decision paths based on the input patient data; Tree pruning: Reduce overfitting through pruning and improve the generalization ability of the model.

[0124] 3. Decision generation: Input data: Input the real-time data of the patient into the decision tree model; Path generation: Generate specific decision paths according to the decision tree model; Suggestion generation: Generate psychological counseling decisions and suggestions for optimizing the nursing plan according to the decision paths. For example, it is recommended to increase the frequency of psychological counseling.

[0125] The specific steps for generating psychological counseling decisions and suggestions for optimizing the nursing plan using a machine learning classifier include: 1. Data preprocessing: Data cleaning: Remove duplicate data, correct incorrect data, and fill in missing values; Data normalization: Normalize the data to the same range (0 to 1).

[0126] 2. Feature extraction: Feature selection: Select features related to the target variable, including emotional state, quality of life level, and nursing operation effect; Feature generation: Generate new features through combination or transformation.

[0127] 3. Model selection: Select classifier: Select a suitable machine learning classifier, including logistic regression, support vector machine (SVM), and random forest.

[0128] 4. Model training: Training data: Use historical nursing data to train the classifier and learn the relationship between the nursing plan and the patient's recovery effect; Model optimization: Optimize the model parameters through cross-validation methods to improve the accuracy and generalization ability of the model.

[0129] 5. Suggestion generation: Input data: Input the real-time data of the patient into the trained classifier; Prediction result: Generate psychological counseling decisions and suggestions for optimizing the nursing plan according to the output of the classifier.

[0130] The specific steps for optimizing the decision-making process using a reinforcement learning algorithm include: 1. Environment definition: State space: Define the current state of the patient, including emotional state, quality of life level, and nursing operation effect; Action space: Define possible nursing operations, including increasing the frequency of psychological counseling and adjusting the specific content of nursing operations.

[0131] 2. Model selection: Select algorithm: Select a suitable reinforcement learning algorithm, including Q-learning and Deep Q-Network (DQN).

[0132] 3. Model Training: Initialization: Initialize the Q-value or policy network; Exploration and Exploitation: Select actions and observe the environmental feedback through exploration and exploitation strategies; Policy Update: Update the Q-value or policy network based on the feedback to learn the optimal decision-making strategy.

[0133] 4. Policy Generation: Input Data: Input the real-time status of the patient into the reinforcement learning model; Action Selection: Select the optimal nursing operation according to the output of the model; Recommendation Generation: Generate specific nursing recommendations, including adjusting the frequency or content of nursing operations.

[0134] The specific steps for optimizing the decision-making process using a feedback loop mechanism include: 1. User Feedback Collection; Feedback Interface: Collect the feedback of nursing staff on decision-making recommendations through the user interface, such as marking whether the recommendations are effective and putting forward improvement suggestions.

[0135] 2. System Operation Data Monitoring: Real-time Monitoring: Real-time monitor the implementation effect of the nursing plan, detect the actual effect of nursing operations and the patient's recovery situation.

[0136] 3. Rule Optimization Feedback: Feedback Analysis: Analyze user feedback and system operation data to generate feedback information for rule optimization; Rule Update: Adjust the decision-making rules according to the feedback information to optimize the decision-making process.

[0137] 4. Dynamic Adjustment: Real-time Adjustment: Dynamically adjust decision-making recommendations based on monitoring results and user feedback to ensure the scientificity and adaptability of the recommendations.

[0138] The specific steps for real-time monitoring and early warning using a real-time data stream processing framework include: 1. Data Stream Definition: Data Source: Define the data source, including the patient's vital sign monitoring devices and nursing operation recording systems; Data Stream: Define the format and transmission method of the data stream, and use Apache Kafka or Apache Flink to process real-time data streams.

[0139] 2. Data Collection: Real-time Collection: Real-time collect the patient's recovery data through the data stream processing framework, including vital signs and nursing operation records.

[0140] 3. Data Preprocessing: Preprocess the collected data, including removing noise and filling in missing values.

[0141] 4. Data Transmission: Transmit the preprocessed data to the anomaly detection system to ensure the real-time and accuracy of the data.

[0142] The specific steps for real-time monitoring and early warning using a real-time anomaly detection algorithm include: 1. Algorithm Selection: Select appropriate anomaly detection algorithms, including statistics-based anomaly detection and machine learning-based anomaly detection.

[0143] 2. Model Training: Use historical rehabilitation data to train the anomaly detection model and learn the characteristics of normal and abnormal data; Model Optimization: Optimize the model parameters through cross-validation methods to improve the accuracy and generalization ability of the model.

[0144] 3. Risk Assessment: Real-time Monitoring: Evaluate the patient's rehabilitation risk based on real-time data streams and identify potential anomalies; Anomaly Detection: Use the anomaly detection model to detect abnormal points in the data, including abnormal vital signs and ineffective nursing operations.

[0145] 4. Warning Signal Generation: Risk Warning: Generate warning signals based on the risk assessment results to alert nursing staff to potential risks; Warning Display: Display the warning signals through the user interface and provide detailed risk information and recommended measures.

[0146] 5. Analysis of Potential Risk Factors: Factor Identification: Analyze potential risk factors, including the patient's emotional state, quality of life level, and nursing operation effectiveness; Early Warning: Based on the identified risk factors, give early warnings of possible rehabilitation risks to help nursing staff take timely measures.

[0147] VI. Interactive User Interface Module The interactive user interface module includes a data entry unit, a data query unit, and a result display unit; The data entry unit is used for nursing staff to enter nursing data. The data entry is realized through voice recognition technology and gesture recognition technology, which can support multiple voice input languages and gesture operation methods, improving the data entry efficiency. The voice input languages include Chinese, English, Spanish, French, German, Japanese, and Korean, and the gesture operation methods include sliding, clicking, long pressing, zooming, and rotating; The data query unit is used for nursing staff to query patient information and nursing data. The data positioning is realized through a full-text search engine and a data filtering algorithm, which supports fuzzy search, multi-condition filtering, and historical query record functions, and can help nursing staff find the required information; The result display unit is used to display the data analysis results and intelligent decision-making suggestions. The intuitive display of the results is realized through data visualization tools and speech synthesis technology, which supports multiple visualization charts and voice broadcast functions. The visualization charts include bar charts, line charts, and heat maps.

[0148] It should be noted that speech recognition technology is a technology that converts human speech into text. By analyzing and processing speech signals through computer algorithms, it identifies and converts them into corresponding text content. Its main functions include speech-to-text conversion: converting speech input into text format for further processing and storage; voice commands: supporting voice command operations, such as "Record the patient's body temperature as 37.5 degrees"; multi-language support: supporting speech recognition in multiple languages.

[0149] Gesture recognition technology is a technology that analyzes and processes gesture actions through computer algorithms, identifies and converts them into corresponding operations, and is usually used on touch screen devices or somatosensory devices. Its main functions include gesture operations: supporting gesture operations such as swiping, tapping, long-pressing, zooming, and rotating; user interaction: achieving user interface interaction through gesture operations, including browsing data and selecting options.

[0150] Swipe: Swiping is a gesture operation where the user moves their finger horizontally or vertically on the touch screen to execute. Swiping operations are usually used for quickly browsing content or switching pages. In an interactive user interface, swiping operations can be used to quickly browse patient information, nursing records, or analysis results.

[0151] Tap: Tapping is a gesture operation where the user touches the touch screen once to execute. Tapping operations are usually used to select or activate interface elements. In an interactive user interface, tapping operations can be used to select specific options, buttons, or links.

[0152] Long-press: Long-pressing is a gesture operation where the user presses on the touch screen for a period of time to execute, usually used to trigger more complex operations or menus. In an interactive user interface, long-pressing operations can be used to trigger more options or detailed information.

[0153] Zoom: Zooming is a gesture operation where the user uses two fingers on the touch screen to perform an enlarging or reducing action. Zooming operations are usually used to adjust the display size of content. In an interactive user interface, zooming operations can be used to adjust the display size of charts, images, or text.

[0154] Rotate: Rotating is a gesture operation where the user uses two or more fingers on the touch screen to perform a rotating action, usually used to adjust the orientation or angle of content. In an interactive user interface, rotating operations can be used to adjust the orientation of charts or images.

[0155] A full-text search engine is a technology used to quickly find specific content in a large amount of text data. By building an index, it supports efficient text search.

[0156] A data filtering algorithm is a technology used to screen out data that meets specific conditions from a large amount of data. By setting filtering conditions, it can quickly eliminate data that does not meet the requirements.

[0157] Fuzzy search: Fuzzy search is a search function that allows users to enter keywords that are not completely accurate for querying. By matching the similarity of keywords, it can return relevant results. Even if the entered keywords have spelling mistakes or are incomplete, in the data query unit, fuzzy search can help caregivers quickly find the information they need, even if they are not sure about the specific keywords.

[0158] Multi-condition filtering: Multi-condition filtering is a query function that allows users to perform data filtering according to multiple conditions simultaneously, which can improve the accuracy and efficiency of the query. In the data query unit, multi-condition filtering can help caregivers perform data filtering according to multiple conditions (patient name, care date, care operation type) simultaneously.

[0159] Historical query record function: The historical query record function is a query function. The system will save the user's query history, which is convenient for users to quickly review and repeat queries. In the data query unit, the historical query record function can help caregivers quickly review previous query records and avoid re-entering the same query conditions.

[0160] A data visualization tool is a technology that presents data in the form of graphs or charts. Through intuitive visual effects, it helps users quickly understand the meaning and trends of the data.

[0161] Speech synthesis technology is a technology that converts text content into speech, generating natural language speech output through computer algorithms.

[0162] Bar chart: A bar chart is a common visualization chart that represents the size of data through vertical or horizontal bars. The height or length of each bar represents the corresponding data value. In the result display unit, bar charts can be used to show the comparison of different categories of data.

[0163] Line chart: A line chart is a common visualization chart that represents the change trend of data by connecting data points with a line. Line charts are especially suitable for showing time series data. In the result display unit, line charts can be used to show the change trend of a patient's vital signs, including body temperature and heart rate. Caregivers can quickly understand the change situation of the patient's vital signs through the line chart.

[0164] Heat map: A heat map is a visualization chart that represents the size of data through color changes. It is usually used to show the distribution of two-dimensional data. The depth of the color represents the size of the data value. In the result display unit, heat maps can be used to show the distribution of a patient's emotional state or quality of life level.

[0165] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included within the protection scope of the present application.

Claims

1. A nursing data management optimization system for cancer patients, characterized in that, Including: A nursing data integration module, which is used to collect and integrate the nursing data of cancer patients. By classifying and transforming unstructured data into structured data, it unifies data from different sources into a standard format to achieve the integration of multi-modal data; A dynamic data standardization module, which is used to perform real-time standardization processing on the integrated multi-modal data. It generates standardization rules through a common data model CDM and machine learning algorithms, and can automatically adjust the standardization rules according to user feedback and system operation data; A secure storage module, which is used to store the standardized nursing data in a cloud distributed database. It realizes the storage, management and recovery of data through a distributed architecture, encryption technology, access control mechanism and backup technology; An analysis and prediction module, which is used to analyze and predict trends in nursing data. By using deep learning algorithms and data analysis algorithms, it extracts potential patterns and correlation relationships, evaluates the patient's emotional state, quality of life level and the effect of nursing operations, predicts the rehabilitation trend and generates a personalized nursing plan; An intelligent decision support module, which is used to provide intelligent decision support for nursing staff based on the analysis and prediction results. It can dynamically adjust decision-making suggestions according to the feedback of nursing staff, monitor the implementation effect of the nursing plan in real time, analyze potential risk factors and issue warning signals; An interactive user interface module, which is used to support nursing staff to input nursing data through voice recognition and gesture operations, query patient information and nursing data, and intuitively display the data analysis results and intelligent decision-making suggestions.

2. The optimized system for nursing data management for tumor patients according to claim 1, wherein The nursing data integration module includes: A nursing data collection unit, which collects nursing data in real time from multiple data sources through a multi-source data access interface. The multi-source data access interface includes API interfaces, Wi-Fi communication protocols, file transfer protocols and web service interfaces. The data sources include hospital information systems, wearable devices, nursing record forms and patient self-reported voice records. The nursing data includes patient basic information, nursing operation records, vital sign data, psychological assessment data and patient self-reported data; A data classification and transformation unit, which classifies nursing data into structured data and unstructured data through a data type recognition algorithm, uses natural language processing technology to transform unstructured text data into structured data, and uses image recognition technology to transform unstructured image data into structured data. The unstructured data includes unstructured text data and unstructured image data; A nursing data fusion unit, which fuses the transformed unstructured data with the original structured data. Through data mapping and format conversion technologies, it unifies data from different sources into a standard format to achieve the fusion of multi-modal data.

3. The optimized system for nursing data management for cancer patients according to claim 1, characterized in that, The dynamic data standardization module includes: A standardization rule generation unit, which is used to generate standardization rules to standardize data. It defines the structure and content of data according to the common data model CDM, and automatically generates standardization rules using machine learning algorithms. The standardization rules include data format specifications, field mapping rules, data type conversion rules, data integrity constraints and data consistency constraints; A rule update unit is used to optimize the rule generation process, dynamically update the standardization rules, collect user feedback and system operation data through a feedback loop mechanism, optimize the standardization rule generation process, and use an incremental learning algorithm to dynamically update the standardization rules. The feedback loop mechanism includes user feedback collection, system operation data monitoring, and rule optimization feedback. The incremental learning algorithm includes online learning and incremental learning.

4. The optimized nursing data management system for tumor patients according to claim 1, wherein The secure storage module includes: A distributed storage unit is used to distribute and store the standardized nursing data. Through a distributed architecture, data scalability is achieved. The distributed architecture includes a distributed file system, a distributed database, data sharding, a replica mechanism, and load balancing. A data encryption unit is used to encrypt the stored data. The SHA-256 hashing algorithm is used to hash the data, the asymmetric encryption technology is used to encrypt the data, and the blockchain technology is combined to store the data in a chain-like manner to prevent data tampering. An access control unit is used to manage users' access rights to the data. A role-based access control mechanism is adopted to assign different access rights according to the user roles. A multi-factor authentication mechanism is used to ensure that only authorized users can access and operate the data through the authentication and authorization mechanisms. A data backup unit is used to regularly back up the stored data. The regular snapshot technology is used to take snapshot backups of the data, the off-site backup technology is used to store the backup data on servers in different geographical locations, and the multi-copy backup technology is used to achieve redundancy and recoverability of the backup data.

5. The optimized system for nursing data management for cancer patients according to claim 1, wherein The analysis and prediction module includes: A data analysis unit uses deep learning algorithms and data analysis algorithms to analyze the nursing data. The deep learning algorithms include convolutional neural networks and long short-term memory networks. The convolutional neural network is used to analyze image data, and the long short-term memory network is used to analyze time series data. The data analysis algorithms include data mining techniques to extract potential patterns and correlation relationships from the nursing data, and to evaluate the patient's emotional state, quality of life level, and nursing operation effect. A trend prediction unit is used to predict the patient's recovery trend based on historical data and analysis results. The ARIMA model and deep learning prediction models are adopted, combined with the patient's individual characteristics and historical recovery data, to generate a personalized recovery path and provide a visual display of the recovery trend. A plan generation unit is used to generate personalized nursing plan suggestions based on the analysis and prediction results. A rule-based inference engine and a machine learning recommendation system are adopted to comprehensively consider the patient's emotions, quality of life, nursing operation effect, and recovery trend to generate a nursing plan.

6. The optimized nursing data management system for tumor patients according to claim 1, characterized in that, The intelligent decision support module includes: A decision recommendation unit is used to provide intelligent decision support for nursing staff based on the analysis and prediction results. A rule-based decision tree and a machine learning classifier are adopted to generate psychological counseling decisions and nursing plan optimization suggestions. The feedback adjustment unit is used to automatically adjust the decision-making suggestions according to the feedback from the nursing staff. By adopting the reinforcement learning algorithm and the feedback loop mechanism, it optimizes the decision-making process, can monitor the execution effect of the nursing plan in real time, dynamically adjust the content of the suggestions, and adapt to the patient's rehabilitation situation; The risk warning unit is used to send warning signals to the nursing staff according to the patient's rehabilitation risk assessment results. By adopting the real-time data stream processing framework and the anomaly detection algorithm, it conducts real-time monitoring and warning, and can analyze potential risk factors to warn of possible rehabilitation risks in advance.

7. The optimized system for nursing data management for cancer patients according to claim 1, wherein The interactive user interface module includes: The data entry unit is used for the nursing staff to enter nursing data. It realizes data entry through voice recognition technology and gesture recognition technology, can support multiple voice input languages and gesture operation methods, improves data entry efficiency. The voice input languages include Chinese, English, Spanish, French, German, Japanese and Korean, and the gesture operation methods include swiping, clicking, long pressing, zooming and rotating; The data query unit is used for the nursing staff to query patient information and nursing data. It realizes data positioning through the full-text search engine and the data filtering algorithm, supports fuzzy search, multi-condition filtering and historical query record functions, and can help the nursing staff find the required information; The result display unit is used to display the data analysis results and intelligent decision-making suggestions. It realizes the intuitive display of the results through data visualization tools and speech synthesis technology, supports multiple visualization charts and voice broadcast functions. The visualization charts include bar charts, line charts and heat maps.

Images