Analysis system based on tumor nursing field data model

Abstract

The invention relates to the technical field of tumor nursing analysis, in particular to an analysis system based on a data model in the field of tumor nursing. The system comprises a tumor nursing field data model construction module, a nursing intervention effect evaluation module, a tumor nursing risk estimation module and a model optimization module. The tumor nursing field data model construction module constructs a tumor nursing field data model by acquiring historical tumor nursing field data and image data; the tumor nursing field data model construction module constructs a tumor nursing field data model by acquiring historical tumor nursing field data and image data; the tumor nursing risk estimation module evaluates the nursing risk and calculates the nursing complication probability based on the nursing intervention effect data, and identifies the potential risk in advance; the model optimization module is used for optimizing the data model by detecting nursing abnormity; according to the invention, by optimizing the data model in the field of tumor nursing, more accurate tumor nursing is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of tumor nursing analysis, and in particular to an analysis system based on a data model in the field of tumor nursing. Background Art

[0002] Tumor care is a complex and personalized process involving many factors, and personalized nursing intervention and formulation are becoming increasingly important. Traditional tumor care often has a one-size-fits-all problem. Historical tumor clinical data, such as physiological indicators, pathology reports, imaging results, etc., cover a wealth of information, but how to efficiently analyze this information, conduct risk assessment, and optimize the nursing path is still a difficult problem. Data models in the field of tumor care are gradually becoming more precise and automated. These models can dynamically adjust care based on specific data, evaluate nursing effects, and even predict nursing risks and complications, optimize nursing resource allocation, and achieve personalized, accurate, and efficient tumor care. However, a traditional analysis system based on a data model in the field of tumor care has the problem of inaccurate analysis of historical tumor conditions and inaccurate analysis of nursing risks. Summary of the invention

[0003] Based on this, it is necessary to provide an analysis system based on the data model in the field of tumor nursing to solve at least one of the above technical problems.

[0004] To achieve the above purpose, an analysis system based on a data model in the field of tumor nursing includes the following steps: A tumor nursing field data model construction module is used to obtain historical tumor nursing field data; collect historical tumor images based on historical tumor nursing field data to obtain historical tumor image data; construct a tumor nursing field data model based on historical tumor image data and historical tumor nursing field data to obtain a tumor nursing field data model; The nursing intervention effect evaluation module is used to analyze the historical tumor structure characteristics using historical tumor imaging data, thereby obtaining historical tumor structure characteristic data; to formulate personalized nursing for the historical tumor structure characteristic data using the tumor nursing field data model, thereby obtaining personalized tumor nursing data; to evaluate the nursing intervention effect based on the personalized tumor nursing data and historical tumor structure characteristic data, thereby obtaining tumor nursing intervention effect data; The tumor nursing risk estimation module is used to estimate the tumor nursing risk based on the tumor nursing intervention effect data, thereby obtaining tumor nursing risk data; calculate the probability of tumor nursing complications based on the tumor nursing risk data, thereby obtaining tumor nursing complication probability data; The model optimization module performs personalized nursing anomaly detection on tumor personalized nursing data based on tumor nursing complication probability data and tumor nursing risk data to obtain tumor personalized nursing anomaly data; and optimizes the data model in the tumor nursing field according to the tumor personalized nursing anomaly data to obtain the tumor nursing field data optimization model.

[0005] The present invention can effectively optimize the tumor nursing process and improve nursing efficiency through the collaborative work of each module. The tumor nursing field data model construction module provides important basic data support by acquiring historical tumor nursing field data and image data, thereby providing strong guarantee for nursing intervention effect evaluation and risk estimation. The nursing intervention effect evaluation module uses the combination of historical image data and personalized nursing data to provide multi-angle data support, making the formulation of nursing more refined and personalized, thereby helping to improve the accuracy of nursing effect. The tumor nursing risk estimation module can evaluate potential nursing risks in advance, identify deviations and inconsistencies in the nursing process, promote safety and standardization in nursing work, and reduce the probability of accidents in the nursing process through the analysis of nursing intervention effect data. The model optimization module detects and adjusts the abnormality of personalized nursing data, can continuously optimize the nursing process, improve the quality and efficiency of overall nursing, and ensure that nursing is better adapted to different needs and nursing environments through continuous feedback optimization. Overall, this system provides an efficient and dynamic nursing management method through the integration and feedback of data models, which can provide more accurate information support for nursing staff, thereby improving the quality and efficiency of nursing services. Therefore, the present invention optimizes the traditional analysis system based on the data model in the field of tumor nursing, solves the problem of inaccurate analysis of historical tumor conditions and inaccurate analysis of nursing risks in the traditional analysis system based on the data model in the field of tumor nursing, and improves the accuracy of historical tumor condition analysis and nursing risk analysis.

[0006] Preferably, the tumor nursing field data model construction module includes the following functions: Access to historical oncology nursing field data; Perform historical tumor nursing desensitization processing based on historical tumor nursing field data, thereby obtaining historical tumor nursing field desensitization data; Use historical desensitized data in the field of oncology nursing to collect pathways in the field of oncology nursing, thereby obtaining oncology nursing pathway data; The historical tumor physical signs and conditions are collected based on the desensitized data in the field of historical tumor nursing, thereby obtaining the historical tumor physical signs and conditions data; Collect historical tumor images based on the desensitized data in the field of historical tumor nursing, thereby obtaining historical tumor image data; Based on historical tumor sign status data, historical tumor imaging data and tumor nursing pathway data, a data model for the tumor nursing field is constructed to obtain a data model for the tumor nursing field.

[0007] The present invention effectively optimizes the processing and application of nursing data through a series of steps, improves the transparency and efficiency of the nursing process, and the acquisition of historical tumor nursing field data provides a sufficient data basis for subsequent model construction, ensuring that the system has good data support. Through the desensitization of historical tumor nursing, privacy is protected, and the integrity and applicability of the data can be guaranteed, which is conducive to subsequent safe use. Then, the collection of tumor nursing pathway data provides a detailed record of each link in the nursing process, which can provide key data support for the analysis and adjustment of nursing. The collection of historical tumor sign status data and imaging data helps to depict the comprehensive characteristics of the nursing object and provides more dimensional data input for the multi-dimensional construction of the data model. By integrating these data, a comprehensive and detailed tumor nursing field data model is constructed, which improves the relevance and operability of the data, so that the nursing process can be dynamically adjusted and optimized according to the trend of historical data. The system provides a data-driven, flexible optimization tool for nursing staff, further improving the accuracy and efficiency of nursing services.

[0008] Preferably, the construction of a tumor nursing field data model based on historical tumor sign status data, historical tumor imaging data, and tumor nursing pathway data includes: Calculating the progress of the tumor nursing pathway according to the tumor nursing pathway data, thereby obtaining the progress data of the tumor nursing pathway; According to the progress data of tumor nursing pathway, the tumor nursing change process is recorded for the historical tumor physical sign status data, so as to obtain the tumor nursing change process record data; Based on the tumor nursing change process record data and tumor nursing pathway progress data, the tumor nursing change law analysis is performed on the historical tumor imaging data to obtain the historical tumor nursing change law data; A data model for the field of tumor nursing is constructed based on the historical data on the changing rules of tumor nursing and the data on the recording of the changing process of tumor nursing, and a data model for the field of tumor nursing is obtained.

[0009] The present invention can effectively track each link in the nursing process through tumor nursing pathway progress calculation, provide a clear view of nursing progress, and enable the nursing team to accurately grasp the nursing progress and the focus of each stage. This process can not only provide real-time nursing status feedback for nursing staff, but also help identify potential bottlenecks in the nursing process and promote efficient management of the nursing process. Further, based on the nursing pathway progress data, the nursing change process record of historical tumor sign status data can reveal subtle changes in the nursing process, provide detailed tracks for subsequent data analysis, and help establish an accurate nursing model. On this basis, the nursing change law analysis of historical tumor imaging data is carried out, the influence of nursing intervention on tumor changes is deeply mined, and valuable trend information is extracted to provide a scientific basis for the optimization of nursing models and the improvement of future data models. The data model in the field of tumor nursing is constructed by integrating the historical tumor nursing change law data and the change process record data, which can not only improve the accuracy of the data model in multiple dimensions, but also help improve the flexibility and adaptability of nursing management, thereby promoting the continuous improvement of nursing service quality.

[0010] Preferably, the nursing intervention effect evaluation module includes: Using historical tumor imaging data to collect historical tumor ultrasonic images, thereby obtaining historical tumor ultrasonic image data; Perform historical tumor structural feature analysis based on historical tumor ultrasonic image data, thereby obtaining historical tumor structural feature data; Use the tumor nursing field data model to formulate personalized nursing care for historical tumor structural feature data, thereby obtaining personalized tumor nursing data; The effectiveness of nursing intervention is evaluated based on personalized tumor nursing data and historical tumor structural characteristic data to obtain tumor nursing intervention effectiveness data.

[0011] The present invention uses historical tumor ultrasonic image data to collect ultrasonic images, providing detailed basic imaging data for the subsequent nursing process, helping to accurately describe the structural characteristics of the tumor and ensure the objectivity and scientificity of the data. This process provides reliable imaging data support for analyzing the evolution of the tumor, and helps to identify the key features of tumor morphology and changes. Analysis of tumor structural characteristics based on historical tumor ultrasonic image data can reveal important information about the morphology, boundaries, texture, etc. of the tumor, and provide data support for the design of further nursing models. By using data models in the field of tumor nursing to formulate personalized nursing for these tumor structural characteristic data, nursing interventions can be optimized according to different specific needs, thereby improving the personalization and accuracy of the nursing process.

[0012] Preferably, the historical tumor structural feature analysis based on historical tumor ultrasonic image data includes: Performing ultrasound image contrast enhancement processing on historical tumor ultrasound image data, thereby obtaining tumor ultrasound image contrast enhanced data; Calculating the roughness of the tumor boundary of the ultrasound image according to the contrast enhancement data of the tumor ultrasound image to obtain the roughness data of the tumor boundary; Using the tumor boundary roughness data to perform ultrasound image tumor edge feature analysis, thereby obtaining ultrasound image tumor edge feature data; Calculate the external morphological features of the tumor based on the tumor edge feature data and tumor boundary roughness data of the ultrasound image to obtain the external morphological feature data of the tumor; Performing tumor texture feature analysis based on tumor ultrasonic image contrast enhancement data to obtain tumor texture feature data; Historical tumor structure feature data is analyzed based on tumor texture feature data and tumor external morphology feature data to obtain historical tumor structure feature data.

[0013] The present invention effectively improves the image details by contrast enhancement processing of historical tumor ultrasonic image data, increases the distinction between the tumor area and the surrounding tissue, and thus obtains clearer image data. This processing provides higher quality image information for subsequent tumor boundary roughness calculation, helps to accurately evaluate the morphological characteristics of the tumor boundary, and thus reveals the growth law of the tumor. By analyzing the tumor boundary roughness data, the subtle structural changes of the tumor edge can be effectively described, thereby providing a scientific basis for analyzing the edge characteristics of the tumor. This analysis process helps to deeply understand the boundary characteristics of the tumor, improve the sensitivity to tumor edge changes, and lay a data foundation for subsequent morphological feature calculations. Further, by analyzing the tumor edge characteristics and roughness data, the external morphology of the tumor can be more comprehensively quantified, thereby providing more dimensional data support for the comprehensive feature analysis of the tumor. At the same time, by analyzing the texture characteristics of the tumor, the detailed characteristics of the tumor are revealed, thereby obtaining important information about the heterogeneity of tumor tissue. Combining the tumor texture feature data with the external morphological feature data, a comprehensive historical tumor structural feature analysis is performed to obtain a comprehensive and detailed structural cognition of the tumor.

[0014] Preferably, the tumor texture feature analysis based on the tumor ultrasound image contrast enhancement data includes the following: Calculate the local gradient of the ultrasound image tumor according to the contrast enhancement data of the ultrasound image of the tumor to obtain the local gradient data of the ultrasound image tumor; Decomposing the multi-scale features of the tumor according to the local gradient data of the tumor in the ultrasound image, thereby obtaining the multi-scale feature data of the tumor; Based on the multi-scale feature data of the tumor and the local gradient data of the tumor in the ultrasound image, the hierarchical structure of the tumor texture is detected to obtain the hierarchical texture structure data of the tumor; The tumor local texture change rate is calculated using the tumor layered texture structure data, thereby obtaining the tumor local texture change rate data; Tumor texture feature analysis is performed based on the tumor local texture change rate data and the tumor hierarchical texture structure data to obtain tumor texture feature data.

[0015] The present invention significantly improves the visibility of the tumor area by contrast enhancement processing on the tumor ultrasonic image, making its local gradient features more obvious, thereby providing more accurate image data for the local gradient calculation of the tumor. This calculation helps to reveal the local changes inside the tumor, thereby providing detailed gradient information for multi-scale feature decomposition, so that the different scale features of the tumor are comprehensively presented. These multi-scale feature data help to analyze the microstructure of the tumor in detail and further promote the multi-level feature exploration of tumor tissue. With the help of the combination of tumor multi-scale feature data and tumor local gradient data, hierarchical structure detection of tumor texture is realized, providing reliable data support for in-depth understanding of the tissue heterogeneity of the tumor. Next, by calculating the local texture change rate of the tumor, the dynamic change characteristics of the tumor area are revealed, providing more accurate texture information, helping to further evaluate the growth characteristics of the tumor, combining the local texture change rate data of the tumor with the hierarchical texture structure data, and performing tumor texture feature analysis, which can fully understand the texture characteristics of the tumor and its changing laws.

[0016] Preferably, the method of using the tumor nursing field data model to formulate personalized nursing care for historical tumor structural feature data includes: Calculating the tumor structure volume according to historical tumor structure characteristic data, thereby obtaining tumor structure volume data; The degree of tumor vascularization is detected according to historical tumor structural characteristic data, thereby obtaining the data of the degree of tumor vascularization; Tumor invasiveness is estimated based on tumor vascularization degree data and tumor structure volume data, thereby obtaining tumor invasiveness data; Analyze the abnormal state of tumor structure according to the tumor invasion data and the tumor vascularization degree data, so as to obtain the abnormal state of tumor structure; The tumor nursing field data model is used to formulate personalized nursing care for tumor structural abnormality data, thereby obtaining personalized tumor nursing data.

[0017] The present invention calculates the tumor structure volume based on historical tumor structure feature data, accurately assesses the size and shape of the tumor, and provides basic data for further analysis of the tumor growth pattern. At the same time, the detection of the degree of tumor vascularization helps to reveal the development of the vascular network in the tumor, reflects the blood supply of the tumor, and thus provides important information about the tumor microenvironment. These vascularization degree data combined with the tumor structure volume data help to more comprehensively assess the invasive characteristics of the tumor, infer the extent of its impact on surrounding tissues, and provide a reference for subsequent personalized nursing formulation. By combining tumor invasiveness data with vascularization degree data, the analysis of abnormal tumor structure conditions is realized, and abnormal changes in tumors are revealed, thereby guiding the key directions of nursing intervention. Using the data model in the field of tumor nursing, personalized nursing is formulated for abnormal tumor structure data, and nursing is tailored and optimized according to the specific characteristics of each tumor, providing scientific support for personalized nursing needs, thereby achieving better nursing results.

[0018] Preferably, the tumor nursing risk estimation module includes the following functions: According to the tumor nursing intervention effect data, the abnormal effect of tumor nursing intervention is estimated, so as to obtain the abnormal effect data of tumor nursing intervention; Estimating tumor nursing risk based on abnormal effect data of tumor nursing intervention, thus obtaining tumor nursing risk data; Based on the abnormal effect data of tumor nursing intervention and tumor nursing risk data, tumor nursing side effects are evaluated to obtain tumor nursing side effects data; The probability of tumor nursing complications is calculated based on the tumor nursing side effect data and tumor nursing risk data to obtain the tumor nursing complication probability data.

[0019] The present invention estimates the abnormal effects of tumor nursing intervention effect data and identifies abnormal reactions or deviations in the nursing process, thereby providing a scientific basis for the adjustment of nursing measures. Nursing risk estimation is performed in combination with abnormal effect data of tumor nursing intervention, which can effectively evaluate potential risk factors in the nursing process and provide support for the optimization of subsequent intervention measures. On this basis, further analysis of tumor nursing side effect data is helpful to understand the adverse reactions in the nursing process and evaluate the safety and effectiveness of nursing measures. Combining tumor nursing side effect data and nursing risk data, the probability of tumor nursing complications is calculated, which helps to quantify the probability of complications, provide early warning of potential complications, and facilitate timely adjustment of nursing.

[0020] Preferably, the tumor nursing risk estimation based on abnormal effect data of tumor nursing intervention includes: Based on the abnormal effect data of tumor nursing intervention, nonlinear change analysis of tumor nursing effect was performed to obtain nonlinear change data of tumor nursing effect; According to the nonlinear change data of tumor nursing effect, the normal distribution imbalance of tumor nursing is calculated to obtain the normal distribution imbalance data of tumor nursing; Tumor nursing deviation is calculated based on the imbalanced data of tumor nursing normal distribution and the nonlinear change data of tumor nursing effect, so as to obtain tumor nursing deviation data; Based on the abnormal effect data of tumor nursing intervention, abnormal influencing factors of tumor nursing were analyzed to obtain abnormal influencing factors of tumor nursing; Tumor nursing risk estimation is performed based on abnormal tumor nursing influencing factor data and tumor nursing deviation data to obtain tumor nursing risk data.

[0021] The present invention reveals the complex change rules of nursing effects by performing nonlinear change analysis on abnormal effect data of tumor nursing intervention, and provides a theoretical basis for optimizing nursing. Calculating the imbalance of normal distribution of tumor nursing effects helps to identify the imbalance in nursing effects, and then more accurately adjust nursing intervention measures. By calculating the nursing deviation data, the difference between the nursing effect and the expectation is quantified, providing data support for subsequent nursing adjustments. Based on the analysis of abnormal influencing factors of tumor nursing, the key factors affecting the nursing effect can be identified, so that corresponding measures can be taken in a targeted manner. By combining the abnormal influencing factor data and the nursing deviation data, nursing risk estimation is performed, which helps to accurately assess potential nursing risks.

[0022] Preferably, the model optimization module includes the following functions: Based on the probability data of tumor nursing complications and the risk data of tumor nursing, personalized nursing anomaly detection is performed on the personalized tumor nursing data to obtain the abnormal data of personalized tumor nursing; Adjusting the personalized tumor care data according to the abnormal personalized tumor care data to obtain personalized tumor care adjustment data; The data model in the field of tumor nursing is optimized according to the abnormal data of personalized tumor nursing and the adjusted data of personalized tumor nursing, so as to obtain the data optimization model in the field of tumor nursing.

[0023] The present invention detects abnormalities in personalized tumor nursing data based on tumor nursing complication probability data and tumor nursing risk data, timely identifies potential problems in the nursing process, and ensures the effectiveness and safety of nursing measures. Personalized nursing adjustments are made to the detected abnormal nursing data to ensure nursing intervention optimization, enhance nursing adaptability and flexibility, and thus improve nursing effects. Optimizing the data model in the field of tumor nursing based on abnormal nursing data and nursing adjustment data helps to improve the accuracy and reliability of the data model.

[0024] The present invention is that, through the collaborative work of each module, it can effectively optimize the tumor nursing process and improve the nursing efficiency. The tumor nursing field data model construction module provides important basic data support by acquiring historical tumor nursing field data and image data, thereby providing strong guarantee for nursing intervention effect evaluation and risk estimation. The nursing intervention effect evaluation module uses the combination of historical image data and personalized nursing data to provide multi-angle data support, making the formulation of nursing more refined and personalized, thereby helping to improve the accuracy of nursing effect. The tumor nursing risk estimation module can evaluate potential nursing risks in advance, identify deviations and inconsistencies in the nursing process, promote safety and standardization in nursing work, and reduce the probability of accidents in the nursing process through the analysis of nursing intervention effect data. The model optimization module detects and adjusts the abnormality of personalized nursing data, can continuously optimize the nursing process, improve the quality and efficiency of overall nursing, and ensure that nursing better adapts to different needs and nursing environments through continuous feedback optimization. Overall, this system provides an efficient and dynamic nursing management method through the integration and feedback of data models, which can provide more accurate information support for nursing staff, thereby improving the quality and efficiency of nursing services. Therefore, the present invention optimizes the traditional analysis system based on the data model in the field of tumor nursing, solves the problem of inaccurate analysis of historical tumor conditions and inaccurate analysis of nursing risks in the traditional analysis system based on the data model in the field of tumor nursing, and improves the accuracy of historical tumor condition analysis and nursing risk analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a module diagram of an analysis system based on a data model in the field of tumor nursing; Figure 2 for Figure 1 Functional flow chart of the nursing intervention effect evaluation module; Figure 3 for Figure 1 Schematic diagram of the functional flow of the tumor nursing risk estimation module; The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0026] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0027] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different network and / or processor methods and / or microcontroller methods.

[0028] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0029] To achieve this, please refer to Figures 1 to 3 , an analysis system based on a data model in the field of tumor nursing, the system comprising the following modules: A tumor nursing field data model construction module is used to obtain historical tumor nursing field data; collect historical tumor images based on historical tumor nursing field data to obtain historical tumor image data; construct a tumor nursing field data model based on historical tumor image data and historical tumor nursing field data to obtain a tumor nursing field data model; The nursing intervention effect evaluation module is used to analyze the historical tumor structure characteristics using historical tumor imaging data, thereby obtaining historical tumor structure characteristic data; to formulate personalized nursing for the historical tumor structure characteristic data using the tumor nursing field data model, thereby obtaining personalized tumor nursing data; to evaluate the nursing intervention effect based on the personalized tumor nursing data and historical tumor structure characteristic data, thereby obtaining tumor nursing intervention effect data; The tumor nursing risk estimation module is used to estimate the tumor nursing risk based on the tumor nursing intervention effect data, thereby obtaining tumor nursing risk data; calculate the probability of tumor nursing complications based on the tumor nursing risk data, thereby obtaining tumor nursing complication probability data; The model optimization module performs personalized nursing anomaly detection on tumor personalized nursing data based on tumor nursing complication probability data and tumor nursing risk data to obtain tumor personalized nursing anomaly data; and optimizes the data model in the tumor nursing field according to the tumor personalized nursing anomaly data to obtain the tumor nursing field data optimization model.

[0030] In the embodiment of the present invention, reference Figure 1 FIG. 1 is a schematic diagram of an analysis system based on a tumor nursing field data model of the present invention. In this example, the analysis system based on a tumor nursing field data model includes the following modules: S1: Tumor nursing field data model construction module, used to obtain historical tumor nursing field data; collect historical tumor images based on historical tumor nursing field data, so as to obtain historical tumor image data; construct a tumor nursing field data model based on historical tumor image data and historical tumor nursing field data, so as to obtain a tumor nursing field data model; In the embodiment of the present invention, the tumor nursing field data model construction module realizes the establishment of the nursing data model by collecting and processing historical data. The acquisition of historical tumor nursing field data needs to rely on standardized data storage systems, such as clinical data storage servers, medical image databases, etc. These data include tumor nursing pathway data, medical records, nursing intervention data, physical sign monitoring data, and related image information. When extracting data, a data extraction tool is used to parse structured and unstructured data from the original storage medium, and the data is stored in a standardized data table after cleaning. Subsequently, based on the historical tumor nursing field data in the data storage system, the image acquisition system is used to perform the extraction of historical tumor images, and the image formats include CT, MRI, ultrasound images, etc. Image acquisition uses a DICOM format parsing tool to convert image files into matrix data that can be calculated and analyzed, and normalizes the image files in combination with historical tumor nursing field data to unify the scale and ensure the comparability of data from different sources. After the data preparation is completed, based on the historical tumor image data and the historical tumor nursing field data, a data association algorithm is used to perform feature mapping and establish a data index so that the image features and nursing data form a computable association relationship. The data model is constructed by using data mining algorithms to screen out highly relevant nursing indicators, and the storage structure of the nursing data model is defined based on the database architecture. Computer data modeling tools are used to generate standardized tumor nursing data models and perform integrity verification.

[0031] S2: Nursing intervention effect evaluation module, which is used to use historical tumor imaging data to analyze historical tumor structural characteristics, thereby obtaining historical tumor structural characteristic data; use tumor nursing field data model to formulate personalized nursing for historical tumor structural characteristic data, thereby obtaining tumor personalized nursing data; evaluate nursing intervention effects based on tumor personalized nursing data and historical tumor structural characteristic data, thereby obtaining tumor nursing intervention effect data; In the embodiment of the present invention, the core task of the nursing intervention effect evaluation module is to analyze the effect of nursing intervention based on image data, pre-process the historical tumor image data using image processing software, and use image enhancement methods, including histogram equalization, edge enhancement, noise removal, etc., to improve image quality and ensure analysis accuracy. Subsequently, feature extraction is performed on the image data, and the tumor area is segmented using methods such as regional growth, K-means clustering or convolution kernel filtering to obtain structural features such as tumor boundaries, morphology, and density, and form historical tumor structural feature data. Based on the structural feature data, personalized nursing is formulated using the tumor nursing field data model, and the basis for personalized nursing is set, such as the trend of physical signs, the trend of image data, etc., and the nursing intervention measures are described in a quantitative manner. Subsequently, based on historical data and personalized nursing data, the nursing intervention effect is analyzed, and the change amplitude of tumor morphology, size, density and other features after nursing intervention is calculated using methods such as change trend calculation and statistical regression analysis, and the results are stored as tumor nursing intervention effect data.

[0032] S3: Tumor nursing risk estimation module, used to estimate tumor nursing risk based on tumor nursing intervention effect data, so as to obtain tumor nursing risk data; calculate the probability of tumor nursing complications based on tumor nursing risk data, so as to obtain tumor nursing complication probability data; In an embodiment of the present invention, the tumor nursing risk estimation module calculates risk data and evaluates the probability of complications by quantifying the impact of nursing intervention, using nursing intervention effect data, and adopting a time series analysis method to predict the trend of tumor change characteristics after nursing intervention, and calculate the fluctuation range of the intervention effect. Subsequently, the tumor nursing risk is calculated based on statistical methods, including the fluctuation range of physiological indicators under the influence of nursing measures, the trend of image data changes, etc., and the samples of nursing-related risks in historical data are compared to calculate the risk probability and generate tumor nursing risk data. On this basis, based on the risk data, the conditional probability calculation method is used to calculate the probability of nursing complications in combination with the risk factors recorded in the nursing process. The calculation of the probability of complications is based on the historical data statistics in the nursing field, and regression analysis is performed on the nursing-related events that have occurred. A complication probability model is established, and it is verified by actual nursing data to obtain tumor nursing complication probability data.

[0033] S4: Model optimization module, which performs personalized nursing anomaly detection on tumor personalized nursing data based on tumor nursing complication probability data and tumor nursing risk data to obtain tumor personalized nursing anomaly data; optimizes the tumor nursing field data model according to the tumor personalized nursing anomaly data to obtain the tumor nursing field data optimization model.

[0034] In an embodiment of the present invention, a model optimization module is used to detect abnormal situations in personalized nursing, and optimize and adjust the data model. The personalized nursing data is analyzed using tumor nursing complication probability data and nursing risk data, and anomaly detection algorithms such as Mahalanobis distance, KNN anomaly detection, etc. are used to identify abnormal situations in nursing data and extract personalized nursing abnormal data. The judgment of abnormal data is based on multi-factor analysis, including fluctuations in physiological indicators after nursing intervention, abnormal trends in imaging data, etc. After identifying the abnormality, data correction is required, and the abnormal nursing data is compared and analyzed with historical data. The data smoothing method is used to eliminate the impact of random fluctuations, and adjustments are made based on similar nursing cases to ensure the stability of nursing data. After the data is adjusted, the data model in the field of tumor nursing is optimized, and methods such as data cleaning, feature selection, and parameter adjustment are used to ensure that the model can adapt to new nursing situations and improve the accuracy of personalized nursing data.

[0035] Preferably, the tumor nursing field data model construction module includes the following functions: Access to historical oncology nursing field data; Perform historical tumor nursing desensitization processing based on historical tumor nursing field data, thereby obtaining historical tumor nursing field desensitization data; Use historical desensitized data in the field of oncology nursing to collect pathways in the field of oncology nursing, thereby obtaining oncology nursing pathway data; The historical tumor physical signs and conditions are collected based on the desensitized data in the field of historical tumor nursing, thereby obtaining the historical tumor physical signs and conditions data; Collect historical tumor images based on the desensitized data in the field of historical tumor nursing, thereby obtaining historical tumor image data; Based on historical tumor sign status data, historical tumor imaging data and tumor nursing pathway data, a data model for the tumor nursing field is constructed to obtain a data model for the tumor nursing field.

[0036] In an embodiment of the present invention, the acquisition of historical tumor nursing field data depends on a multi-source data storage system, including an electronic medical record database, a nursing information management system, a medical imaging storage system, etc. The first step in data acquisition is to access the nursing data repository and extract structured data such as nursing records, medical records, nursing measures, and intervention plans related to tumor nursing. The data storage system provides standardized interfaces, such as SQL queries or REST API interfaces, to achieve batch extraction of data. For unstructured data, such as nursing reports, handwritten records, etc., natural language processing (NLP) technology is used for text parsing to extract key nursing information from unstructured text and store it in a structured data table. After data extraction, the data format is standardized and converted, such as a unified unit system, time format, etc., to ensure data consistency and computability. The data cleaning tool fills or deletes outliers and missing values ​​to improve data quality. All historical tumor nursing field data is stored in the central data warehouse, and all acquired historical tumor nursing field data must be desensitized. Desensitization involves data anonymization, pseudo-anonymization, and encryption of sensitive information. Data mapping methods are used to convert directly identifiable information such as patient names, ID numbers, and contact information into unique identifiers (UUIDs) to ensure that the data cannot be directly traced back to specific individuals. Subsequently, hash encryption technology is used to irreversibly encrypt specific data fields such as medical record numbers. For text data that leaks privacy, such as personalized descriptions in nursing records, natural language processing algorithms are used for automatic de-identification processing, and information related to patient characteristics is generalized, such as replacing "45-year-old males" with "40-50-year-old males". All desensitized data is stored in a dedicated desensitized data warehouse. After data desensitization is completed, historical desensitized data in the field of oncology nursing is used to extract nursing pathways. Nursing pathway collection is based on the nursing process management system to extract key time nodes and nursing measures in different nursing stages, including admission assessment, nursing plan formulation, nursing intervention implementation, follow-up records, etc. Data extraction uses a time series analysis method to arrange nursing data in chronological order and calculate indicators such as the time interval, duration, and frequency of nursing measures of different nursing interventions to form complete nursing pathway data. The nursing pathway data storage adopts a structured database mode and sets a unique index to ensure the retrievability of the pathway data. The collection of historical tumor sign status data relies on the sign monitoring system and nursing record data, and extracts historical daily sign monitoring data from the nursing database, including blood pressure, body temperature, pulse, respiratory rate, blood oxygen saturation, etc. The data collection adopts batch query mode, performs time series analysis on long-term monitoring data, and calculates the trend of sign changes. After the sign data is extracted, combined with the nursing records, the relationship between nursing intervention and sign changes is analyzed, such as the change range of sign indicators after the implementation of a certain nursing measure.After data cleaning, the sign data is cleaned, outliers are removed, and the data is stored in the sign data repository. The collection of historical tumor image data depends on the medical image storage and communication system (PACS). The image storage management interface is used to retrieve image data that matches the desensitized nursing data, including CT, MRI, and ultrasound images. The image data is extracted using the DICOM format parsing tool to convert the image file into a standardized data format. Subsequently, the image is preprocessed using an image processing algorithm, including denoising, contrast enhancement, and size normalization, to ensure the availability of the image. After the image data is prepared, the association index between the image and the nursing data is established in combination with the nursing pathway data and the sign status data, and stored in the image database. Based on the historical tumor sign status data, historical tumor image data, and tumor nursing pathway data, a data model is constructed, and feature selection is performed on all data to screen out key features related to nursing, such as nursing measures, sign changes, and image feature changes. Subsequently, data fusion technology is used to multi-dimensionally associate the sign data, image data, and nursing pathway data, such as establishing a causal relationship between nursing measures and sign changes, and a time series relationship between image changes and nursing pathways. After data fusion, define the storage structure of the data model, including field definition and index setting of the data table, use data modeling tools to generate a complete data model for the field of tumor nursing, and perform consistency verification.

[0037] Preferably, the construction of a tumor nursing field data model based on historical tumor sign status data, historical tumor imaging data, and tumor nursing pathway data includes: Calculating the progress of the tumor nursing pathway according to the tumor nursing pathway data, thereby obtaining the progress data of the tumor nursing pathway; According to the progress data of tumor nursing pathway, the tumor nursing change process is recorded for the historical tumor physical sign status data, so as to obtain the tumor nursing change process record data; Based on the tumor nursing change process record data and tumor nursing pathway progress data, the tumor nursing change law analysis is performed on the historical tumor imaging data to obtain the historical tumor nursing change law data; A data model for the field of tumor nursing is constructed based on the historical data on the changing rules of tumor nursing and the data on the recording of the changing process of tumor nursing, and a data model for the field of tumor nursing is obtained.

[0038] In the embodiment of the present invention, the tumor nursing pathway data includes multiple nursing links, each of which corresponds to specific nursing measures and time nodes. The tumor nursing pathway progress calculation is based on the time series analysis method, and the time progress calculation is performed on different nodes in the nursing pathway to determine the actual execution of the nursing pathway. The nursing pathway data is sorted according to the time sequence of the nursing plan, and key data such as the execution time, nursing frequency, and nursing goal completion of the nursing intervention measures are extracted. Subsequently, the nursing plan time is compared with the actual execution time, the time deviation is calculated, and the completion ratio of each nursing stage is counted. The progress calculation adopts a cumulative calculation method to accumulate the completed nursing links to obtain the execution progress of the overall nursing pathway. For data with abnormal deviations, an interpolation algorithm is used to correct the abnormalities to ensure the stability and accuracy of the nursing pathway progress data. After the calculation is completed, the tumor nursing pathway progress data is stored in the nursing progress database. Based on the tumor nursing pathway progress data, the nursing change process of the historical tumor physical sign status data is recorded to form a complete nursing change process data. The nursing change process record needs to perform stage-by-stage statistics on the physical sign data, and divide the physical sign data into different nursing stages according to the time nodes of the nursing pathway. For example, in the postoperative care stage, the physical sign data include changes in blood pressure, blood oxygen, heart rate, etc. during the postoperative recovery period; in the radiotherapy and chemotherapy care stage, the physical sign data include changes in immune function, etc. Data collection uses batch query to extract time series data from the physical sign monitoring database and calculate the change trend of physical sign indicators, such as growth rate, fluctuation amplitude, etc. Subsequently, the nursing pathway data is associated with the physical sign data using data mapping technology to record the physical sign change process corresponding to each nursing intervention measure, forming a complete nursing change process record data. The data is stored using a time series database. After obtaining the nursing change process record data, the historical tumor imaging data is combined with the tumor nursing pathway progress data to analyze the tumor nursing change law. Image data analysis uses a DICOM format parsing tool to preprocess medical images at different time points, remove noise, enhance contrast, and perform image alignment processing to ensure that images at different time points have the same spatial resolution. Subsequently, image feature extraction technology is used to extract key features of the image, such as lesion boundary changes, density changes, and vascular structure changes. Through time series analysis, the changing trend of image features with the progress of nursing is calculated, and matched with the nursing change process record data to analyze the impact of nursing measures on image changes. For example, at a certain stage of the nursing pathway, whether nursing intervention leads to stability or change of image features, whether there are obvious boundary changes of tumor lesions in the image data, etc. Based on the historical tumor nursing change law data and tumor nursing change process record data, the data model of tumor nursing field is constructed. The first step in data model construction is feature fusion, which integrates the nursing pathway progress data, nursing change process data, and image change law data, and establishes data association according to the time sequence of the nursing pathway.Subsequently, data normalization is performed to ensure that all data are in a unified standard unit and format, such as the progress of the nursing path is expressed as a percentage, the change of physical signs is expressed as a change rate, and the change of images is expressed as the change rate of pixel features. After the data is integrated, multidimensional data storage technology is used to establish data indexes for rapid retrieval and query, and all nursing data are stored in a structured form in the nursing data warehouse to form a complete data model in the field of oncology nursing.

[0039] Preferably, the nursing intervention effect evaluation module includes: Using historical tumor imaging data to collect historical tumor ultrasonic images, thereby obtaining historical tumor ultrasonic image data; Perform historical tumor structural feature analysis based on historical tumor ultrasonic image data, thereby obtaining historical tumor structural feature data; Use the tumor nursing field data model to formulate personalized nursing care for historical tumor structural feature data, thereby obtaining personalized tumor nursing data; The effectiveness of nursing intervention is evaluated based on personalized tumor nursing data and historical tumor structural characteristic data to obtain tumor nursing intervention effectiveness data.

[0040] As an example of the present invention, refer to Figure 2 As shown, Figure 1 The functional flow diagram of the nursing intervention effect evaluation module in the embodiment includes: S21: collecting historical tumor ultrasonic images using the historical tumor image data, thereby obtaining historical tumor ultrasonic image data; In an embodiment of the present invention, ultrasonic image acquisition is performed based on historical tumor image data, mainly including retrieval, extraction and formatting of ultrasonic image data, screening of ultrasonic examination data from the historical tumor image database, and extraction of qualified ultrasonic images through a time screening mechanism. The screening conditions include the examination time of the ultrasonic image, the location of the tumor, the parameters of the ultrasonic equipment, etc. After the screening is completed, the original ultrasonic data is parsed in DICOM format using an ultrasonic image data parsing tool, and key image information such as grayscale value, echo intensity, tissue boundary, etc. is extracted. After denoising and contrast enhancement processing, the image data is stored in the historical tumor ultrasonic image database, and image indexing is performed to obtain historical tumor ultrasonic image data.

[0041] S22: performing historical tumor structural feature analysis based on historical tumor ultrasonic image data, thereby obtaining historical tumor structural feature data; In an embodiment of the present invention, based on historical tumor ultrasonic image data, the tumor structural features in the ultrasonic image are analyzed to obtain complete tumor structural feature data, and an ultrasonic image segmentation algorithm is used to extract the tumor tissue area through threshold segmentation and region growing methods, and the boundary contour is calculated. Subsequently, texture feature analysis is performed, and the gray level co-occurrence matrix (GLCM) is used to extract uniformity, contrast, energy and other indicators in the image, and the acoustic characteristics of the tumor tissue are analyzed in combination with the echo intensity characteristics unique to ultrasound. For blood flow-related ultrasound images, color Doppler analysis methods are used to extract parameters such as blood flow velocity and vascular density. After the feature data is extracted, feature vectorization processing is used to convert the ultrasound image features into structured data.

[0042] S23: Use the tumor nursing field data model to formulate personalized nursing care for historical tumor structural feature data, thereby obtaining personalized tumor nursing data; In an embodiment of the present invention, a data model in the field of tumor nursing is used to formulate personalized nursing for historical tumor structural feature data to generate nursing data for specific nursing needs. Based on the structural feature data, the imaging indicators of the tumor are classified and mapped to different nursing categories. For example, for high-density lesion areas shown by imaging features, the corresponding nursing measures data are associated and the execution records of the nursing path are matched. Subsequently, based on the historical data of the nursing path, the data of similar cases are extracted, and the nursing change process records corresponding to different nursing measures are compared. A data matching method is used to compare the similarity between individual nursing data and historical data, and the nursing intervention is adjusted based on the matching degree. The nursing formulation results are stored in a structured manner in a personalized nursing database, and nursing intervention data is generated.

[0043] S24: Evaluate the effectiveness of nursing interventions based on personalized tumor care data and historical tumor structural characteristic data to obtain tumor nursing intervention effectiveness data.

[0044] In an embodiment of the present invention, the effect of nursing intervention is evaluated based on personalized nursing data and historical tumor structure feature data to obtain nursing intervention effect data. The evaluation process compares the characteristic changes of image data before and after nursing, such as whether the tumor boundary is clear, whether the blood flow parameters are stable, etc. The comparison method uses image registration technology to perform pixel-level alignment on ultrasound images before and after nursing, and calculate the changes in the tumor area. Subsequently, combined with the nursing path progress data, the correlation between the nursing execution process and the image changes is analyzed. For example, for specific nursing intervention measures, the time point of nursing implementation is compared with the time point of image feature change, and the time delay of nursing effect is calculated. For abnormal data that occurs during the nursing process, an anomaly detection algorithm is used for analysis to identify nursing execution deviations. The nursing intervention effect data is stored in a time series format and used for optimization of the nursing data model and adjustment of the nursing path.

[0045] Preferably, the historical tumor structural feature analysis based on historical tumor ultrasonic image data includes: Performing ultrasound image contrast enhancement processing on historical tumor ultrasound image data, thereby obtaining tumor ultrasound image contrast enhanced data; Calculating the roughness of the tumor boundary of the ultrasound image according to the contrast enhancement data of the tumor ultrasound image to obtain the roughness data of the tumor boundary; Using the tumor boundary roughness data to perform ultrasound image tumor edge feature analysis, thereby obtaining ultrasound image tumor edge feature data; Calculate the external morphological features of the tumor based on the tumor edge feature data and tumor boundary roughness data of the ultrasound image to obtain the external morphological feature data of the tumor; Performing tumor texture feature analysis based on tumor ultrasonic image contrast enhancement data to obtain tumor texture feature data; Historical tumor structure feature data is analyzed based on tumor texture feature data and tumor external morphology feature data to obtain historical tumor structure feature data.

[0046] In an embodiment of the present invention, ultrasound image contrast enhancement processing is performed on historical tumor ultrasound image data to obtain clearer image data, and the historical ultrasound image data is read and converted into a grayscale image format. The pixel distribution of the image is adjusted by using the histogram equalization method to make the low-contrast image details more prominent. Subsequently, the adaptive histogram equalization (CLAHE, Contrast Limited AdaptiveHistogram Equalization) technology is used to locally enhance the ultrasound image to avoid information loss caused by overall over-enhancement. For ultrasound images with large noise, Gaussian filtering or non-local mean filtering is used for denoising to reduce the influence of artifacts. The enhanced ultrasound image is stored in a database with the index information of the original image. Based on the contrast enhancement data of the tumor ultrasound image, the boundary roughness of the tumor area is calculated to quantify the boundary features, and the tumor boundary is extracted using an image segmentation method. The tumor area is segmented using an active contour model (Active Contour Model, ACM) or a region growing algorithm (Region Growing) to obtain a boundary contour. Subsequently, the fractal dimension is calculated based on the contour curve to evaluate the complexity of the boundary. Furthermore, the boundary roughness is calculated by the boundary gradient change rate, including boundary curvature analysis and Fourier transform spectrum analysis to identify the subtle undulations of the boundary. The calculated boundary roughness data is stored in the database and associated with the ultrasound image index. Based on the tumor boundary roughness data, the tumor edge features in the ultrasound image are deeply analyzed, and the edge of the tumor area in the ultrasound image is extracted using the Canny Edge Detection algorithm to identify the main contour features of the boundary. Then, the coherence of the edge structure is analyzed by the morphological gradient operation, and the integrity parameters of the edge, such as the degree of edge closure and edge smoothness, are calculated. For complex edge morphology, the Gabor filter is used to perform edge texture analysis to extract the directional characteristics of the tumor edge.After edge feature data is extracted, it is stored in a matrix form and matched with the tumor boundary roughness data. Based on the ultrasound image tumor edge feature data and tumor boundary roughness data, the external morphological features of the tumor are calculated to obtain the overall morphological parameters of the tumor. The aspect ratio of the tumor is calculated using the elliptical fitting method to evaluate the regularity of its morphology. Subsequently, parameters such as roundness and irregularity of the tumor are calculated, including calculation of the boundary perimeter and area ratio, and moment invariant analysis. For specific morphological features, such as lobulated and spiculated, a deep learning edge feature extraction network (such as U-Net) is used for pattern recognition. Based on the contrast enhancement data of tumor ultrasound images, the texture features inside the tumor are analyzed to extract the information of tissue structure. The texture parameters of ultrasound images, including energy, contrast, entropy, etc., are calculated using the Gray-Level Co-occurrence Matrix (GLCM) to quantify the fineness of tumor tissue. Subsequently, the local binary pattern (LBP) is used to analyze the local texture changes of the image to evaluate the uniformity of the tissue. For cases with strong internal heterogeneity of the tumor, wavelet transform is used for multi-scale feature extraction to identify the distribution of high-frequency and low-frequency information. Based on the tumor texture feature data and the external morphological feature data of the tumor, the historical tumor structure features are comprehensively analyzed to obtain complete tumor structure feature data, and the texture feature data is compared and analyzed with the morphological feature data. The principal component analysis (PCA) method is used to reduce the dimensionality of the high-dimensional feature data to extract the most representative feature combination. Subsequently, clustering algorithms (such as K-means or hierarchical clustering) were used to classify different feature combinations to identify different types of tumor tissue structure patterns. For time series data, dynamic time warping (DTW) was used to analyze the trend of historical tumor images and calculate the feature changes before and after nursing intervention.

[0047] Preferably, the tumor texture feature analysis based on the tumor ultrasound image contrast enhancement data includes the following: Calculate the local gradient of the ultrasound image tumor according to the contrast enhancement data of the ultrasound image of the tumor to obtain the local gradient data of the ultrasound image tumor; Decomposing the multi-scale features of the tumor according to the local gradient data of the tumor in the ultrasound image, thereby obtaining the multi-scale feature data of the tumor; Based on the multi-scale feature data of the tumor and the local gradient data of the tumor in the ultrasound image, the hierarchical structure of the tumor texture is detected to obtain the hierarchical texture structure data of the tumor; The tumor local texture change rate is calculated using the tumor layered texture structure data, thereby obtaining the tumor local texture change rate data; Tumor texture feature analysis is performed based on the tumor local texture change rate data and the tumor hierarchical texture structure data to obtain tumor texture feature data.

[0048] In the embodiment of the present invention, in the process of calculating the local gradient of the ultrasound image tumor, grayscale normalization processing is performed on the contrast enhancement data of the tumor ultrasound image to make the pixel value distribution more uniform, so as to reduce the influence of the brightness difference between different image data on the subsequent calculation. Then, the processed ultrasound image is subjected to edge feature extraction, and the image gradient calculation method is adopted to obtain the local gradient information of the tumor area in the ultrasound image by detecting the brightness change around the pixel point. Specifically, the edge detection operator is used to calculate the gradient of the image in the horizontal and vertical directions to obtain the local grayscale change characteristics of the tumor tissue boundary. After obtaining the local gradient data, the gradient distribution is statistically analyzed to extract the numerical information that can reflect the boundary characteristics of the tumor tissue. Based on the local gradient data of the ultrasound image tumor, the multi-scale feature decomposition of the tumor area is performed to extract the texture feature information at different spatial scales, and the ultrasound image is subjected to multi-level fuzzy segmentation. The multi-level image structure is constructed using image data of different resolutions to ensure that the local and overall features of the tumor tissue can be obtained at different scales. Then, the texture feature decomposition method is adopted to split the image data according to different frequency components, and the overall tissue structure information of the low-frequency part and the detailed edge features of the high-frequency part are extracted. Further, the decomposed image data is processed into sub-regions, and the texture distribution characteristics of each sub-region are calculated, including parameters such as brightness change, edge density and regional uniformity. After completing the multi-scale feature decomposition, the feature data at different scales are stored in the database and associated with the local gradient data of the tumor. Based on the multi-scale feature data of the tumor and the local gradient data of the ultrasound image tumor, the texture hierarchical structure detection of the tumor area is performed to identify the texture changes of the tumor tissue at different depth levels, and the ultrasound image is hierarchically segmented. By setting analysis windows of different depths, the texture information of different levels from shallow to deep in the ultrasound image is extracted. Then, the texture hierarchical detection method is used to calculate the uniformity, contrast and energy distribution of each level respectively to analyze the texture feature performance of the tumor tissue at each level. Further, the texture distribution of different levels is compared and analyzed to identify the differences in texture features between the levels, and the texture change trend of the tumor tissue at different depth levels is judged in combination with the local gradient changes. Using the hierarchical texture structure data of the tumor, the local texture change rate of different layers of the tumor is calculated to analyze the texture change amplitude of the tumor tissue at different depths. The different layers of the ultrasound image are pixel-segmented, the different depth layers of the tumor tissue are extracted, and the texture features of each layer are refined and calculated. Then, the texture feature differences between adjacent layers are compared, the grayscale change degree, edge sharpness change, and texture uniformity change amplitude of each layer are statistically analyzed, and the local texture change rate of the tumor tissue at different depths is calculated. Furthermore, the calculated local texture change rate data is subjected to time series analysis, and the change rate data at multiple time points are compared to analyze the dynamic evolution of the tumor tissue.Based on the local texture change rate data of the tumor and the hierarchical texture structure data of the tumor, the overall texture features of the tumor tissue are comprehensively analyzed to obtain complete tumor texture feature data, and the texture feature data at different levels are fused, and the representative texture feature parameters are extracted using the data dimensionality reduction method. Then, the fused texture feature data are classified and analyzed to identify the texture change pattern of the tumor tissue at different stages of care, and its change trend is summarized. Furthermore, the current texture feature data is compared with the historical texture feature data using the data matching method to analyze the long-term texture evolution of the tumor tissue.

[0049] Preferably, the method of using the tumor nursing field data model to formulate personalized nursing care for historical tumor structural feature data includes: Calculating the tumor structure volume according to historical tumor structure characteristic data, thereby obtaining tumor structure volume data; The degree of tumor vascularization is detected according to historical tumor structural characteristic data, thereby obtaining the data of the degree of tumor vascularization; Tumor invasiveness is estimated based on tumor vascularization degree data and tumor structure volume data, thereby obtaining tumor invasiveness data; Analyze the abnormal state of tumor structure according to the tumor invasion data and the tumor vascularization degree data, so as to obtain the abnormal state of tumor structure; The tumor nursing field data model is used to formulate personalized nursing care for tumor structural abnormality data, thereby obtaining personalized tumor nursing data.

[0050] In an embodiment of the present invention, when calculating the volume of the tumor structure, based on the historical tumor structure feature data, the tumor image data at continuous time points are screened out, and the tumor area is reconstructed in three dimensions to ensure the accuracy of the volume calculation. Then, the medical image segmentation method is used to extract the contour of the tumor area in the ultrasound image, CT image or MRI image to exclude the interference of the non-target area. Next, the voxel accumulation method is used to calculate the three-dimensional spatial volume of the tumor, that is, by counting the number of voxels marked as tumor tissues, and combining the spatial resolution parameters of the image, the overall structural volume data of the tumor is obtained. Based on the historical tumor structure feature data, the degree of tumor vascularization is detected to obtain the vascular distribution of the tumor area, and the vascular structure inside and around the tumor is extracted using Doppler ultrasound images or contrast imaging data, and the blood flow signal is enhanced to more clearly observe the vascular network. Then, the vascular segmentation method is used to separate the vascular area in the image, the vascular density in the tumor area is counted, and the key parameters such as the blood flow velocity and the number of vascular branches are calculated. Further, the vascularization degree data at different time points are compared to analyze the change trend of the blood supply of the tumor tissue. When estimating tumor invasiveness, the growth pattern of tumor tissue is analyzed based on the data of tumor vascularization degree and tumor structure volume data, the changes in tumor volume at consecutive time points are compared, the volume growth rate of the tumor is calculated, and whether there is an accelerated growth trend is evaluated. Combined with the data of vascularization degree, whether the blood supply inside the tumor is abnormally enhanced is analyzed, and whether angiogenesis is positively correlated with the structural changes of the tumor is judged. Further, based on the data of tumor edge characteristics, whether the tumor tissue has irregular diffusion characteristics, such as blurred boundaries, local invasion, etc. is evaluated. According to the data of tumor invasiveness and tumor vascularization degree, the abnormal state of tumor structure is analyzed to identify the morphological abnormal characteristics of tumor tissue, extract the boundary morphological data of the tumor, compare the boundary changes at different time points, and judge whether there is a structural mutation phenomenon, such as local hyperplasia of the tumor edge or the appearance of new protrusions. Then, combined with the data of vascularization degree, whether the distribution of vascular network has abnormal expansion is analyzed, and whether the tumor tissue has a trend of rapid vascularization is evaluated. Further, combined with the image texture feature data, whether the internal structure of the tumor tissue has heterogeneous changes is analyzed, such as the density of some areas is significantly different from that of other parts. When formulating personalized nursing care, we analyze the effects of different nursing measures on tumor tissue characteristics based on the data of abnormal tumor structure, and screen nursing measures that match the current tumor structure characteristics in combination with historical data of nursing interventions, and compare the changes in tumor tissue under different nursing methods. Then, we formulate corresponding nursing adjustments for the specific manifestations of abnormal tumor structure, such as excessive volume growth and abnormal vascularization, and evaluate the feasibility of nursing measures.

[0051] Preferably, the tumor nursing risk estimation module includes the following functions: According to the tumor nursing intervention effect data, the abnormal effect of tumor nursing intervention is estimated, so as to obtain the abnormal effect data of tumor nursing intervention; Estimating tumor nursing risk based on abnormal effect data of tumor nursing intervention, thus obtaining tumor nursing risk data; Based on the abnormal effect data of tumor nursing intervention and tumor nursing risk data, tumor nursing side effects are evaluated to obtain tumor nursing side effects data; The probability of tumor nursing complications is calculated based on the tumor nursing side effect data and tumor nursing risk data to obtain the tumor nursing complication probability data.

[0052] As an example of the present invention, refer to Figure 3 As shown, Figure 1 The functional flow chart of the tumor nursing risk estimation module in the embodiment includes: S31: estimating abnormal effects of tumor nursing intervention according to the tumor nursing intervention effect data, thereby obtaining abnormal effect data of tumor nursing intervention; In an embodiment of the present invention, when estimating the abnormal effect of tumor nursing intervention, the image data, vital sign records, nursing records and other related information after the nursing intervention are screened out based on the tumor nursing intervention effect data. Then, by using the time series analysis method, the tumor image data at different time points are compared to observe the change trend of the tumor tissue, and the changes in the tissue morphology, degree of vascularization and image texture characteristics after the implementation of the nursing measures are analyzed in combination with the nursing measures records. Further, a data deviation detection method is used to compare the amplitude of the data changes before and after the nursing intervention, and calculate whether there is an abnormal nursing intervention effect. For example, if the tissue morphology changes after the nursing intervention exceed the normal range of the historical nursing data, it is marked as an abnormal nursing intervention effect and classified into a specific abnormal category, such as enhanced tissue reaction, aggravated local edema, abnormally enhanced blood supply, etc.

[0053] S32: Estimating tumor nursing risk based on abnormal effect data of tumor nursing intervention, thereby obtaining tumor nursing risk data; In an embodiment of the present invention, tumor nursing risk estimation is performed based on abnormal effect data of tumor nursing intervention to identify risk factors caused by nursing intervention, extract abnormal nursing intervention effect data, analyze the incidence, duration and severity of abnormal phenomena, and compare them with historical nursing data to evaluate whether there are specific risk trends in current nursing measures. Then, according to the category of abnormal nursing effects, combined with imaging data, analyze whether there is an accumulation of risk factors, such as whether local edema continues to worsen, whether tissue necrosis expands, whether vascular hyperplasia is overactive, etc. Further, the nursing risk attribution analysis method is used to evaluate the correlation between specific nursing measures and abnormal effects to determine high-risk nursing intervention factors.

[0054] S33: Evaluate the side effects of tumor nursing based on the abnormal effect data of tumor nursing intervention and the risk data of tumor nursing, so as to obtain the side effects data of tumor nursing; In an embodiment of the present invention, when evaluating the side effects of tumor nursing, based on the abnormal effect data of tumor nursing intervention and the tumor nursing risk data, the adverse reactions caused by the nursing measures are analyzed, the nursing risk data is extracted, the abnormal effect categories that are highly correlated with the nursing measures are screened out, and the probability of side effects is calculated in combination with historical data. Then, for different abnormal effects, their effects on tumor tissue and surrounding normal tissue are analyzed, such as whether the nursing intervention causes tissue damage, vascular blockage, local inflammation, etc. Further, the distribution of side effects under different nursing measures is compared to evaluate whether a specific nursing method has a higher risk of side effects.

[0055] S34: Calculate the probability of tumor nursing complications based on the tumor nursing side effect data and the tumor nursing risk data, so as to obtain the tumor nursing complication probability data.

[0056] In an embodiment of the present invention, the probability of occurrence of tumor nursing complications is calculated based on tumor nursing side effect data and tumor nursing risk data to evaluate the long-term impact of nursing interventions. Based on the side effect data, abnormal nursing interventions that lead to complications are screened out, and the development trend of these abnormalities is analyzed. Then, combined with the nursing risk data, the potential triggering factors of complications under specific nursing measures are evaluated, and the probability of occurrence of specific complications is calculated, such as local infection, tissue fibrosis, chronic inflammatory response, etc. Further, using long-term nursing data, the incidence of complications of different nursing care is compared to evaluate whether there is a long-term risk for a specific nursing method.

[0057] Preferably, the tumor nursing risk estimation based on abnormal effect data of tumor nursing intervention includes: Based on the abnormal effect data of tumor nursing intervention, nonlinear change analysis of tumor nursing effect was performed to obtain nonlinear change data of tumor nursing effect; According to the nonlinear change data of tumor nursing effect, the normal distribution imbalance of tumor nursing is calculated to obtain the normal distribution imbalance data of tumor nursing; Tumor nursing deviation is calculated based on the imbalanced data of tumor nursing normal distribution and the nonlinear change data of tumor nursing effect, so as to obtain tumor nursing deviation data; Based on the abnormal effect data of tumor nursing intervention, abnormal influencing factors of tumor nursing were analyzed to obtain abnormal influencing factors of tumor nursing; Tumor nursing risk estimation is performed based on abnormal tumor nursing influencing factor data and tumor nursing deviation data to obtain tumor nursing risk data.

[0058] In an embodiment of the present invention, when performing nonlinear change analysis of tumor nursing effect, abnormal effect data of tumor nursing intervention are extracted, and these data include tumor imaging changes, vital sign fluctuations, and nursing intervention records after nursing intervention. Next, a time series analysis method is used to select data points of multiple nursing cycles to detect changes in the volume, structure, texture and other characteristics of the tumor. By comparing the data of each time node, the nonlinear change pattern of the tumor nursing effect is identified. Specifically, a nonlinear regression analysis method is used to fit the change trend of the nursing effect with the growth pattern of the tumor, and identify which nursing measures lead to the acceleration or slowdown of tumor changes. This process does not rely on traditional linear models, but accurately captures the nonlinear change characteristics of nursing effects at different stages through a more flexible nonlinear modeling method, and generates nonlinear change data of tumor nursing effects. According to the nonlinear change data of tumor nursing effects, the data distribution is standardized to ensure that the data meets the basic conditions of normal distribution. Then, the skewness and kurtosis of the nursing effect data are analyzed, and the data statistical method is used to check whether it deviates from the normal distribution. If it is found that there is an obvious imbalance in the distribution of the data, such as most of the data is concentrated on one side, or the distribution form is too skewed, it is judged that the normal distribution is unbalanced. The degree of imbalance in the nursing effect data is quantified by using the normal distribution imbalance calculation method and recorded in the data model. This process uses statistical methods to determine whether the nursing effect data presents an asymmetric distribution or a skewed distribution, and calculates its degree of deviation, such as the skewness and kurtosis of the normal distribution. Finally, the tumor nursing normal distribution imbalance data is generated. Through the tumor nursing normal distribution imbalance data and the tumor nursing effect nonlinear change data, the tumor nursing deviation calculation is performed to analyze the overall trend of the nursing effect data and judge whether the results of the nursing intervention meet expectations. If the nursing effect fluctuates greatly or deviates from the expected effect at a specific time point, the deviation of the data is calculated based on the normal distribution imbalance data. The deviation is expressed by measuring the distance between the data point and the average value, and further combined with the nonlinear change data to measure whether the deviation belongs to the normal fluctuation range of the nursing intervention. Through these analyses, it is possible to identify which nursing intervention measures lead to a large deviation between the nursing effect and the expected target, and calculate the deviation value, thereby generating tumor nursing deviation data. In the analysis of abnormal influencing factors of tumor nursing, the key factors affecting the nursing effect are extracted from the abnormal effect data of tumor nursing intervention. These factors include the implementation process of nursing measures, tumor type, tumor location, etc. Through data mining methods, the factors that have the greatest impact on nursing effects are identified. Specific analysis methods include cluster analysis, factor analysis, etc., which can effectively extract potential factors that affect tumor nursing effects. Then, through multidimensional analysis, the specific degree of influence of these factors on nursing effects is determined. For example, some nursing measures have a greater impact on specific types of tumors, but less on other types.After the analysis is completed, the data of abnormal influencing factors of tumor nursing are obtained. When estimating the risk of tumor nursing, the data of abnormal influencing factors of tumor nursing are combined with the data of tumor nursing deviation to identify the links that cause risks in the nursing intervention process. By comprehensively analyzing the mechanism of action of abnormal influencing factors and the deviation of nursing effects, it is determined which nursing intervention measures have higher risks. For example, some nursing measures lead to significant deviations in nursing effects due to differences, which in turn cause adverse reactions. Through the risk assessment model, combined with the deviation calculation results, the risks of different nursing measures are quantified, and their potential threats are evaluated, tumor nursing risk data is generated, and the risk level of each nursing intervention is recorded.

[0059] Preferably, the model optimization module includes the following functions: Based on the probability data of tumor nursing complications and the risk data of tumor nursing, personalized nursing anomaly detection is performed on the personalized tumor nursing data to obtain the abnormal data of personalized tumor nursing; Adjusting the personalized tumor care data according to the abnormal personalized tumor care data to obtain personalized tumor care adjustment data; The data model in the field of tumor nursing is optimized according to the abnormal data of personalized tumor nursing and the adjusted data of personalized tumor nursing, so as to obtain the data optimization model in the field of tumor nursing.

[0060] In an embodiment of the present invention, when performing abnormal detection of personalized tumor nursing, the probability of potential complications in the process of receiving nursing is extracted from the probability data of tumor nursing complications, including but not limited to fluctuations in physical signs, concurrent infections, etc., and combined with tumor nursing risk data, the risk factors affecting the effect of nursing intervention are analyzed, such as the type of drugs used in the nursing intervention process, the intensity of nursing means, changes in the nursing process, etc. The personalized tumor nursing data are monitored and analyzed using statistical analysis and outlier detection methods. By comparing the nursing data with the preset threshold of the standard nursing model, abnormal data points that exceed the normal range are identified. For example, if abnormal physical signs change during the nursing process, it is determined to be nursing abnormality. By analyzing the nursing records one by one, the abnormal data of personalized tumor nursing are obtained. After the abnormal data of personalized tumor nursing is obtained, the next step is to adjust the personalized tumor nursing data according to these abnormal data, and combine the historical data of nursing intervention and the abnormal detection results to identify the specific factors that cause the abnormality, such as the state of inappropriate intensity of certain nursing measures, or the mismatch between nursing means and health conditions. Then, the abnormal data is corrected according to the nursing adjustment. For example, if a nursing measure is found to lead to an abnormal nursing response, the nursing method is adjusted according to the specific situation, such as reducing the intensity of the nursing measure, adjusting the nursing interval, etc. Using data mining and regression analysis techniques, combined with the setting of personalized nursing, the nursing is re-drafted. In this process, it is also necessary to simulate and verify the adjusted nursing data to ensure that the adjusted nursing measures have a more balanced and controllable impact. After the adjustment process is completed, the personalized nursing adjustment data for tumors is generated. After the abnormal data and the nursing adjustment data for personalized nursing for tumors are generated, the data model in the field of tumor nursing is optimized. The existing data model in the field of tumor nursing is evaluated using data analysis technology to determine the shortcomings of the current model, especially the adaptability problems when facing abnormal nursing situations and adjusting data. Combined with abnormal data, analyze whether the current model can effectively reflect the changes caused by differences in the nursing intervention process, and adjust some parameters in the model. For example, by introducing more personalized features, the input variables of the model are expanded. Then, the optimization algorithm, such as genetic algorithm or particle swarm optimization algorithm, is used to iteratively optimize the model and update the weight parameters in the model to improve its prediction accuracy and adaptability. Through these optimizations, we ensure that the data model can better cope with the complex changes in the tumor nursing process, improve the predictive ability of nursing intervention effects, and obtain a data optimization model in the field of tumor nursing. This optimization model can more accurately reflect the relationship between multiple variables in the tumor nursing process.

[0061] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. An analysis system based on a data model in the field of tumor nursing, characterized in that: Includes the following modules: A tumor nursing field data model construction module is used to obtain historical tumor nursing field data; collect historical tumor images based on historical tumor nursing field data to obtain historical tumor image data; construct a tumor nursing field data model based on historical tumor image data and historical tumor nursing field data to obtain a tumor nursing field data model; The nursing intervention effect evaluation module is used to analyze the historical tumor structure characteristics using historical tumor imaging data, thereby obtaining historical tumor structure characteristic data; to formulate personalized nursing for the historical tumor structure characteristic data using the tumor nursing field data model, thereby obtaining personalized tumor nursing data; to evaluate the nursing intervention effect based on the personalized tumor nursing data and historical tumor structure characteristic data, thereby obtaining tumor nursing intervention effect data; The tumor nursing risk estimation module is used to estimate the tumor nursing risk based on the tumor nursing intervention effect data, thereby obtaining tumor nursing risk data; calculate the probability of tumor nursing complications based on the tumor nursing risk data, thereby obtaining tumor nursing complication probability data; The model optimization module is used to perform personalized nursing anomaly detection on tumor personalized nursing data based on tumor nursing complication probability data and tumor nursing risk data to obtain tumor personalized nursing anomaly data; and to optimize the data model in the tumor nursing field according to the tumor personalized nursing anomaly data to obtain the tumor nursing field data optimization model.

2. The analysis system based on the tumor nursing field data model according to claim 1 is characterized in that: The tumor nursing domain data model construction module specifically performs the following steps: Access to historical oncology nursing field data; Perform historical tumor nursing desensitization processing based on historical tumor nursing field data, thereby obtaining historical tumor nursing field desensitization data; Use historical desensitized data in the field of oncology nursing to collect pathways in the field of oncology nursing, thereby obtaining oncology nursing pathway data; The historical tumor physical signs and conditions are collected based on the desensitized data in the field of historical tumor nursing, thereby obtaining the historical tumor physical signs and conditions data; Collect historical tumor images based on the desensitized data in the field of historical tumor nursing, thereby obtaining historical tumor image data; Based on historical tumor sign status data, historical tumor imaging data and tumor nursing pathway data, a data model for the tumor nursing field is constructed to obtain a data model for the tumor nursing field.

3. The analysis system based on the tumor nursing field data model according to claim 2 is characterized in that: The following steps are performed to build a data model in the field of tumor nursing based on historical tumor signs and conditions data, historical tumor imaging data, and tumor nursing pathway data: Calculating the progress of the tumor nursing pathway according to the tumor nursing pathway data, thereby obtaining the progress data of the tumor nursing pathway; According to the progress data of tumor nursing pathway, the tumor nursing change process is recorded for the historical tumor physical sign status data, so as to obtain the tumor nursing change process record data; Based on the tumor nursing change process record data and tumor nursing pathway progress data, the tumor nursing change law analysis is performed on the historical tumor imaging data to obtain the historical tumor nursing change law data; A data model for the field of tumor nursing is constructed based on the historical data on the changing rules of tumor nursing and the data on the recording of the changing process of tumor nursing, and a data model for the field of tumor nursing is obtained.

4. The analysis system based on the tumor nursing field data model according to claim 1 is characterized in that: The nursing intervention effect evaluation module specifically performs the following steps: Using historical tumor imaging data to collect historical tumor ultrasonic images, thereby obtaining historical tumor ultrasonic image data; Perform historical tumor structural feature analysis based on historical tumor ultrasonic image data, thereby obtaining historical tumor structural feature data; Use the tumor nursing field data model to formulate personalized nursing care for historical tumor structural feature data, thereby obtaining personalized tumor nursing data; The effectiveness of nursing intervention is evaluated based on personalized tumor nursing data and historical tumor structural characteristic data to obtain tumor nursing intervention effectiveness data.

5. The analysis system based on the tumor nursing field data model according to claim 4 is characterized in that: The following steps are performed to analyze the structural characteristics of historical tumors based on historical tumor ultrasound image data: Performing ultrasound image contrast enhancement processing on historical tumor ultrasound image data, thereby obtaining tumor ultrasound image contrast enhanced data; Calculating the roughness of the tumor boundary of the ultrasound image according to the contrast enhancement data of the tumor ultrasound image to obtain the roughness data of the tumor boundary; Using the tumor boundary roughness data to perform ultrasound image tumor edge feature analysis, thereby obtaining ultrasound image tumor edge feature data; Calculate the external morphological features of the tumor based on the tumor edge feature data and tumor boundary roughness data of the ultrasound image to obtain the external morphological feature data of the tumor; Performing tumor texture feature analysis based on tumor ultrasonic image contrast enhancement data to obtain tumor texture feature data; Historical tumor structure feature data is analyzed based on tumor texture feature data and tumor external morphology feature data to obtain historical tumor structure feature data.

6. The analysis system based on the tumor nursing field data model according to claim 5 is characterized in that: The following steps are performed to analyze the tumor texture features based on the contrast enhancement data of the tumor ultrasound image: Calculate the local gradient of the ultrasound image tumor according to the contrast enhancement data of the ultrasound image of the tumor to obtain the local gradient data of the ultrasound image tumor; Decomposing the multi-scale features of the tumor according to the local gradient data of the tumor in the ultrasound image, thereby obtaining the multi-scale feature data of the tumor; Based on the multi-scale feature data of the tumor and the local gradient data of the tumor in the ultrasound image, the hierarchical structure of the tumor texture is detected to obtain the hierarchical texture structure data of the tumor; The tumor local texture change rate is calculated using the tumor layered texture structure data, thereby obtaining the tumor local texture change rate data; Tumor texture feature analysis is performed based on the tumor local texture change rate data and the tumor hierarchical texture structure data to obtain tumor texture feature data.

7. The analysis system based on the tumor nursing field data model according to claim 4 is characterized in that: The following steps are performed to formulate personalized nursing care based on historical tumor structure feature data using the tumor nursing data model: Calculating the tumor structure volume according to historical tumor structure characteristic data, thereby obtaining tumor structure volume data; The degree of tumor vascularization is detected according to historical tumor structural characteristic data, thereby obtaining the data of the degree of tumor vascularization; Tumor invasiveness is estimated based on tumor vascularization degree data and tumor structure volume data, thereby obtaining tumor invasiveness data; Analyze the abnormal state of tumor structure according to the tumor invasion data and the tumor vascularization degree data, so as to obtain the abnormal state of tumor structure; The tumor nursing field data model is used to formulate personalized nursing care for tumor structural abnormality data, thereby obtaining personalized tumor nursing data.

8. The analysis system based on the tumor nursing field data model according to claim 1, characterized in that: The cancer nursing risk estimation module specifically performs the following steps: According to the tumor nursing intervention effect data, the abnormal effect of tumor nursing intervention is estimated, so as to obtain the abnormal effect data of tumor nursing intervention; Estimating tumor nursing risk based on abnormal effect data of tumor nursing intervention, thus obtaining tumor nursing risk data; Based on the abnormal effect data of tumor nursing intervention and tumor nursing risk data, tumor nursing side effects are evaluated to obtain tumor nursing side effects data; The probability of tumor nursing complications is calculated based on the tumor nursing side effect data and tumor nursing risk data to obtain the tumor nursing complication probability data.

9. The analysis system based on the tumor nursing field data model according to claim 8, characterized in that: The following steps are performed to estimate the risk of cancer nursing based on the abnormal effect data of cancer nursing intervention: Based on the abnormal effect data of tumor nursing intervention, nonlinear change analysis of tumor nursing effect was performed to obtain nonlinear change data of tumor nursing effect; According to the nonlinear change data of tumor nursing effect, the normal distribution imbalance of tumor nursing is calculated to obtain the normal distribution imbalance data of tumor nursing; Tumor nursing deviation is calculated based on the imbalanced data of tumor nursing normal distribution and the nonlinear change data of tumor nursing effect, so as to obtain tumor nursing deviation data; Based on the abnormal effect data of tumor nursing intervention, abnormal influencing factors of tumor nursing were analyzed to obtain abnormal influencing factors of tumor nursing; Tumor nursing risk estimation is performed based on abnormal tumor nursing influencing factor data and tumor nursing deviation data to obtain tumor nursing risk data.

10. The analysis system based on the tumor nursing field data model according to claim 7, characterized in that: The model optimization module specifically performs the following steps: Based on the probability data of tumor nursing complications and the risk data of tumor nursing, personalized nursing anomaly detection is performed on the personalized tumor nursing data to obtain the abnormal data of personalized tumor nursing; Adjusting the personalized tumor care data according to the abnormal personalized tumor care data to obtain personalized tumor care adjustment data; The data model in the field of tumor nursing is optimized according to the abnormal data of personalized tumor nursing and the adjusted data of personalized tumor nursing, so as to obtain the data optimization model in the field of tumor nursing.

Images