Patient nursing grade intelligent evaluation system based on high-dimensional tumor data

Abstract

The invention discloses a patient nursing grade intelligent evaluation system based on high-dimensional tumor data, and relates to the technical field of nursing grade evaluation. The nursing grade of the patient is intelligently evaluated through a comprehensive evaluation algorithm. The AHP can help the system to carry out hierarchical processing on factors influencing the nursing grade, and the entropy weight rule carries out objective optimization on the weight of each evaluation factor, so that a comprehensive nursing grade evaluation result is generated. The process is automatically completed, manual intervention is not needed, and the evaluation efficiency and precision are greatly improved.

Description

Technical Field

[0001] The present invention belongs to the field of nursing level assessment, and in particular, relates to an intelligent patient nursing level assessment system based on high-dimensional tumor data. Background Art

[0002] In the current medical care field, the assessment of a patient's care level is a complex process that usually requires consideration of multiple factors, such as the patient's health status, tumor characteristics, mobility, etc. Traditional care level assessment methods often rely on manual judgment, are highly subjective and unstable, and lack real-time and dynamic adaptability. With the development of data science, more and more intelligent assessment systems have begun to be introduced into care level assessment, in order to improve the accuracy and consistency of assessment through data-driven methods.

[0003] Traditional nursing level assessment methods usually rely on expert experience or simple weighted average methods, which leads to several major problems: first, poor data integration, it is difficult to effectively integrate information from different sources (such as tumor data, activity data, etc.), resulting in one-sided or incomplete assessment results; second, the lack of dynamic adjustment mechanism, unable to make rapid adjustments according to the real-time changes in the patient's health status, resulting in delayed assessment; finally, over-reliance on expert judgment, easily affected by subjective factors, resulting in inconsistent and inaccurate assessment results. Therefore, traditional methods are difficult to adapt to complex patient health conditions and cannot provide accurate, real-time, and standardized nursing level assessments.

[0004] In view of this, the present invention is proposed. Summary of the invention

[0005] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and provide a patient care level intelligent assessment system based on high-dimensional tumor data, thereby solving the problems raised in the above-mentioned background technology.

[0006] In order to solve the above technical problems, the basic concept of the technical solution adopted by the present invention is:

[0007] An intelligent patient care level assessment system based on high-dimensional tumor data includes: a data acquisition module for collecting high-dimensional tumor data and activity data of patients;

[0008] A feature extraction module, used to filter out features related to nursing level assessment from the high-dimensional tumor data and activity data, and perform preliminary processing on the data;

[0009] A nursing level assessment module is used to intelligently assess the patient's nursing level by combining the screened features with a preset comprehensive assessment algorithm. The assessment algorithm hierarchically processes the factors affecting the nursing level through the analytic hierarchy process, and objectively optimizes the weight of each assessment factor based on the entropy weight method to generate a comprehensive nursing level assessment result;

[0010] An assessment result display module is used to display the patient's nursing level assessment results in text form and automatically remind nursing staff when the assessment result is high-risk;

[0011] The data interface module is used to interact with other medical systems or devices to ensure the timeliness and accuracy of data acquisition and transmission.

[0012] Optionally, the steps of screening out features related to nursing level assessment from the high-dimensional tumor data and activity data and performing preliminary processing on the data are:

[0013] Obtain various high-dimensional data related to the patient's tumor, including but not limited to tumor markers, tumor size, stage, pathology results, imaging data, and genomic data;

[0014] After data collection is completed, the data needs to be cleaned and preprocessed, and then the features related to nursing level assessment in high-dimensional tumor data are analyzed to select those features that play an important role in nursing level prediction;

[0015] After feature screening, feature engineering and data transformation are performed. For continuous features, the distribution of data can be adjusted through logarithmic transformation or differential processing. Finally, all processed, screened and transformed features are integrated into a final data set for subsequent care level evaluation.

[0016] Optionally, the patient care level is intelligently evaluated by combining the screened features with a preset comprehensive evaluation algorithm, and the evaluation algorithm hierarchically processes the factors affecting the care level through the hierarchical analysis method as follows:

[0017] According to historical data, an n-order judgment matrix A is constructed, and its expression is: Among them, a ij It represents the relative importance evaluation of factor i and factor j. The matrix is ​​a symmetric matrix, and all elements on the diagonal are 1;

[0018] By solving the eigenvalue and eigenvector of the constructed judgment matrix A through the formula A·ω=λ·ω, the eigenvector of the matrix can be obtained, where λ is the eigenvalue and w is the eigenvector, which represents the weight distribution of each factor. The element w of the eigenvector is i represents the weight of the i-th factor in the decision-making process;

[0019] In order to make the sum of the weight vectors equal to 1, we need to use the formula The obtained eigenvector w is normalized, and the normalized ω=(ω1, ω2, ..., ω n ) is the final weight vector, where each w i represents the relative weight of the i-th factor in the overall decision;

[0020] In order to ensure the consistency of the judgment matrix, the formula A consistency check is required. If CR < 0.1, the consistency of the judgment matrix is ​​considered acceptable, indicating that the weight allocation has high reliability and rationality.

[0021] Optionally, the weights of each evaluation factor are objectively optimized based on the entropy weight method, and the steps for generating the comprehensive nursing level evaluation results are:

[0022] The entropy value E of each indicator based on the information entropy method j Calculate and calculate the information entropy by combining the proportion of multiple data sources through a multi-level information fusion algorithm to optimize the indicator weight. The expression is: Among them, w k is the weighting coefficient of different data sources, p ijk It represents the proportion of the jth indicator in the i-th solution according to the k-th data source; according to the entropy value E j , through the formula Calculate the weight of each indicator Among them, the entropy value E j It is dynamically updated based on real-time monitoring data to improve the real-time adaptability of the assessment model.

[0023] Combining the weights obtained by the hierarchical analysis method and the entropy weight method, the comprehensive weight of each indicator ω is calculated using adaptive weighted regression and weighted average method. j , and the consistency of the judgment matrix is ​​optimized by genetic algorithm. The optimized weighted coefficient is: Where β is the adaptive adjustment coefficient, Adapt(X) is the adaptive adjustment function, which optimizes the weighting coefficient α based on data changes, indicator correlation and expert feedback during the evaluation process;

[0024] The genetic algorithm optimizes the weight distribution in the judgment matrix through crossover and mutation operations to minimize the consistency ratio CR and improve the evaluation accuracy. Then, the evaluation score Sc of the nursing level is obtained through weighted summation, and its calculation formula is:

[0025] Optionally, when the assessment result is high risk, first, a high risk judgment threshold is set based on the comprehensive results of historical assessment data and expert feedback. When the comprehensive score is lower than a certain set threshold, the assessment result is judged to be high risk. When the assessment result is high risk, the system will automatically start the high risk reminder function.

[0026] Optionally, alarm prompts include but are not limited to sound alarms, visual alarms, and vibration reminders to ensure that nursing staff can promptly notice the patient's high-risk assessment results.

[0027] Optionally, historical nursing level assessment data include historical assessment results, the patient's nursing level score and corresponding assessment indicators, real-time data feedback, after each assessment, the nursing staff's feedback on the assessment results based on the patient's actual nursing needs, expert evaluation data, and the expert's revised opinions on the nursing level assessment results based on clinical experience and assessment data.

[0028] After adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art. Of course, any product implementing the present invention does not necessarily need to achieve all the advantages described below at the same time:

[0029] 1. The present invention uses a comprehensive evaluation algorithm to intelligently evaluate the patient's nursing level. AHP can help the system hierarchically process the factors that affect the nursing level, and the entropy weight rule objectively optimizes the weights of each evaluation factor to generate a comprehensive nursing level evaluation result. This process is completed automatically without manual intervention, which greatly improves the efficiency and accuracy of the evaluation.

[0030] The specific implementation modes of the present invention are further described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The drawings described below are only some embodiments. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0032] Figure 1 Block diagram of the system for intelligent assessment of patient care level.

[0033] It should be noted that these drawings and textual descriptions are not intended to limit the conceptual scope of the present invention in any way, but are intended to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0034] The present invention will now be described in further detail with reference to the accompanying drawings.

[0035] See also Figure 1As shown, in this embodiment, a patient care level intelligent assessment system based on high-dimensional tumor data is provided, including a data acquisition module for collecting high-dimensional tumor data and activity data of patients;

[0036] A feature extraction module, used to filter out features related to nursing level assessment from the high-dimensional tumor data and activity data, and perform preliminary processing on the data;

[0037] A nursing level assessment module is used to intelligently assess the patient's nursing level by combining the screened features with a preset comprehensive assessment algorithm. The assessment algorithm hierarchically processes the factors affecting the nursing level through the analytic hierarchy process, and objectively optimizes the weight of each assessment factor based on the entropy weight method to generate a comprehensive nursing level assessment result;

[0038] An assessment result display module is used to display the patient's nursing level assessment results in text form and automatically remind nursing staff when the assessment result is high-risk;

[0039] The data interface module is used to interact with other medical systems or devices to ensure the timeliness and accuracy of data acquisition and transmission.

[0040] In this embodiment, the steps of screening out features related to nursing level assessment from the high-dimensional tumor data and activity data and performing preliminary processing on the data are as follows:

[0041] Obtain various high-dimensional data related to the patient's tumor, including but not limited to tumor markers, tumor size, stage, pathology results, imaging data, and genomic data;

[0042] After data collection is completed, the data needs to be cleaned and preprocessed, and then the features related to nursing level assessment in high-dimensional tumor data are analyzed, and those features that play an important role in nursing level prediction are selected; for example, tumor size, stage, gene mutation, etc. may directly affect the nursing level.

[0043] After completing feature screening, feature engineering and data transformation are performed. For continuous features, the distribution of the data can be adjusted through logarithmic transformation or differential processing. Finally, all processed, screened and transformed features are integrated into a final data set for subsequent nursing level evaluation. The analytic hierarchy process (AHP) is used to decompose complex nursing level evaluation problems into multiple levels, each of which contains relevant evaluation factors and sub-factors. The core advantage of AHP is that it can quantify the relative importance of different factors by constructing a judgment matrix, helping the system to clarify the degree of influence of each factor on the final nursing level evaluation. By hierarchicalizing the problem, the analytic hierarchy process not only simplifies the complex evaluation process, but also effectively clarifies the relationship between the patient's health data and nursing needs, so that each factor can play its due role in the overall evaluation.

[0044] In this embodiment, the patient care level is intelligently evaluated by combining the screened features and a preset comprehensive evaluation algorithm. The evaluation algorithm hierarchically processes the factors affecting the care level through the hierarchical analysis method. The steps are as follows:

[0045] According to historical data, an n-order judgment matrix A is constructed, and its expression is: Among them, a ij Represents the relative importance evaluation of factor i and factor j. The matrix is ​​a symmetric matrix, and all elements on the diagonal are 1.

[0046] The judgment matrix is ​​constructed by comparing the relative importance of different factors. Each element in the matrix represents the relative importance between two factors.

[0047] By solving the eigenvalue and eigenvector of the constructed judgment matrix A through the formula A·ω=λ·ω, the eigenvector of the matrix can be obtained, where λ is the eigenvalue and ω is the eigenvector, which represents the weight distribution of each factor. i represents the weight of the i-th factor in the decision-making process;

[0048] In order to make the sum of the weight vectors equal to 1, we need to use the formula The obtained eigenvector ω is normalized, and the normalized eigenvector ω=(ω1, ω2, ..., ω n ) is the final weight vector, where each ω i represents the relative weight of the i-th factor in the overall decision;

[0049] In order to ensure the consistency of the judgment matrix, the formula A consistency check is required. If CR < 0.1, the consistency of the judgment matrix is ​​considered acceptable, indicating that the weight allocation has high reliability and rationality.

[0050] In this embodiment, the weights of each evaluation factor are objectively optimized based on the entropy weight method, and the steps of generating the comprehensive nursing level evaluation result are as follows:

[0051] The entropy value E of each indicator based on the information entropy method j Calculate and calculate the information entropy by combining the proportion of multiple data sources through a multi-level information fusion algorithm to optimize the indicator weight. The expression is: Among them, w k is the weighting coefficient of different data sources, p ijk It represents the proportion of the jth indicator in the i-th solution according to the k-th data source; according to the entropy value E j , through the formula Calculate the weight of each indicator Among them, the entropy value E j It is dynamically updated based on real-time monitoring data to improve the real-time adaptability of the assessment model.

[0052] Combining the weights obtained by the hierarchical analysis method and the entropy weight method, the comprehensive weight of each indicator ω is calculated using adaptive weighted regression and weighted average method. j , and the consistency of the judgment matrix is ​​optimized by genetic algorithm. The optimized weighted coefficient is: Where β is the adaptive adjustment coefficient, Adapt(X) is the adaptive adjustment function, and the weighting coefficient α is optimized based on data changes, indicator correlation and expert feedback during the evaluation process. It should be noted that the weighting coefficient α is dynamically adjusted through adaptive weighted regression and optimized according to real-time data feedback during the evaluation process, making the evaluation process more flexible and adaptable.

[0053] The genetic algorithm optimizes the weight distribution in the judgment matrix through crossover and mutation operations to minimize the consistency ratio CR and improve the evaluation accuracy. Then, the evaluation score Sc of the nursing level is obtained through weighted summation, and its calculation formula is: Based on the score and comprehensive weight of each indicator, the evaluation results of the nursing level are further optimized by considering multiple evaluation objectives. As another key algorithm, the entropy weight method objectively optimizes the weight distribution in the nursing level evaluation by quantifying the information entropy of each evaluation factor. The advantage of the entropy weight method is that it can automatically determine which factors contribute more to the nursing level evaluation by calculating the "information concentration" of each evaluation factor. Specifically, the factors with smaller information entropy have higher weights because these factors can provide more effective information about the patient's health status. On the contrary, factors with larger information entropy are given lower weights, thereby reducing their impact on the final evaluation results. Through this objective data-driven approach, the entropy weight method avoids the bias that may be caused by manually setting weights and ensures the fairness and accuracy of the evaluation process.

[0054] The combined use of AHP and entropy weight method can integrate data from different sources (such as patients' tumor markers, mobility, quality of life, etc.) into an accurate and comprehensive nursing level assessment result through scientific weight distribution and multi-level processing. This process is fully automated and does not require human intervention, which greatly improves the efficiency of the assessment. The automated assessment process not only saves nursing staff time and reduces the occurrence of human errors, but also enables rapid assessment of a large number of patients in a short period of time, thereby improving the overall efficiency of patient care management.

[0055] The present invention can more accurately evaluate the patient's nursing level by integrating the comprehensive evaluation model of AHP and entropy weight method, and improve the evaluation efficiency and accuracy through automated and intelligent processes, reduce the risk of human intervention, and ultimately provide patients with more personalized and timely nursing services.

[0056] In this embodiment, when the evaluation result is high risk, first, a high risk judgment threshold is set based on the comprehensive results of historical evaluation data and expert feedback. When the comprehensive score is lower than a certain set threshold, the evaluation result is judged to be high risk. When the evaluation result is high risk, the system will automatically start the high risk reminder function.

[0057] In this embodiment, the alarm prompts include but are not limited to sound alarms, visual alarms, and vibration reminders, ensuring that nursing staff can promptly notice the patient's high-risk assessment results.

[0058] In this embodiment, the historical nursing level assessment data includes historical assessment results, the patient's nursing level score and the corresponding assessment indicators, real-time data feedback, after each assessment, the nursing staff's feedback on the assessment results based on the patient's actual nursing needs, expert evaluation data, and the expert's revised opinions on the nursing level assessment results based on clinical experience and assessment data.

[0059] In the prior art, common nursing level assessment methods mainly include the following schemes:

[0060] Existing solution 1: Traditional nursing level assessment method based on expert evaluation

[0061] This method mainly relies on the clinical experience of experts to evaluate the health status of patients. Usually, experts score patients on multiple assessment dimensions (such as pain, mobility, vital signs, etc.) based on their symptoms, signs, and underlying diseases. Ultimately, experts determine the patient's care level (such as low risk, medium risk, high risk). The main disadvantage of this method is that the evaluation results are affected by subjective factors and lack systematicity and accuracy.

[0062] Existing solution 2: Nursing level assessment method based on simple weighting

[0063] This method relies on experts to set fixed weights for each evaluation factor and combine the scores of each factor with the weights to calculate the final level of care. However, the weight allocation of this method often lacks dynamic optimization and is easily affected by data bias, making it difficult to fully consider the patient's personalized health status.

[0064] Existing solution 3: Evaluation method based on traditional analytic hierarchy process (AHP) and simple entropy weight method

[0065] In this method, the nursing level assessment is carried out by combining AHP and entropy weight method. AHP is used to hierarchically process the factors affecting the nursing level, while entropy weight method is used to calculate the weight of each assessment factor. Although this method is more objective than traditional expert evaluation and weighting methods, it lacks dynamic adaptive capabilities and does not make full use of real-time data to adjust weights, resulting in limited accuracy and flexibility of the assessment.

[0066] In order to fairly compare the effects of this solution with the existing technical solution, the following are the same conditions in the experiment:

[0067] Data source: All experiments are based on the same batch of patient data, including the patients' high-dimensional tumor data. Evaluation indicators: All schemes use nursing grade score as the evaluation indicator and are evaluated using the same scoring criteria. Experimental design: The experiment includes 3 groups: 1) Traditional scheme based on expert evaluation; 2) Traditional scheme based on weighted summation; 3) Traditional scheme based on AHP and entropy weight method; 4) This scheme: A scheme that combines AHP and entropy weight method for dynamic adaptive weight adjustment. Evaluation dimensions: All methods consider the following evaluation dimensions: tumor-related data (such as tumor size, stage, genomic data);

[0068] Experimental condition control: All experiments were conducted on the same equipment and platform, and the same algorithm was used (through unified code implementation and standardized parameter settings).

[0069] The data processing and result calculation methods are consistent to ensure the fairness of the experiment.

[0070] 3. After comparison, show the beneficial effects of this solution through data or charts

[0071] 3.1 Comparative Experimental Results

[0072] In the experiment, the above four schemes were used to evaluate the nursing level of the same group of patients, and the results were compared by calculating the evaluation accuracy and evaluation error.

[0073] Comparison of evaluation accuracy

[0074] Assuming there are 100 samples, the evaluation accuracy refers to the degree of match between the evaluation results and the actual nursing needs (i.e. the actual nursing level). The following are the experimental results:

[0075]

[0076] Evaluation error comparison

[0077] Evaluation error refers to the difference between the nursing level evaluation result and the actual demand (real level). Experimental data show that this scheme can significantly reduce the evaluation error through dynamic optimization and real-time data adjustment. The traditional expert evaluation method is greatly affected by subjective factors, resulting in low evaluation accuracy and large errors. Although the weighted sum evaluation method introduces objective weights, it lacks dynamic adjustment capabilities, so the error is high. The evaluation method combining AHP and entropy weight method is more objective, but it still lacks the adaptability and adjustment mechanism of real-time data, resulting in relatively large errors.

[0078] By introducing adaptive learning algorithms and dynamic optimization mechanisms, this scheme can automatically adjust weights according to real-time data and historical data, reduce evaluation errors, and ultimately achieve the highest evaluation accuracy. Compared with existing traditional methods and combined methods based on AHP+entropy weight method, this scheme (combining AHP, entropy weight method and dynamic optimization) can significantly improve the accuracy of nursing level evaluation and reduce evaluation errors. Traditional expert evaluation methods and weighted summation methods have low evaluation accuracy due to the lack of dynamic data adjustment mechanism. The evaluation method combining AHP and entropy weight method has improved compared with traditional methods, but still has not reached the accuracy and adaptability of this scheme. By introducing a dynamic optimization mechanism and combining real-time data and historical data, this scheme significantly improves the flexibility and accuracy of evaluation, especially in high-risk evaluation, and can provide accurate results more timely.

[0079] The present invention is not limited to the above-mentioned embodiments. Anyone should be aware that any structural changes made under the enlightenment of the present invention, and any technical solutions that are the same or similar to the present invention, fall within the protection scope of the present invention. The technology, shape, and structural parts not described in detail in the present invention are all well-known technologies.

Claims

1. An intelligent patient care level assessment system based on high-dimensional tumor data, characterized in that: include: A data acquisition module, used to collect high-dimensional tumor data and activity data of patients; A feature extraction module, used to filter out features related to nursing level assessment from the high-dimensional tumor data and activity data, and perform preliminary processing on the data; A nursing level assessment module is used to intelligently assess the patient's nursing level by combining the screened features with a preset comprehensive assessment algorithm. The assessment algorithm hierarchically processes the factors affecting the nursing level through the analytic hierarchy process, and objectively optimizes the weight of each assessment factor based on the entropy weight method to generate a comprehensive nursing level assessment result; An assessment result display module is used to display the patient's nursing level assessment results in text form and automatically remind nursing staff when the assessment result is high-risk; The data interface module is used to interact with other medical systems or devices to ensure the timeliness and accuracy of data acquisition and transmission.

2. According to claim 1, a patient care level intelligent assessment system based on high-dimensional tumor data is characterized in that: The steps of selecting features related to nursing level assessment from the high-dimensional tumor data and activity data and performing preliminary data processing are as follows: Obtain various high-dimensional data related to the patient's tumor, including but not limited to tumor markers, tumor size, stage, pathology results, imaging data, and genomic data; After data collection is completed, the data needs to be cleaned and preprocessed, and then the features related to nursing level assessment in high-dimensional tumor data are analyzed to select those features that play an important role in nursing level prediction; After feature screening, feature engineering and data transformation are performed. For continuous features, the distribution of data can be adjusted through logarithmic transformation or differential processing. Finally, all processed, screened and transformed features are integrated into a final data set for subsequent care level evaluation.

3. The patient care level intelligent assessment system based on high-dimensional tumor data according to claim 1 is characterized in that: The patient care level is intelligently evaluated by combining the screened features and a preset comprehensive evaluation algorithm. The evaluation algorithm uses a hierarchical analysis method to hierarchically process the factors that affect the care level. The steps are as follows: According to historical data, an n-order judgment matrix A is constructed, and its expression is: Among them, a ij It represents the relative importance evaluation of factor i and factor j. The matrix is ​​a symmetric matrix, and all elements on the diagonal are 1; By solving the eigenvalue and eigenvector of the constructed judgment matrix A through the formula A·ω=λ·ω, the eigenvector of the matrix can be obtained, where λ is the eigenvalue and ω is the eigenvector, which represents the weight distribution of each factor. i represents the weight of the i-th factor in the decision-making process; In order to make the sum of the weight vectors equal to 1, we need to use the formula The obtained eigenvector ω is normalized, and the normalized ω=(ω1,ω2,…,ωn) is the final weight vector, where each ω i represents the relative weight of the i-th factor in the overall decision; In order to ensure the consistency of the judgment matrix, the formula A consistency check is required. If CR < 0.1, the consistency of the judgment matrix is ​​considered acceptable.

4. The patient care level intelligent assessment system based on high-dimensional tumor data according to claim 1 is characterized in that: Based on the entropy weight method, the weights of each evaluation factor are objectively optimized, and the steps to generate the comprehensive nursing level evaluation results are as follows: The entropy value E of each indicator based on the information entropy method j Calculate and calculate the information entropy by combining the proportion of multiple data sources through a multi-level information fusion algorithm to optimize the indicator weight. The expression is: Among them, w k is the weighting coefficient of different data sources, p ijk It represents the proportion of the jth indicator in the i-th solution according to the k-th data source; according to the entropy value E j , through the formula Calculate the weight of each indicator Among them, the entropy value E j It is dynamically updated based on real-time monitoring data to improve the real-time adaptability of the assessment model. Combining the weights obtained by the hierarchical analysis method and the entropy weight method, the comprehensive weight of each indicator ω is calculated using adaptive weighted regression and weighted average method. j , and the consistency of the judgment matrix is ​​optimized by genetic algorithm. The optimized weighted coefficient is: Where β is the adaptive adjustment coefficient, Adapt(X) is the adaptive adjustment function, which optimizes the weighting coefficient α based on data changes, indicator correlation and expert feedback during the evaluation process; The genetic algorithm optimizes the weight distribution in the judgment matrix through crossover and mutation operations to minimize the consistency ratio CR and improve the evaluation accuracy. Then, the evaluation score Sc of the nursing level is obtained through weighted summation, and its calculation formula is:

5. The patient care level intelligent assessment system based on high-dimensional tumor data according to claim 1 is characterized in that: When the assessment result is high risk, first, a high-risk judgment threshold is set based on the comprehensive results of historical assessment data and expert feedback. When the comprehensive score is lower than a certain set threshold, the assessment result is judged to be high risk. When the assessment result is high risk, the system will automatically start the high-risk reminder function.

6. The patient care level intelligent assessment system based on high-dimensional tumor data according to claim 1 is characterized in that: Alarm prompts include but are not limited to sound alarms, visual alarms and vibration reminders to ensure that caregivers can promptly pay attention to the patient's high-risk assessment results.

7. The patient care level intelligent assessment system based on high-dimensional tumor data according to claim 1 is characterized in that: Historical nursing level assessment data includes historical assessment results, patients’ nursing level scores and corresponding assessment indicators, real-time data feedback, after each assessment, nursing staff’s feedback on the assessment results based on the patients’ actual nursing needs, expert assessment data, and experts’ revised opinions on the nursing level assessment results based on clinical experience and assessment data.

Images